Professional Documents
Culture Documents
Michael 2015 - Freud's Theory of Dreams
Michael 2015 - Freud's Theory of Dreams
Dialog-on-Freud Series
Series Editor: M. Andrew Holowchak
The Dialog-on-Freud series invites authors to explore the history and practice of analytic
therapies through critical analysis of and expatiation on the seminal work of Freud. It
seeks books that critically scrutinize the numerous facets of Freud’s work over the course
of his life, that investigate how or to what extent Freud’s thinking causally gave rise to the
various sorts of therapies that currently exist, and that examine the relevance of Freud’s
thinking today for those therapies.
A Philosophico-Scientific Perspective
Michael T. Michael
All rights reserved. No part of this book may be reproduced in any form or by any
electronic or mechanical means, including information storage and retrieval systems,
without written permission from the publisher, except by a reviewer who may quote
passages in a review.
Michael, Michael T.
Freud's theory of dreams : a philosophico-scientific perspective / Michael T. Michael.
pages cm. -- (Dialog-on-freud series)
Includes bibliographical references and index.
ISBN 978-1-4422-3044-6 (cloth : alk. paper) -- ISBN 978-1-4422-3045-3 (electronic)
1. Freud, Sigmund, 1856-1939. 2. Dream interpretation. 3. Dreams. 4. Psychoanalysis. I. Title.
BF1078.M456 2053 2015
154.6'3092--dc23
2015027319
TM
The paper used in this publication meets the minimum requirements of American
National Standard for Information Sciences Permanence of Paper for Printed Library
Materials, ANSI/NISO Z39.48-1992.
Acknowledgments ix
Introduction: Alchemy or Calculus? xi
vii
Acknowledgments
My first thanks go to Mark Holowchak for asking me to write this book and
for his fine editorship. Thanks also to Alison Pavan, Amy King, Kayla Rid-
dleberger, and the other people at Rowman & Littlefield for their good com-
munication, patience, and hard work throughout the process. Beyond this,
there are numerous colleagues who have helped me greatly by reading drafts
and offering feedback, or simply by answering questions. The biggest thanks
in this regard go to Tim Fuller and Loren Goodman, but also Brad Bow, Paul
Chang, Chad Denton, Carroll Brooks, Neeraja Sankaran, and Axel Gelfert.
Thanks to Lu Yu for assisting me with the references and Aeree Kim with the
cover and indexing. My gratitude also to all my students at Yonsei University
for helping me become a better communicator, and to the university for the
opportunities it has given me. Finally, my special heartfelt thanks to Sang
Hyun for her unstinting support and understanding, without which I would
never have been able to do this.
ix
Introduction
Alchemy or Calculus?
core claims turned out to be false: the philosopher’s stone does not exist, base
metal cannot be turned into gold in a chemical reaction, and there is no elixir
of life. But these claims were not due to superstition and irrationality; they
were, with respect to the science of the day—in particular, Aristotle’s theory
of elements—eminently reasonable. Moreover, the use of controlled experi-
mentation by alchemical practitioners, in contrast to the Aristotelian prefer-
ence for naturalistic observation, was vital to the development of science.
Could psychoanalysis be like this? Like alchemy, it is first and foremost a
practical art. Like alchemy, it advocates techniques not accepted by main-
stream science. Like alchemy, it promises forbidden secrets. And like alche-
my in the latter years of the scientific revolution, it is hugely controversial.
Adepts defend it with a passion from the astonishingly fierce attack of oppo-
nents, while those who stand in between continue to point out the flaws that
bar it from the respectability of scientific acceptance.
In the practice of dream interpretation we see a pursuit analogous to that
of turning base metal into gold. Here the base metal is the chaotic material of
our dream-lives, and the gold is the deep understanding of ourselves that
interpretation promises. This centuries-long pursuit has its apotheosis in
Freud’s theory of dreams, which offers to be the philosopher’s stone of the
mind. But its promise is false. The understanding it provides is, like the
alchemist’s gold, an alloy made by mixing genuine insight with the ordinary
mental outpourings of the dreamer. Even great men, a Karl Popper or an
Albert Einstein (see Chapters 5 and 6), may be fooled by it. But the cold,
sober judgment of posterity will be just as it was with alchemy: an ingenious
but ultimately mistaken pursuit.
Or perhaps that will not be its fate. Perchance we have chosen the wrong
metaphor. Maybe Freud’s theory of dreams is like a different kind of philoso-
pher’s stone: the hard, smooth stone of calculus. The term “calculus” comes
from the Latin for the small pebbles used in an abacus. This is an apt choice
to describe a method that matches that of the abacus in its precision and
practical value. Just as the abacus has for centuries been invaluable to those
endeavoring to apply mathematics to everyday life, so today the calculus is
invaluable to anyone endeavoring to apply even moderately advanced mathe-
matics to real-world problems. And just as the abacus, when used properly,
can be relied upon to yield the right result, so too modern calculus is a
paragon of exactitude.
But this has not always been the case. The high regard in which we hold
the calculus, though widespread among the mathematicians who used it in
the early part of the eighteenth century, was not shared by all. The eminent
philosopher Bishop George Berkeley, one of the three great British empiri-
cists, famously attacked it in his essay The Analyst, arguing that it was
replete with obscurities, mysteries, and unsound inferences. He wrote acerbi-
cally:
Introduction xiii
And what are these Fluxions [that is, derivatives]? The Velocities of evanes-
cent Increments? And what are these same evanescent Increments? They are
neither finite Quantities nor Quantities infinitely small, nor yet nothing. May
we not call them the ghosts of departed quantities? (Analyst, §35)
both sides have their points, and the verdict is “unproven.” But “unproven”
comes in many shades, and appreciating which shade applies is vital to
determining the appropriate attitude one should take toward the theory. It is
the quest for the right shade that is often left out of debates. This is the quest
we will engage in.
The quest will take the form of three simple questions. Is what Freud had
to say about dreams reasonable? Is it scientific? Is it still relevant?
The first question concerns the plausibility of the theory and method.
Many critics dismiss the theory even before examining the evidence. One
reason for this is the oft-repeated accusation that the theory is based on
outdated ideas—that it derives from Freud’s abandoned Project for a Scien-
tific Psychology, a work of now-obsolete speculative neuroscience. The in-
fluence of this work and these ideas is undeniable, but it is, I think, unjust to
hold Freud’s eagerness to ground his ideas in neuroscience against him. He
abandoned the Project because he realized that the neuroscientific concepts
that it encompassed could not be relied upon. But that is not to say that,
transposed to a psychological context, from whence many of them anyway
emerged, they cannot be illuminating, and cannot form the basis of theories
that subsequently gain credence from other sources of enquiry. A proper
assessment requires close examination of Freud’s rationale.
Other critics dismiss Freud’s method of interpretation. They maintain that
the superficial similarities that make an interpretation look plausible can
easily be generated, but cannot tell us anything about causes. On this basis,
the philosopher Clark Glymour concludes, “The method itself is worth-
less . . . the objections to it are obvious ones. . . . It is hard to believe that they
did not occur to Freud himself. The whole business seems the cheapest of
rhetorical tricks” (Glymour 1983, 65). But is that all there is to Freud’s
method? If it were, then those who claim to have interpreted dreams along
the lines drawn by Freud are, wittingly or unwittingly, participants in this
charade. Again a closer examination is in order.
The second question asks if the theory and method are scientific. Freud
was adamant that they were. He was frustrated at those who demanded
proofs beyond those he gave, as if the evidence of his interpretations was not
good enough. It was, in his mind, plenty good enough: his method of inter-
pretation was analogous to the microscope, which would reveal a world
unseen to those skilled enough to master its use. But critics remained, and
still remain, unmoved. The method is unscientific, so its “proofs” are irrele-
vant. And even the theory itself is, in the eyes of many, of a kind that cannot
be admitted into science because it lacks an essential characteristic.
Since much of the critical appraisal of psychoanalysis has revolved
around this question of scientificity, we would do well to enquire of what
historians and philosophers of science have to say on the matter. I speak of
historians and philosophers of science, and not scientists themselves, because
Introduction xv
often scientists are too narrowly focused to put their own subject into appro-
priate perspective. An example of this is the “Dream Debate” between the
anti-Freudian psychiatrist Hobson and the pro-Freudian neuroscientist Solms
on the value of Freud’s theory in the light of recent scientific research on
dreams (see Chapter 7). At one point in the debate, Hobson resorted to the
Popperian objection that Freud’s theory is “unfalsifiable,” to which Solms
responded by pointing out that Hobson had spent most of the time up until
then trying to show that Freud’s theory had in fact been empirically “fal-
sified.” Hobson’s appeal to Popper’s famous criterion for distinguishing sci-
ence from non-science is typical, but sadly ignorant of recent philosophy of
science, which has shown Popper’s ideas to be problematic. A better under-
standing of this philosophy may have obviated Hobson’s making such an
objection. For us it will provide insight into why many scientists do try to
have it both ways—“falsifiability,” like the concept of “scientificity” that it
attempts to capture, is far less clear than it initially appears. The history of
science, too, casts a different light on the issue of scientificity. Some of our
most successful scientific theories were once regarded, effectively, as
pseudoscientific. This illustrates the value of the perspective of the history
and philosophy of science. One may not be able to say of Freud’s work that it
is scientific, but that is not necessarily to relegate it to the same category as
creation science and astrology.
The third question is about how Freud’s theory measures up to modern
scientific research on dreams. Since the discovery of rapid-eye-movement
sleep in the 1950s, sleep science has flourished. With it have come numerous
supposedly evidence-based theories of dreaming. Perhaps the most promi-
nent of these is that of the aforementioned psychiatrist, Hobson. He and
many other scientists have turned their backs on Freud’s theory in favor of
ones grounded in modern neuroscientific techniques. Are they right to do so?
We will see. Other critics have amassed evidence from multiple lines of
research against Freud’s theory, regarding it as overwhelmingly contradicted
by the scientific findings. Are they right? If they are, then the theory, even if
plausible, even if scientific, is no longer relevant. It is, maybe, a historical
curiosity, a testament to imaginative genius, but not something to take seri-
ously, not something to trust as providing insight into our inner nature. But,
again, we will see.
What we are engaged in, then, is an epistemological enterprise. My aim is
to evaluate Freud’s theory on his own terms—as a theory that is plausible,
scientific, and lasting. This is, therefore, not a book on the history of Freud’s
theory, or simply reportage of the controversy. It is a book that seeks an-
swers. It is not a polemical book. I have no axe to grind or vested interest to
serve, no need to take sides. It is an attempt at an objective appraisal, an
effort to engage sincerely with the criticisms. Where Freud’s theory is defen-
sible, I will defend it; where it is not, I will assess the degree of damage
xvi Introduction
rendered by the objection. By doing this, and by placing the theory within the
context of the history and philosophy of science, I hope to offer nuance and
perspective to the debate.
I will not, and cannot, address all objections brought against the theory.
So, for example, while I discuss the prominent critic Adolf Grünbaum’s
arguments about the scientific acceptability of Freud’s method, I do not
discuss all his objections to the theory of dreams. I have a view on these, but
leave it to another occasion. The three questions that I have laid out order and
determine what I include and what I do not. To those whose arguments I
bypass: I will be happy to engage with you, but in another format, at another
time.
I will also have little to say about the so-called hermeneutic conception of
psychoanalytic theory. The philosophers Jürgen Habermas and Paul Ricoeur
have argued that Freud is guilty of a “scientistic self-misunderstanding,” and
have attempted to reinterpret his theory upon a different construal of its
causal claims. But this book is about Freud’s theory of dreams, not some
reinterpretation of it. It is an attempt to be faithful to how Freud understood
his own theory. There are in any case many problems with the hermeneutic
conception, which others, in particular Grünbaum (1984, 1–94), have written
about. I will simply pass over the subject in silence. My concerns are avow-
edly scientistic, in the non-pejorative use of the term.
In keeping with the series theme of “Dialogs on Freud,” the book, though
not written in dialog form, is a dialog in essence. It is a dialog with Freud,
with critics of Freud, and with defenders of Freud. As with most dialogs,
there will be no pithy, easy-to-summarize conclusion. It is the process of
examination, not the outcome, that is of greatest value. The logician Doug
Walton puts it most aptly when he describes reasoned dialog as “the vehicle
that enables one to come to better understand one’s own position on impor-
tant issues, one’s own reasoned basis behind one’s deeply held convictions”
(1988, 253). I believe the reader to whom the question of the validity of
Freud’s theory of dreams is important will gain much from the present exam-
ination: a clearer view as to which objections to take seriously and how
seriously, and a better perspective on what attitude is appropriate to Freud’s
claims.
Dreams in Therapy
relevant. What one believes and the assurance with which one believes it
undoubtedly influence one’s interpretation of one’s experience.
There has unquestionably already been a strong influence on therapy by
critics such as Adolf Grünbaum and J. Allan Hobson. The psychoanalyst and
scholar Rachel Blass makes this point. She argues that criticisms of the
theory have “had a latent effect on clinical practice” (2001, 115). This is
manifested in two ways. One is on the take-up of the hermeneutic position on
psychoanalytic interpretation that I mentioned earlier. This position rejects
Freud’s view that an interpretation picks out causal states (as normally
understood in natural science), holding instead that it is more like a meta-
phor, an imaginative creation, thereby transcending the need for justification.
The other effect is the “present-day focus on the dream as a form of discourse
and communication” (ibid., 146). This is an approach whereby what matters
is what the analysand makes of the dream, rather than what caused the
dream. It may occur in conjunction with the hermeneutic approach, but may
also occur separate from it. Both of these approaches depart from Freud’s
emphasis on discovery, of uncovering a real and objective hidden meaning,
in favor of an ostensibly more pragmatic approach. On Blass’s analysis, the
ascendance of these positions is a consequence of theoretical attacks on
Freud’s theory.
Blass’s point is well taken. While it is difficult to quantify the degree of
influence that critics have had, it is almost certain that had they not been
taken as seriously as they have the situation would have been different. It is
therefore all the more important to determine whether these criticisms are
sound. Thus though our quest will barely brush the surface of the many
issues about dreams that therapists have been grappling with over the years
since Freud introduced his theory—such as the centrality of dreams to thera-
py, the role of the ego in dreams, and the importance of the manifest content
to interpretation—it nevertheless addresses the most fundamental issue: the
extent to which Freud’s theory of dreams is justified. Without clarity about
this, there can be little clarity about anything else.
Though this is not a book on the history of Freud’s theory, it would be amiss
not to say something of this. Freud, as is well-known, was the originator not
just of a single theory, but of an entire movement, one that has had an
astonishing influence on Western thought for over a century. His contribu-
tions—addressing subjects as diverse as the causes of psychopathology, sex-
ual development in infancy, slips of the tongue, jokes, and religious belief—
are numerous. Yet Freud thought his theory of dreams the greatest of all his
discoveries. He wrote of it, “Insight such as this falls to one’s lot but once in
xviii Introduction
a lifetime” (1900, S.E., IV: xxxii). This “insight” was, without doubt, among
his proudest moments.
How exactly such insight occurs is a useful historical question to ask. In a
flash, some would say. Freud himself emphasized a particular event: his
interpretation of the dream of Irma’s injection, the “specimen dream” that
was later to form the centerpiece of his Interpretation of Dreams. In a letter
to his friend and confidant Wilhelm Fliess on June 12, 1900, he wrote of the
house in which he had this dream, “Do you suppose that someday one will
read on a marble tablet on this house: Here, on July 24, 1895, the secret of
the dream revealed itself to Dr. Sigm. Freud” (Masson 1985, 417). The
background to this revelation is as follows. Freud had been in the habit of
writing down his dreams since his youth, but at the time of the dream this
practice had taken on a new urgency. He had developed a novel method of
analyzing hysterical symptoms through the use of free associations, and in
the course of applying this method, his patients would quite naturally talk
about their dreams. Thus he found himself interpreting dreams as well as
symptoms. He had touched upon dream interpretation in Studies on Hysteria,
the book co-written with his mentor, Joseph Breuer, by which he had intro-
duced his controversial new ideas to the world. In this he indicated that
dreams operated according to associative rules, but he suspected there was
more. On that fateful night, his suspicions were confirmed. He discovered
that dreams are wish-fulfillments.
This heroic story, however, needs correcting in at least two ways. First, it
presents Freud as the sole, intrepid discoverer, but this is misleading. Almost
all components of his theory can be found in the works of predecessors. Even
before the Irma dream, Freud had drawn a link between dreams and psycho-
sis and hence—as this is what he believed psychotic hallucination to be—
with wish-fulfillment. Some thirty years earlier, the psychiatrist Wilhelm
Griesinger had observed the very same connection, a fact that Freud may
well have learned from his teacher, Theodor Meynert (after whom a form of
psychosis, Meynert’s amentia, is named). Freud’s awareness of such influ-
ences was gradual. In 1909 he wrote, “I am really very ignorant about my
predecessors. If we ever meet up above they will certainly greet me ill as a
plagiarist” (in Jones 1955, 443). Fourteen years later, it seems his knowledge
had grown. In a short paper, he acknowledged that the engineer Josef Pop-
per-Lynkeus had expressed the “essential part of my theory of dreams” that
the distorted, seemingly meaningless, nature of most dreams was due to
censorship of forbidden ideas. In this particular case, Freud considered it an
independent discovery, but he also admitted that other elements of the theory
were not his own. In relation to the “hidden and long-forgotten sources”
behind seemingly original theories, he wrote, “in my case, too, the originality
of many of the new ideas employed by me in the interpretation of dreams and
in psychoanalysis has evaporated in this way” (1923, S.E., XIX: 261). The
Introduction xix
evidence for the full version of his theory, which included the claim that the
meaning of the dream was an infantile wish. It was, of course, deeply unsatis-
factory to leave such a large evidential gap in the most controversial area of
his theory.
The more sober reviews of the book reflect this. A picture has emerged,
promoted by Freud himself as well as his early followers, that the book met
with a frosty reception from the scientific establishment. In reality, though,
the reception was mixed. There were favorable reviews, including ones that
described the work as “epoch-making” and “profound,” as well dismissive
ones (Sulloway 1992, 347). But even the favorable reviews pointed out the
lack of direct evidence for some of Freud’s claims. This was hardly some-
thing that Freud could dispute—he himself admitted, “There is too much that
is new and incredible, and too little strict proof” (ibid.). In particular, there
was too little strict proof of the most controversial assertion, that dreams
were driven by infantile wishes. Carl Jung reflected the mood when he later
urged Freud to include fully interpreted dreams in order that the “ultimate
real motives” be “ruthlessly disclosed” (ibid.).
The sales of the book disappointed Freud—only 351 in the first two years,
rising to 600 in eight. This, though, may have had more to do with Freud’s
decision to publish a briefer summary work, On Dreams, just a year after the
Interpretation was published. 1 In any case, in due time the reputation of the
book grew, in line with that of psychoanalysis in general. The Interpretation
of Dreams was one of only two books (the other being Three Essays of the
Theory of Sexuality, 1905) that Freud kept up to date. He did so through
seven additional editions, beginning with the second in 1909, and ending
with the eighth in 1930. The first English translation was A. A. Brill’s, in
1913, following a Russian one in 1904, and followed by Spanish, French,
Swedish, Japanese, Hungarian, and Czech translations within Freud’s life-
time. It has since gone through countless other editions and translations,
becoming a global classic of the twentieth century.
The Interpretation of Dreams was, of course, not the only text by which
Freud expounded his theory or offered evidence in its support. From the
famous Dora case study, to other major works, such as Jokes and Their
Relation to the Unconscious, to his many lectures and overviews of psycho-
analysis, Freud continued to write about dream interpretation. Through these
writings, as well as through the numerous editions of the Interpretation, he
also continued to develop the theory. While he remained true to the core of it,
he changed it in small but significant ways—most prominently by placing
greater emphasis on symbolism in dreams, revising his original wish-fulfill-
ment claim in recognition of the traumatic elements within many dreams, and
by gradually assigning an increasing role to the ego in the formation of
dreams (Weiss 1992). These trends have been continued by his followers
after his death (Brenner 1969).
Introduction xxi
Our focus, however, will be on that essential core that remains definitive
of Freud’s theory. This will be described in detail in the coming chapters.
However, from the beginning Freud was offering more than just a theory; he
was initiating his readers into a practice, a craft. It is appropriate therefore to
precede our analysis of the theory with an exhibit of Freud’s technique of
dream interpretation. There is no more natural an exhibit to provide than
Freud’s very own specimen case.
At the time of the dream, Freud was at a precarious stage in life. He had long
given up his youthful ambitions to be a physiology researcher and had in-
stead opened a medical practice, where he specialized in neurotic illnesses.
The income this provided was irregular, depending on the ever-fluctuating
number of patients who came to him. His family was growing—at the time of
the dream his wife was pregnant with their sixth child—and he was anxious
about supporting them. He aspired to become a professor in order to ease his
economic worries, 2 but he faced anti-Semitic discrimination and hostility
against his ideas about the relation between sexuality and neuroses. Even
Breuer was distancing himself from him. This period was also one of profes-
sional self-doubt. Just a few months before, a patient of his, Emma Eckstein,
had suffered a near-fatal hemorrhage due to the medical negligence of
Freud’s closest friend, Wilhelm Fliess, to whom he had referred Emma. 3 At
the time of the dream of Irma’s injection, Freud was without fame, without a
following, without financial security, without a great deal of self-assurance,
and without the intellectual backing of his old friend Breuer. This was the
general context of the dream that he had on the night of July 23–24, 1895.
The immediate context was that he had been treating a young lady, Irma, 4
about whose case he felt pressured due to his close acquaintance with her
family. He had partially cured her of her hysterical symptoms, but was anx-
ious because the cure was incomplete. The day before his dream, a friend and
fellow physician, Otto, 5 who had seen Irma while she was on vacation, had
responded to Freud’s enquiries about her by remarking that, while she was
better, she was not yet well. Freud thought this a reproof, which angered him.
It prompted him to spend much of the night writing a long letter of self-
justification to Breuer (who appears as Dr. M. in the dream). Shortly after, he
had the dream.
myself that after all I must be missing some organic trouble. I took her to the
window and looked down her throat, and she showed signs of recalcitrance,
like women with artificial dentures. I thought to myself that there was really no
need for her to do that.—She then opened her mouth properly and on the right
I found a big white patch; at another place I saw extensive whitish grey scabs
upon some remarkable curly structures which were evidently modeled on the
turbinal bones of the nose.—I at once called in Dr. M., and he repeated the
examination and confirmed it. . . . Dr. M. looked quite different from usual; he
was very pale, he walked with a limp and his chin was clean-shaven. . . . My
friend Otto was now standing beside her as well, and my friend Leopold was
percussing her through her bodice and saying: ‘She has a dull area low down
to the left.’ He also indicated that a portion of the skin on the left shoulder was
infiltrated. (I noticed this, just as he did, in spite of her dress.) . . . M. said:
‘There’s no doubt it’s an infection, but no matter; dysentery will supervene
and the toxin will be eliminated.” . . . We were directly aware, too, of the
origin of the infection. Not long before, when she was feeling unwell, my
friend Otto had given her an injection of a preparation of propyl, propyls . . .
propionic acid . . . trimethylamin (and I saw before me the formula for this
printed in heavy type). . . . Injections of that sort ought not to be made so
thoughtlessly. . . . And probably the syringe had not been clean. (1900, S.E.,
IV: 107; his ellipses)
Freud’s analysis of this dream is the most exhaustive of all those he gives
in The Interpretation of Dreams, consisting of thirteen pages in which he
reports his thoughts as they occurred to him in relation to each element of the
dream. Here I can give but a flavor of this. 6
“I was alarmed . . .”
Associating with the element, “I was alarmed and looked at her,” Freud
reports that “a faint doubt crept into my mind . . . that my alarm was not
entirely genuine” (ibid., 109). Following this line of thought, he comes to
realize that he actually harbors a wish that the symptom has an organic cause,
as this would relieve him of responsibility for not being able to cure Irma
(since his is a psychological cure).
Irma
The presence of Irma by the window brings to Freud’s mind other people.
One is Irma’s friend, a woman he admires. He had chanced upon her one
evening while she was standing by the window just as Irma was in the dream.
Also, he knows that she suffers from hysterical choking, as Irma does in the
dream (but not in real life). Freud remarks that he wanted to have her as his
patient, but that she was too reserved—too recalcitrant—to ask him to treat
her, and in any case, her condition was not serious enough: there was no need
for her to do it. It occurs to Freud that Irma had acquired characteristics of
Introduction xxiii
her friend in the dream because he wished to exchange her for that other lady.
The other lady, being wiser, would not have disputed his solution to her
hysteria as Irma had done—by which thought, Freud realizes that he is again
passing on the responsibility for failing to cure Irma, this time by attributing
the blame to her.
But perhaps it is not just Irma that he wishes to exchange, for other
characteristics—the pale, puffy face and the pains in the abdomen—remind
him of his pregnant wife, whose birthday party is the event the dream is
anticipating.
Dr. M.
The Dr. M. of the dream also brings to mind another person: Freud’s elder
brother, who is similarly clean-shaven, has recently had a limp, and other-
wise closely resembles Dr. M. as he appears in the dream. In real life there is
in Freud’s mind one salient connection between these two older men: they
had both rejected a suggestion he had made.
This rejection, it seems, bears a relation to the absurd prognosis that
Freud attributes to M. in the dream, that “it’s an infection, but no matter.
Dysentery will supervene and the toxin will be eliminated.” The reference to
dysentery reminds Freud of a case in which a patient of his, whose symptoms
he had diagnosed as hysterical, was, on a trip abroad, alternatively diagnosed
by an “ignorant” doctor as having dysentery. This in turn reminds Freud of
an anecdote that M. had told him, of when a colleague of his, failing to
recognize the seriousness of a patient’s symptoms, had foolishly declared,
“No matter, the albumen will soon be eliminated.” Freud concludes from this
that he was mocking M. in his dream for failing to realize that Irma’s symp-
toms were hysterical. He realizes he is wishing a loss of face on both M. and
his brother for failing to accept his theories. 7
Otto
The greatest opprobrium is reserved for Otto. In the dream, Leopold (a rela-
tive of Otto), 8 observes that there is “a dull area low down on the left.” As
Freud reports, this agrees “in every detail” with a case that he remembers in
which Leopold impressed him with his thoroughness. This contrasts with
Otto’s hasty judgment of Freud. Freud is thus comparing Otto unfavorably
with his relative. Later in the dream he reports that Irma was feeling unwell
because “Otto had given her an injection.” In this way he blames Otto direct-
ly for her illness. The words in his dream, “injections of that sort ought not to
be made so thoughtlessly,” strike him as distinctly similar to the thought he
had in response to Otto’s implied rebuke of him, that such judgment ought
not to be made so thoughtlessly. Furthermore, the words, “probably the sy-
ringe had not been clean,” bring to mind tidings from the day before, in
xxiv Introduction
which he had learned that a former patient of his, an old lady, had become ill
due (he guessed) to being injected with an unclean syringe. This negligence
he attributes to Otto in the dream, in contrast to his own scrupulousness—he
prides himself on never having caused an infection in this way. Otto—the
little squirt 9 —is thus blamed and denigrated in several ways in the dream.
The Chemicals
We see from this analysis how multiple lines of association seem to converge
on the overarching theme of Freud’s dodging blame for failing to cure Irma’s
illness. The associations he makes with the elements of the dream bring to
light a series of excuses—that Irma’s symptoms could not be cured because
she had failed to accept his solution (in contrast to what the other, wiser, lady
would have done); because her symptoms had an organic rather than psycho-
logical cause; because they had been caused (twice-over) by Otto’s careless-
ness; and because of her widowhood. Freud wishes to free himself of respon-
sibility for failing to cure Irma, and the dream seems to fulfill this in numer-
ous, if incompatible, ways. He summarizes the meaning thus: the “content of
the dream was the fulfillment of a wish and its motive was a wish” (ibid.,
119). There are also a number of related wishes that are fulfilled in the
dream: a wish to revenge himself on Otto, both for the perceived reproof and
the foul-smelling liqueur; a wish to revenge himself also on his elder brother
and on Breuer for doubting him; and a wish to be right about the sexual
origins of hysteria. These though can be seen as corollaries to the wish to be
exonerated from blame. 11
What makes Freud’s interpretation, and the sub-interpretations on which
it is based, at all compelling? Might not Freud be reading these interpreta-
Introduction xxv
tions into the material, rather than gleaning them from it? About this issue I
will have much to say in Chapters 3 and 4, but observe for now the different
kinds of link that Freud uses. At one time it is his current emotional response
to a dream element by which a connection is made (“a faint doubt crept into
my mind”). At other times it is a distinctive similarity between the dream-
content and a memory (“I had found her by a window in the situation repro-
duced in the dream”; “The dull area low down on the left seemed to agree in
every detail with one particular case in which Leopold had struck me by his
thoroughness”). At others it is a looser similarity, but one with many points
of contact (“my elder brother . . . who is clean-shaven and whom, if I
remembered right, the M. of the dream closely resembled . . . was walking
with a limp”). There are also occasions in which Freud expresses certainty in
his conclusions (“I could no longer feel any doubt, therefore, that this part of
the dream was expressing derision at physicians who are ignorant of hyster-
ia”), where it is not entirely clear if this is simply an expression of confidence
in his inference, or if it is something more—something perhaps akin to self-
knowledge. Finally, there is the synthetic aptness of his overall interpreta-
tion: it fits the dream by unifying the disparate parts of it.
There is, though, clearly more to this dream than the interpretation Freud
reports. The occasion, recall, is that of his pregnant wife’s birthday, and his
wife appears in the dream in the guise of Irma, who he wants to exchange for
another woman of whom he has a higher opinion. About this other woman he
remarks, suggestively, “She would then have opened her mouth properly.”
He adds in a footnote to this, “I had a feeling that the interpretation of this
part of the dream was not carried far enough to make it possible to follow the
whole of its concealed meaning” (ibid., 111). Also, when discussing the
element in spite of her dress, he breaks off, admitting, “Frankly, I had no
desire to penetrate more deeply at this point” (113). Later, once he has
presented his interpretation, he informs, “I will not pretend that I have com-
pletely uncovered the meaning of the dream or that its interpretation is with-
out a gap” (120–21). The sense that there is a sexual subtext is palpable.
The interpretation of this dream aptly illustrates Freud’s method, but also
brings to light the problems that accompany it. We get a glimpse of the
plausibility of his interpretation, as well as the doubt that may be cast on it.
We perceive too the plight that using his own dreams put him in, with the
need to evidence his full theory calling for a deeper-layer interpretation, but
the needs of discretion preventing him from giving this interpretation. Yet for
Freud, despite the problems, he had done enough to open the gates to the
royal road. By mastering his technique of dream interpretation, people could
gain access to the otherwise hidden truths of psychoanalysis. But over a
hundred years later, both the technique and the theory built upon it remain
heavily disputed. Our task begins here.
xxvi Introduction
NOTES
1. Sales figures for On Dreams are unavailable—but see Sulloway (1992, 349) for reasons
for thinking its impact to be significant.
2. Professor Extraordinarius—a title Freud attained in 1902—carried no duties, but was a
position of prestige that would boost his practice.
3. In line with his eccentric theory that the nose and the female genitals were linked, Fliess
had performed nasal surgery on Emma in order to cure her hysterical symptoms, which he
thought were due to masturbation. He had, however, left a piece of medical gauze in her nose,
causing an infection. This was discovered and removed by another doctor, but she later had
another hemorrhage and had to undergo an emergency operation to save her life (Masson 1985,
116–21).
4. Probably Anna Hammerschlag, whose father was Freud’s Hebrew teacher.
5. A pseudonym for Oscar Rie, physician to Freud’s children.
6. The dream has been extensively analyzed and discussed in the literature. See especially
Erikson (1954) and Anzieu (1986).
7. The German word translated “clean-shaven” in the dream report is bartloss, or beard-
less. Erikson (1954) sees in this a “vengeful castrative impulse on the part of the dreamer.”
8. Ludwig Rosenberg, brother-in-law of Oscar Rie.
9. According to Erikson (1954), the German word for syringe that Freud uses, Spritze, can
also be used with this connotation—a “dirty squirter” or “little squirt.” This relates to Erikson’s
suggested deeper interpretation of the dream, which links it with a childhood event in which
Freud had relieved himself in his parents’ bedroom and thereby incurred his father’s wrath.
Before his beloved mother, his father had exclaimed, “That boy will never amount to any-
thing”—a devastating forecast for a clever child of whom much was expected. Freud writes,
“This must have been a terrible affront to my ambition, but allusions to this scene recur again
and again in my dreams” (1900, S.E., IV: 274). It is eminently plausible that Otto’s remark of
the day before, which Freud interpreted as, “You’re not conscientious; you don’t carry out what
you’ve undertaken,” and which conjured for him a series of medical accidents for which he
could be blamed, might have reignited his childhood anxiety about not fulfilling his promise,
along with the wish that his father be wrong (perhaps along with other, related, wishes. . .).
10. Later (1900, S.E., IV: 294), Freud reports that ‘propyls’ brought to mind ‘Proplylaea,’
which reminded him of the Propylaea in Munich that he had seen when he visited Fliess there.
Erikson (1954) observes that the Greek proplyon (vestibule) symbolizes the female genitals (as
in, vulvar vestibule), while propionic can be associated with priapic (that is, phallic), implying
a sexual theme. Both these observations fit with Freud’s interpretation of trimethylamin.
11. Others have observed analogies between the dream and the Emma Eckstein affair of the
preceding months. These include Otto’s causing an infection through a dirty syringe being like
Fliess’s negligently causing an infection in Emma by leaving a gauze in her nose, and the
examination of the distinctly nasal anatomy of Irma’s throat being like the examinations of
Emma’s nose by Fliess and other doctors. There are contrasting views as to how this might fit
in, or not, with Freud’s interpretation (Schur 1972, 79–89; Anzieu 1986, 71–72; Griffin and
Tyrrell 2007).
Part I
Meaning
Copernicus’s theory of the heavens was neither correct nor original. It was
not original because Aristarchus of Samos had presented the same basic
theory some 1,800 years before Copernicus. It was not correct because,
among other things, one of its fundamental posits was that the planets, in-
cluding the earth, move in uniform circular motion. Why then do most of us
have such a high regard for the theory?
The answer is twofold. Copernicus was partly right. He was wrong about
uniform circular motion, but right that the earth moves around the sun and
about its own axis. Of course, Aristarchus was also partly right. But—and
this is the second half of the equation—Copernicus’s ideas were the begin-
ning of a new way of thinking. His theory was the stimulus for subsequent
astronomers like Galileo and Kepler to develop the arguments that brought
about the demise of the previous view of the heavens and augured a radically
different one, from which we have not looked back. We hold Copernicus in
high regard not because he was right, but because he pointed us in the right
direction.
Copernicus’s theory is not exceptional in this respect. A similar truth
holds of other historically important scientific theories. It holds of Newton’s,
of Lavoisier’s, and of Darwin’s. 1 What is common to all is that they are at
least partly right, and right in a way that bears a direct relation to what came
after. The mark of a good theory is not that it is completely correct, nor that it
is wholly original, but that it focuses attention on hitherto neglected truths
that open up new vistas. It need only be partly right, but those parts it gets
right must be important.
Freud’s theory of dreams may well fit this historical pattern. It is thus an
aim of this part of the book, not just to describe Freud’s theory, but to
determine what about it is most important. This requires engagement with
3
4 Chapter 1
Freud’s reasoning. Did he establish that dreams have meaning? Did he give
any evidence for his view that most dreams are the product of censorship and
disguise? Did he offer good reasons for thinking they are wish-fulfillments?
Such engagement is not readily available in the literature. As Rachel
Blass observes, “an in-depth and detailed study of the very structure of
Freud’s arguments regarding the nature and justification of his theory” is
needed, but “surprisingly, such an in-depth study is absent from the psycho-
analytic literature” (2001, 142). 2 I do not pretend that my account in this
chapter and the next fills this gap, but it is perhaps a beginning. What I offer
is not an in-depth study, but an exposition that delineates the key parts of
Freud’s theory and reasoning. The purpose in so doing is twofold: to facili-
tate a proper assessment of the main criticisms brought against the theory and
to present a positive case for its reasonableness.
Ideally, an in-depth study of the kind that Blass calls for would be histori-
cal, placing the development of Freud’s theory and arguments in their proper
context. I have, however, veered away from this. Much has already been
written about the history of Freud’s theory. 3 My concerns are different. What
I want is not details of how Freud’s ideas arose, but insight into how they
might be justified. This chapter and the next are preparation for this, and in
this respect, clarity of exposition and engagement with reasons matter more
than the subtleties of change and context.
The Theory
Freud made many claims about dreams, but there are five that stand out as
central to his theory. They are as follows (1900, S.E., IV–V: 96; 141–44;
121; 219, 553; 579):
D1. dreams are interpretable;
D2. dream-distortion is the product of censorship;
D3. dreams are wish-fulfillments;
D4. ultimately, dreams are the fulfillments of infantile wishes;
D5. the function of dreams is to preserve sleep.
The order matters, as the later claims depend conceptually and logically on
the earlier ones, with the exception that the last does not depend on D4, since
D4 is a dispensable—if psychoanalytically important—elaboration of D3.
In what follows I explain what each of these claims means and recon-
struct the reasoning that underlies it. I do not thereby endorse any claim or
argument. At this stage, my aim is expository. Nevertheless, in reconstruct-
ing Freud’s reasoning I adhere to the principle of charity, that is, the principle
of interpreting another’s sayings so as, as far as is reasonable, to maximize
their truth and rationality (Davidson 1984). This does not mean overlooking
Meaning 5
poor inferences, but it does mean selecting among the many feasible interpre-
tations that which renders the author’s arguments most cogent.
The claim that dreams are interpretable consists of two smaller claims: that
dreams have a meaning and that this meaning, where not immediately appar-
ent, can be discovered by a process of interpretation. These are not trivial
claims. As Freud points out, most scientists of his time denied them. In so
doing they opposed the folk wisdom of the ages—the extensive traditions of
dream interpretation found in cultures throughout the world. In contrast,
Freud upheld this folk wisdom. But what exactly does Freud mean by “mean-
ing?” And just what is the process of interpretation?
Consider the following examples of dream meanings that Freud gives:
I am not responsible for Irma’s pains, but Otto is. (1900, S.E., IV: 118)
I’d rather not give a supper-party than do so and provide my friend with the
opportunity to grow stouter and hence attract my husband still more. (1900,
S.E., IV: 148)
The reasons that R. and N. have not received professorships do not apply to
me. Therefore I can rejoice at the thought of becoming a professor. (1900,
S.E., IV: 140)
What can we discern from these examples? First, rather obviously, that the
dream meaning is a thought, or thoughts, that can be summarized by one or a
small number of (linguistically) meaningful sentences. Second, that it has to
do with issues of importance to the dreamer in his waking life, issues toward
which he has strong feelings—though as we shall see, not necessarily feel-
ings of which he is currently aware. These features of dream meaning are
not, however, sufficient to adequately define the notion. They are just surface
characteristics. They tell us about the form dream meanings take, but not
about their essence.
To understand the essence we need to enquire about the relationship
between a dream and its meaning. An examination of Freud’s dream inter-
pretations reveals two key characteristics. First, there are connections be-
tween the meaning and the dream such that the former makes sense of the
latter. That is, the connections are such that the dream represents the mean-
ing, in much the same way as a painting can represent the thoughts of an
artist. Second, the meaning is in fact a cause of the dream. A dream meaning
is, in short, a thought (or set of thoughts) that caused the dream and that
sheds light on its content.
6 Chapter 1
Dream meanings come in two varieties. One is meanings that require little
effort to discover. Children’s dreams are a good illustration of this. A mem-
orable example that Freud gives is of his nineteen-month-old daughter, Anna.
Because of illness, Anna had been denied food during the day. At night she
was heard saying in her sleep, “Anna Fweud, stwawbewwies, wild stwaw-
bewwies, omblet, pudden!” (1900, S.E., IV: 130). Assuming these utterances
arose from a dream, the meaning of the dream is obvious: these were foods
that Anna wanted to eat. Her wish to gorge on her favorite foods explains the
content of the dream and, we may surmise, was what caused her to dream in
the first place.
Adults too have dreams whose meanings are straightforward. We may all
remember occasions in which, having eaten salty food the evening before, we
dreamt of drinking water; or occasions in which, having a full bladder, we
dreamt of urinating. In such cases, the dream bears its meaning more or less
on its face.
The vast majority of dreams are not of this variety. Most dreams seem not
to have any meaning at all. According to Freud, however, to take this appear-
ance at face value would be a mistake. Such dreams have a meaning no less
than the others. They have a hidden meaning, though one that is discover-
able. This is what makes his claim about the interpretability of dreams so
interesting. That dreams with a straightforward meaning are interpretable
goes without saying, but not so for those far more numerous dreams with no
straightforward meaning. Freud asserts not only that they are interpretable,
but that they are so by a particular method.
The interpretative method was developed by Freud in order to cure neuro-
tic symptoms. In its application to dreams, it consists of the dreamer focusing
on some specific element in the dream and then free associating with this
element. Free association involves disengaging one’s critical faculties and
allowing one’s thoughts to emerge to consciousness. The process is not to be
treated as an intellectual game. It is not one’s capacity for solving problems
or making judgments that is to be exercised. Indeed, one should not make
any judgments upon, or in any way dismiss or reject, the thoughts that
emerge. Any thought that emerges, no matter how apparently trivial or em-
barrassing or otherwise unpleasant should be accepted as output. Thoughts
that emerge in this way are raw material for interpretation. This process of
free association is to be repeated for all elements of the dream, thereby
providing enough raw material for the dream to be interpreted.
Sometimes, thoughts that emerge from free association are the meaning
of the dream. But often this is not the case and discovery of the meaning
requires something more. This is not an algorithmic procedure. In interpret-
ing a dream one can follow the hints that prior experience of dream interpre-
tation supplies, but arriving at an interpretation is no more formulaic than
solving a rebus puzzle or a detective mystery. It requires ingenuity and
Meaning 7
imagination. One can reasonably believe that one has found a solution inso-
far as the proposed meaning of the dream accounts for all the elements of the
dream and makes sense of them. Recall, making sense, in the present context,
means connecting the elements of the dream to the thoughts that constitute
the proposed meaning—and hence to the rest of the dreamer’s thought-life—
in such a way that it is reasonable to think of the dream as representing the
content of those thoughts.
To summarize, all dreams have a meaning, where a meaning is a (set of)
thought(s) that caused the dream and in the light of which the dream makes
sense; moreover, this meaning, when hidden, can be uncovered by the
psychoanalytic method of interpretation.
Notice that to say that dreams can be uncovered by the psychoanalytic
method is not to say that they necessarily will be. In order for the method to
succeed, at least a couple of conditions must be fulfilled. First, as we have
noted, the method requires skill. Clearly, the requisite skill is not possessed
by everybody—for example, it is not possessed by young children. Second,
free association needs to be carried out correctly. One should avoid employ-
ing one’s critical faculty during this process to censor or dismiss thoughts.
Dream interpretation uncovers both a meaning and various distinct
thoughts that connect the meaning to the dream. Freud distinguishes between
the remembered dream material and the thoughts uncovered by the interpre-
tative process through his concepts of the manifest content and the latent
content of the dream (1901, S.E., IV: 641). By “manifest content,” Freud
means the images and thoughts of which the dreamer was conscious while
dreaming. By “latent content” he means the numerous thoughts that, because
they emerge during the interpretative process, are hypothesized to have been
part of the dream, even though they were not conscious at the time. This
latent content includes the meaning of the dream. Freud sometimes also calls
the manifest content the dream-content, and the thoughts that make up the
latent content, dream-thoughts (e.g., 1900, S.E., IV: 277). 4
Table 1.1.
Rationale for D1
Many would avow that they have tried this experiment and obtained
negative results. They have attempted to interpret their dreams but failed.
Does this not show the claim to be false? Freud would respond by emphasiz-
ing two cautionary points. One is that the technique requires skill. As with
any technique (such as microscopy or telescopy; see Chapter 6) care and
patience is required to learn it properly. The second is that there is a special
obstacle to successful interpretation. Recall that the method demands that
free association be carried out correctly. It requires the suspension of one’s
critical faculties during the process. This is not trivial. But the difficulty is
magnified when one considers that much of the material to be uncovered is,
according to Freud, of a nature to which one would be averse. In the course
of free association, thoughts are likely to arise that one would not want to
admit to consciousness, never mind reflect upon for the purposes of interpre-
tation (or have others reflect upon). They are, on Freud’s account, as we shall
see, thoughts that have been subject to self-censorship. One would therefore
be strongly inclined to reject or dismiss them. This offers a formidable obsta-
cle to any would-be interpreter, whether she is analyzing her own dreams or
those of another. As such, although the failure of an attempted interpretation
disconfirms D1, the degree of disconfirmation must be tempered by the
above considerations. 7
The correspondence of the eminent Swiss psychiatrist Eugen Bleuler de-
scribes experiments that illustrate the above difficulties. Bleuler was sympa-
thetic to Freud’s ideas, but not to the point of uncritical acceptance. As his
letters to Freud explain, he made considerable efforts to test his claims about
dream interpretation. The results were equivocal. In one letter (dated 9 Octo-
ber 1905) he writes:
It is not quite correct that I could not have analyzed anything at all from my
dreams; I was able to interpret several completely as to their meaning. In the
first, to be sure, I had the misfortune that the explanation that seemed entirely
plausible was surely false, because it referred to an event that occurred after
the dream. In one case, I had presented the dream to the assistant doctors and
to my wife. In my presence no progress was made. So I had to leave the room
for quite some time, & when I returned the dream had been construed, but in
10 Chapter 1
such a way that it could not at all correspond to my thinking: it was quite clear
that they had read into it the complexes of my wife, who had taken the leading
role during the analysis. That was the beginning. We did not encounter such
lapses again. Whenever an interpretation or part of it is unclear, in our experi-
ence, it is the dreamer who mostly has the definite feeling of correctness: “the
explanation hits the nail on the head?” (ibid., 163–64)
It seems, then, that Bleuler was able to confirm for himself—as were his
colleagues and his wife for themselves—the qualified claim that some
dreams are interpretable, but not the more general one that all dreams are
interpretable. Despite his only partial success, and the numerous objections
he raises, Freud would surely have approved of his attempts. The admonition
to others would be to do the same: You should test the claim by trying the
method for yourself.
Condensation
The dream is brief, yet the analysis of it occupies several pages and uncovers
a number of thoughts. The thoughts uncovered were inferred by Freud to be
dream-thoughts, and hence the material out of which the dream-content was
fashioned. Chief among the operations that constituted this dream-work was
condensation.
Let us see how condensation is purported to have worked in this dream.
One application of it is manifest from the following observation that Freud
makes: “The face that I saw in the dream was at once my friend R.’s and my
uncle’s. It was like one of Galton’s composite photographs (ibid., 139).” The
uncle in question was his Uncle Josef. With Uncle Josef, Freud associated
two memories: that he had been punished for some illegal activities and that
he had been described by Freud’s father as a simpleton. With R., he associat-
ed the thought that this friend had been recommended for a professorship but
after several years of waiting had yet to receive it. The thought was poignant
for Freud, because he himself had just been recommended for a professorship
and was worried he might suffer the same fate as R. The associations thus far
do not have anything in common, but there was one further association that
did: Uncle Josef and R. had similar aging yellow-grey beards (as did Freud).
On account of this, it seems, condensation had brought together these two
dream-thoughts and reduced them to one image, that of a composite of Uncle
Josef and R.
12 Chapter 1
There is, though, more. Freud associated with R. another colleague, N.,
who, like Freud and R., had been recommended to a professorship but, like
R., had after several years yet to receive it. N.’s explanation of this delay was
that a woman had started legal proceedings against him, which, though dis-
missed, had left a blemish on his character. On the basis of these details,
Freud draws the following conclusion: “My Uncle Josef represented my two
colleagues who had not been appointed to professorships—the one [R.] as a
simpleton and the other [N.] as a criminal” (1900, S.E., IV: 139).
Both these accusations against his colleagues represented egoistic wishes
ensuing from Freud’s deeper wish that his fate would not be the same as
theirs. The first accusation, that R. is a simpleton, is an exaggeration of
Freud’s judgment of his friend; the second, that N. is a criminal, is an indict-
ment stemming from the real-life allegation against him.
Thus several sets of thoughts had, according to Freud, been condensed
into the brief dream-content. This act of dream-work had involved omitting a
large number of details, including the important ones about R.’s and N.’s
recommendations for professorship. It had also involved fusing together
dream-thoughts. This latter process appears to have been a complex one.
Freud’s dream-thoughts had been concerned with providing reasons for why
R. and N. had not obtained the professorships to which they, like him, had
been recommended. He had wanted to avoid the obvious explanation that it
was due to “denominational” reasons—that is, anti-Semitism—since such
considerations applied equally to him, and would make it likely that he too
would fail to gain a professorship. Fortunately, N. had supplied an alternative
reason, the allegation of criminal behavior. This, it seems, had brought to
Freud’s mind his Uncle Josef (the thought of whom may already have been
activated, in virtue of the yellow-grey beard, by his thoughts of R. and
himself). Such being so, and his needing to find a reason to explain away R.’s
failure, he had alighted upon the other salient characteristic of Uncle Josef:
that he was a simpleton. The work of condensation had therefore brought N.
and Uncle Josef together in one movement, and Uncle Josef and R. together
in another, to form the dream element of the composite image of Uncle Josef
and R. The first movement had been based on the common feature between
Uncle Josef and N. of criminal proceedings, and had culminated in the elimi-
nation of N. from the dream altogether. The second movement had been
based on the common features of the yellow-grey beards and the convenience
of the charge of being a simpleton, and had culminated in a composite image.
This is a nice illustration of how condensation works.
Incidentally, the above example illustrates another feature of condensa-
tion, namely, the more dream-thoughts brought together by condensation into
one element of the dream-content, the more intense that element is. This is
exemplified by the yellow beard that “stood out especially clearly” in the
dream. This element had brought together R., Uncle Josef, and Freud him-
Meaning 13
self. It had brought together R. and Freud not just because they both had a
yellow beard, but also because they had both been recommended for profes-
sorships and were both Jewish (hence subject to anti-Semitic discrimination),
and it had brought together R. and Uncle Josef for the reasons given above.
This element of the dream was particularly intense, according to Freud, be-
cause three prominent dream-thoughts had been fused into one element of the
dream.
Of course, the acceptability of this demonstration of condensation de-
pends on the acceptability of Freud’s interpretation. On this issue, again I
must defer to the discussion of Chapters 3 and 4.
Displacement
Displacement refers to two related but distinct processes. In its first and
simplest meaning, it is the operation by which a dream-thought is trans-
formed into another thought on the basis of some superficial similarity be-
tween them. To give an example from one of my dreams, the thought “bar-
barian” was displaced (on my interpretation) to the image of baboons on
account of the superficial similarities in the sound of the two words (see
Chapter 4).
In its second and more important meaning, displacement is the operation
by which the psychical intensity of a dream-thought is displaced to other
thoughts. Psychical intensity refers to the value, or importance, that a thought
has for the dreamer, as measured by the strength of his feelings. Freud posits
displacement in this second sense in order to explain a particular fact. This is
that “what, on the evidence of our feelings, can claim to be the most promi-
nent among the dream-thoughts is either not present at all . . . in the content
of the dream or is only remotely alluded to in some obscure region of it”
(1901, S.E., V: 654). For example, in one dream that Freud recounts, the
central theme is that of a woman making advances on him. But the central
theme of the dream-thoughts is different from this: it is the “wish for once to
enjoy unselfish love, love which ‘costs nothing’” (ibid., 655). Freud thinks
the other operations of the dream-work insufficient to explain this shift in the
“center of mass” of dreams. For this reason he hypothesizes the operation of
displacement.
Freud’s argument for the existence of displacement (in this second sense)
is, in its full version, rather subtle. It is, though, also important, as we shall
see, because displacement plays a pivotal role in Freud’s subsequent claims.
It would therefore be helpful to understand it more fully.
The observation to be explained is as follows: psychical values differ
between dream-thoughts and dream-content. In other words, what is of in-
tense interest in the dream-thoughts is often peripheral in the dream, and
vice-versa. What would explain this? An obvious hypothesis is suggested by
14 Chapter 1
Dramatization
Symbolization
One of the means through which dramatization works is the use of symbols.
In this context, a symbol is a concrete concept that has come to be associated
with another (usually more abstract) concept by regular association. It could
be unique to an individual or it could be culturally widespread.
Freud, while recognizing the relevance of symbols, did not initially put
much emphasis on them. Subsequently he came increasingly to do so, to the
point where symbolization may be considered a distinct operation of the
dream-work. Among the most well-known of symbols that he speaks of are
elongated objects, such as swords, trees, and poles, to represent the penis,
and containers, such as boxes, baskets, and ovens, to represent the uterus.
Where common symbols occur in dreams, they may facilitate interpretation
without the need for a full analysis. Freud cautions, however, against over-
using symbols in this way. What may be taken as a common symbol in a
dream can, upon analysis, turn out to have a different interpretation. Also,
16 Chapter 1
Secondary Revision
According to Freud’s first claim, all dreams have meanings. For a number of
dreams, such as those of children, these meanings are easily discernible. For
others, however, they require the resources of psychoanalysis to interpret. Of
the latter kind, we may ask: Why do such dreams not display their meaning
in a straightforward way, like the first kind? That they do not is what Freud
means by dream-distortion (1900, S.E., IV: 136). The question is, therefore:
Why is there dream-distortion?
One possible answer comes from the operations of dramatization and
symbolization. Dramatization distorts thoughts in the process of rendering
them visual (or in some other way amenable to representation), and symbol-
ization similarly distorts thoughts in the process of replacing them by sym-
bols. Appeal to these operations, however, can only provide a partial solu-
tion. If it were all, then one might expect dreams to be far easier to interpret
than they are. Freud therefore postulates another cause of dream-distortion.
This is the presence of a motive to repress. On his view, the meaning of a
dream often contains thoughts that the dreamer is strongly averse to—or as
Freud more vividly expresses it, thoughts that “surprise me, which I have not
known to be mine, which not only appear foreign to me, but which are
unpleasant, and which I would like to oppose vehemently” (Freud 2001, 32;
author’s italics). The desire to oppose these thoughts manifests as the motive
to keep the thoughts from consciousness. This motive, he believes, is what
causes dream-distortion.
Freud had come upon his theory of repression through his clinical work. It
became most evident to him when, in his attempts to analyze his patients, he
frequently encountered resistance. This resistance, he postulated, was due to
a motive in his patients to keep something from becoming conscious—the
motive to repress. According to Freud, this motive is also present during
sleep.
How, though, does the motive to repress lead to dream-distortion?
Freud’s answer is semi-metaphorical. He conceives of there being two agen-
cies at work within the same individual. One, driven by the motive to repress,
acts to prevent any distressing material from becoming conscious in the
dream. The other, driven by the motive to express, acts to bring about the
expression of the dream meaning in the dream. Dream-distortion is the result
of the conflict between these two agencies. If the dream meaning is to be
expressed despite the opposition of the first agency, it must take a round-
about route. It is to this end that the dream-work, in particular the operation
18 Chapter 1
of displacement (in both its senses), works. The meaning cannot be ex-
pressed in its naked form, as the first agency would prevent this. It therefore
finds an alternative route to expression by displacing to material that bears
superficial connections to it. Thus changed, it can elude the repression of the
first agency and find its expression.
Freud illustrates this idea with the following memorable analogy:
A similar difficulty [to that faced by the second agency] confronts the political
writer who has disagreeable truths to tell to those in authority. If he presents
them undisguised, the authorities will suppress his words—after they have
been spoken, if his pronouncement was an oral one, but beforehand, if he had
intended to make it in print. A writer must beware of the censorship, and on its
account he must soften and distort the expression of his opinion. (1900, S.E.,
IV: 142)
Based on this analogy, Freud calls the first agency the censor. While this is a
metaphorical label, Freud’s talk of agencies is not entirely metaphorical.
What Freud is invoking is a conception of the mind as partitioned. We
normally think of ourselves as a unified whole. On occasions, however, we
conceive of ourselves as divided into parts. The clearest case of this is self-
deception. In self-deception we deceive ourselves in order not to face up to
an unpleasant consequence. The very notion of self-deception carries within
it the idea of the mind as divided. Part of us “knows” the truth to which the
evidence points, but this part also realizes how unpleasant it would be for us
to acknowledge this truth, and so arranges the deception. The deception
would be impossible without our conceiving of the existence of an isolated
part of the self acting without the knowledge of the rest. This part is con-
ceived as an agent acting on the basis of motives. It is an agent within the
agent. Freud, therefore, has as much right as those who employ the common-
sense notion of self-deception to use the idea of agents within agents to
explain the phenomenon of dream-distortion. It is more than a metaphor: it is
an explanatory resource taken from commonsense psychology.
One must remember, however, that talk of agencies is, according to
Freud, ultimately underwritten by a mechanistic account of the mind’s opera-
tions. Thus instead of speaking of agencies, one could instead speak of
countervailing forces. Or, at a level intermediate between the mechanistic
and the agential, one could speak of the censor as a function of the mind. All
of these concepts—forces, functions, agents—simply apply to different lev-
els of description. They are all of a piece, though serving different kinds of
understanding.
Freud’s explanation of dream-distortions, then, is that the majority are a
consequence of the presence of the censor. Ordinarily—that is, during wak-
ing life—the censor would prevent any objectionable thought from finding
conscious expression. But during sleep, its powers are diminished. The sec-
Meaning 19
ond agency, that which seeks expression, attempts to express the otherwise
thwarted thoughts in the dream. The censor, being not completely inert, yet
prevents the thoughts from making their way to consciousness unchanged.
To become consciously accessed, they must find an alternative route in
which distortion by the dream-work has so disguised them that they can
elude the censor’s repressive acts. This, according to Freud, is the primary
reason why most dreams do not openly display their meaning.
Table 1.2.
Rationale for D2
I have already alluded to part of Freud’s rationale for the censorship claim. It
is intended as an explanation of dream-distortion. An alternative explanation
that Freud considers is “that some incapacity exists during sleep for giving
direct expression of our dream-thoughts” (1900, S.E., IV: 136). One such
incapacity is that which I mentioned earlier, the inability of abstract thoughts
to be expressed in dreams but for the aid of dramatization. Freud thinks that
this alternative is an explanation of dream-distortion, but only a partial one.
This is not because of some inherent flaw in the explanation, but because the
evidence provided by dream interpretations points to another source.
As Freud puts it, it is “the analysis of certain dreams” that compels the
censorship explanation of dream-distortion (ibid.). That is, certain dreams
have subtle details that are best explained as the consequence of censorship.
An example will illustrate. Recall the dream of the yellow beard described
earlier. In his report of his interpretation of this dream, Freud mentions a
number of details pertinent to our present concerns. One has to do with his
initial response to the dream. As he explains, “When, during the course of the
morning, the dream came into my head, I laughed aloud and said: ‘The
dream’s nonsense!’” (ibid., 138). This is a normal enough response for most
people, but, given his views, not for Freud. He followed this thought by
putting off any attempt to interpret the dream. By evening, though, he came
to see this as a form of resistance: “I began to reproach myself: ‘If one of
your patients who was interpreting a dream could find nothing better to say
20 Chapter 1
than that it was nonsense, you would take him up about it and suspect that the
dream had some disagreeable story at the back of it which he wanted to avoid
becoming aware of’” (ibid.). The eventual interpretation reinforced this in-
itial suspicion, for it not only revealed an ambition that was surprising to his
self-conception (he had prided himself on not putting much stock on his
recommendation for a professorship), but also a disturbingly hostile judg-
ment of his friend R., whom he had turned into a simpleton.
There remained, however, one puzzling feature of the dream that had yet
to be explained. This was the “great feeling of affection” he had felt toward
R. in the dream. What would explain that? According to Freud, “My affec-
tion for him struck me as ungenuine and exaggerated—like the judgement of
his intellectual qualities which I had expressed by fusing his personality with
my uncle’s, though there the exaggeration had been in the opposite direc-
tion” (ibid., 140–41). Thus there appears to be a relationship between the
exaggerated affection and the exaggerated negative judgment: it is as if the
affection is compensating for the negative judgment. But there is more.
Freud comes to realize that it was largely on account of his bad feelings
about this negative judgment that he had so resisted interpreting the dream.
In light of this, it is plausible that the exaggerated affection in the dream was
there precisely to mask the negative judgment that he so wanted to avoid
acknowledging. This trades on the following simplistic reasoning: if at the
forefront of the dream was a feeling of affection toward R., then it would be
unlikely that the dream involved a negative judgment about this beloved
friend.
Notice here the similarities between this explanation and one that might
be offered in commonsense psychology. When one observes that another
person is expressing herself in an ungenuine and exaggerated way, one’s
suspicions are raised. An explanation that might readily come to mind is that
the exaggeration is there to hide thoughts with precisely the opposite con-
tent—as in the popular understanding of “The lady doth protest too much.”
In Freud’s example, he uses this commonsense explanation to account for the
great feeling of affection in his dream. He is protesting of his affection for R.
too much, indicating that he is actually trying to hide opposite thoughts. It is
as if this element were in the dream due to the dissimulation of part of
himself. On coming up with this explanation, Freud—who after all is the
purported dissimulator, even if only in part—seems to have had no doubt of
its truth.
Interestingly, this explanation bears no relation to the dream’s meaning. It
is, rather, of a piece with Freud’s wakeful resistance. His laughing at the
dream and calling it nonsense, his disinclination to analyze the dream, and
the exaggerated affection in the dream were all attempts to prevent himself
from confronting the negative judgment he had made of his friend. It was, in
other words, his motive to repress, his desire to keep unpleasant thoughts
Meaning 21
from consciousness, that was behind both his wakeful behavior and the re-
maining puzzling element of the dream. Or to put it another way, it was that
part of himself that was acting on the motive to repress—the censor—that
was responsible for this element. The censor was here, on Freud’s view,
directly influencing the dream-content.
As well as providing evidence for dream censorship, the example illus-
trates a general point. Freud observes, “It happens often enough that a pa-
tient, despite all his efforts, cannot remember one of his dreams. But after we
have been able in the course of a piece of analytic work to get rid of a
difficulty which had been disturbing his relation to the analysis, the forgotten
dream suddenly re-emerges” (1933, S.E., XXII: 14). In addition, he observes
that during the process of free association it often happens that “there is a
stoppage and the patient hesitates before bringing out an association, and, if
so, we often have to listen to a long chain of ideas before receiving anything
that helps us to understand the dream” (ibid., 13). These observations are
evidence of resistance. So the resistance that Freud experienced in relation to
his dream in the above example seems to be typical. This makes more plau-
sible the explanation that distortion is the product of censorship—especially
since the above example offers evidence that the force behind resistance can
have an influence on the very content of the dream.
Freud similarly finds evidence of censorship in the subtle details of other
dreams. One may ask, though, if his thoroughly detailed explanation of indi-
vidual dreams would be enough to warrant the general claim that dream-
distortion is mainly due to censorship. To this end, Freud has another argu-
ment to supplement his case-by-case findings.
This second argument begins with a general observation. Freud finds that
all obscure dreams—all dreams for which the meaning is unclear—when
analyzed, eventually turn up secrets. That is, when interpreted, their meaning
is invariably found to contain thoughts that the agent finds highly objection-
able—recall Freud’s description: they are thoughts that “surprise me, which I
have not known to be mine, which not only appear foreign to me, but which
are unpleasant, and which I would like to oppose vehemently.” Freud is
familiar with such thoughts from his clinical work. They are thoughts that
have been repressed. This offers reason to think that obscurity is related to
repression. From this relation, he infers that the motive to repress is the cause
of the obscurity (1901, S.E., V: 672).
This may at first sight appear to be a weak inference, no better than
deducing causation from correlation. It has, though, I think, more merit. At
least, it does when taken in conjunction with a second observation: repressed
thoughts never appear in dreams undistorted. Thus where there is obscurity,
repressed thoughts are present, and where there is no obscurity, repressed
thoughts are absent. This in itself provides reason to infer that the presence of
repressed thoughts is causally relevant to the obscurity of the dream. But
22 Chapter 1
there is more. Freud is concerned to answer the following two questions: (1)
Why should it be that some dreams are obscure? (2) How is it that repressed
thoughts, which ordinarily cannot find expression on account of repression,
are able to find expression in dreams, even if only in a distorted form? The
perfect alignment between dream obscurity and repressed thoughts suggests
an explanation that addresses both questions simultaneously. This is that the
presence of repressed thoughts is indeed causally relevant to the obscurity of
the dream and the thoughts are able to find expression precisely because of
the obscurity thereby produced. The most plausible elucidation of this causal
relevance is the censorship-disguise thesis: that the motive to repress causes
the obscurity by necessitating the disguise of the repressed thoughts. Thus
censorship-disguise can be inferred as the best explanation of the reported
facts.
Of course, this argument depends on the facts being as Freud reports. In
particular, the premise that all obscure dreams have, upon interpretation, a
meaning that contains repressed thoughts is one that can only be established
by induction over numerous dream interpretations. The same epistemological
problems apply here as those we encountered for D1. One cannot simply take
the interpretative evidence from Freud on trust: it needs to be tested by
others. But here the obstacles mentioned in the rationale of D1 are even more
serious, for if Freud is right, then the success of one’s interpretations would
depend on overcoming precisely the aversion that one is testing for.
Thoughts are repressed because they are highly aversive, so if one is to
uncover such thoughts the dreamer must, to some degree, overcome this
aversion. This is both a practical and an epistemological problem. Practical-
ly, it is difficult to overcome the aversion—hence the prevalence of resis-
tance. Epistemologically, a false outcome—an obscure dream that one fails
to interpret as having a meaning involving repressed thoughts—can be ex-
plained as being due to one failing to overcome the aversion, making it
difficult to see how anything can count against the claim.
About the practical problem, little can be done, but about the epistemo-
logical one, Freud has a partial answer. We can know that a dream has not
been successfully interpreted by the presence of unexplained facts. One may
think that one has given an interpretation of the dream, but if some aspect of
the dream remains unexplained, then that shows, on Freud’s theory, that
one’s interpretation is not complete (1900, S.E., IV–V: 269, 553). Converse-
ly, if one’s interpretation adequately explains all aspects of the dream, then
there seems no good reason to regard it as incomplete. This provides a basis
for testing the claim. If one has an interpretation of an obscure dream that
explains all aspects of the dream and yet has no aversive thoughts as part of
its meaning, then the premise is disconfirmed. The problem that remains,
however, is that many of the judgments involved in reaching a correct inter-
Meaning 23
pretation are subtle, and hence the possibility of an erroneous negative judg-
ment is high.
There is solace for Freud despite these problems. Even if it were only
found that some obscure dreams contain repressed thoughts, then, though
that would not support the assertion that censorship-disguise is the explana-
tion of dream obscurity, it would support the assertion that censorship-dis-
guise is an explanation of dream obscurity. For such would account for how
repressed thoughts find expression in dreams and why they do so only in
obscure ones. What would remain is the possibility of additional explana-
tions of dream obscurity. Freud himself allowed the additional explanation
that some obscurity arises due to dramatization and symbolization. As he put
it, “even if there were no dream-censorship dreams would still not be easily
intelligible to us, for we should still be faced with the task of translating the
symbolic language of dreams into that of our waking thought” (1916, S.E.,
XV: 168). His claim was that censorship-disguise was the main explanation,
not the only one. This assertion is supportable as long as most obscure
dreams turn out, upon interpretation, to contain repressed thoughts. Given the
problems highlighted above, though, it may be that, as with D1, one must at
this stage settle for only a qualified version of the claim: at least some dream-
distortion is due to censorship-disguise. For this the interpretative evidence
that Freud provides would suffice. 10
Table 1.3.
3. D2 would explain well (and be the best explanation of) the existence of
dream-distortion and the observation that where there is dream-distortion, the
meaning of the dream always contains repressed thoughts.
4. D2 would explain well (and be the best explanation of) the existence and
nature of displacement.
NOTES
1. Important elements of Newton’s theory were anticipated by Robert Hooke, and the
theory in its exact form has now been transplanted by Einstein’s. Lavoisier’s chemical theory
was similarly wrong in certain fundamental respects and anticipated in part by figures such as
Lomonosov. Evolution had been discussed much before Charles Darwin, not least by his
grandfather Erasmus. Darwin himself admitted that Patrick Matthew had anticipated—though
not properly developed—natural selection in 1831, before he had even entertained the idea. We
now know that important aspects of both evolution and natural selection as Darwin conceived
them are wrong.
2. Blass herself has since attempted such a study (2002, 63–116). Her account of Freud’s
reasoning raises important points, particularly about the justification of his method of interpre-
tation, but is not comprehensive: it neglects many of Freud’s arguments for his other key claims
(D2–D5).
3. For example, Jones (1953), Anzieu (1986), Ellenberger (1970), Sulloway (1992), For-
rester (1997), Grubrich-Simitis (2000), and Marinelli and Mayer (2003).
4. More strictly, latent content “refers to all parts of a dream which are not manifest but are
only discovered through the work of interpretation,” whereas dream-thoughts “refers only to
the preconscious components of the latent content” (Nagera 1969, 28). This technical distinc-
tion need not trouble us, especially as Freud often used the terms interchangeably.
5. In his later writing Freud appears to admit that this is not so: “‘Can we interpret all
dreams by [our method’s] help?’ . . . ‘No, not all; but so many that we feel confident in the
serviceability and correctness of the procedure’” (1933, S.E., XXII: 13; see also 1925, S.E.,
XIX: 128–29).
6. In The Interpretation of Dreams, Freud claims to “already have analysed over a thou-
sand dreams” (1900, S.E., IV: 104).
7. Freud argues for something stronger than this. He argues that one can be justified in
believing that all dreams are interpretable because, in the course of a patient’s treatment, one
often finds that dreams that initially fail to yield an interpretation do so later, and because
children’s dreams are always straightforwardly interpretable (1925, S.E., XIX: 129). The first
reason may be construed as inductive support for believing that failure of interpretation is not
due to uninterpretability. The second reason must be taken in conjunction with claim D2, that
most adult dreams are not straightforwardly interpretable because of censorship and disguise. It
would be strange to suppose that children’s dreams invariably have meaning, but then adult
dreams emerge that do not. Rather, in light of D2 and the other purported facts, the better
explanation for occasional failures to interpret an adult dream is resistance. While these reasons
do offer some support for the claim that all dreams are interpretable, I do not think they are
sufficient justification. The inductive evidence given in his first reason is not particularly
strong. Also, Freud’s sample for determining the character of children’s dreams was likely too
small, and Freud anyway admitted that not all children’s dreams were straightforwardly inter-
pretable (1900, S.E., IV: 127n1; recent scientific evidence suggests this is correct [see Chapter
8]). Moreover, Freud himself acknowledged that some dreams are not interpretable by his usual
method—namely, typical dreams—raising the prospect that there are dreams that are not inter-
pretable at all.
8. Freud gives similar examples of hypnagogic hallucinations based on the experiments of
Herbert Silberer (1900, S.E., V: 344–45; 1933, S.E., XXII: 23).
26 Chapter 1
9. Originally, Freud spoke of secondary revision as part of the dream-work without qualifi-
cation. But later he was to say that it was not strictly part of it. There are two main reasons for
this. First, the dream-work, on its strict conception, works according to a distinctive set of
principles—“primary process laws”—whereas secondary revision works according to a differ-
ent set of principles—“secondary process laws.” Second, condensation, displacement, and
representability, are always operative wherever there is dream-work, but not so with secondary
revision.
10. Through his numerous dream interpretations in The Interpretation of Dreams, Freud
offers considerable evidence that obscure dreams contain repudiated thoughts. Most of these
thoughts, however, do not meet his description of repressed thoughts strictly interpreted—they
are not entirely “foreign” to the dreamer, though they are thoughts of which he was unaware
and would want to oppose. They are, according to the model we will consider in Chapter 2,
preconscious thoughts that have been recently repressed. In the Interpretation, Freud only
occasionally and indirectly indicates thoughts that meet his description more strictly (e.g.,
1900, S.E., V: 583–84)—though in his subsequent writing, he provides more direct evidence of
the presence of such thoughts (e.g., 1905, S.E., VII: 64–111).
Chapter Two
Wishes
Having examined D1 and D2, the theses that dreams are interpretable and are
the products of censorship and disguise, this is an appropriate place to reflect
on how important these are to Freud’s theory. The question of which claims
are most definitive of the theory matters to its appraisal, since many critics,
in dismissing the theory, focus on only one or two of its claims. For example,
Adolf Grünbaum, the leading philosophical critic of Freud, considers the
theory of dreams to be well defined by the assertion that for all dreams the
content is the fulfillment of a wish and the instigator is a repressed infantile
wish. If he is right, then undermining this assertion is sufficient for under-
mining the theory as a whole. But others contest this view, and there is
reason to see D1 and D2 as more important.
Rachel Blass argues that the first claim, in particular, is central: “The
essence of the psychoanalytic theory of dreams is the claim that the dream
contains accessible meanings. The claim that the nature of those meanings is
wishes is secondary, and the claim that wishes are what instigated the dream
is even further removed” (Blass 2002, 7). Freud’s own stated aims in The
Interpretation of Dreams are in line with Blass’s account:
The aim which I have set before myself is to show that dreams are capable of
being interpreted; and any contributions I may be able to make towards the
solution of the problems dealt with in the last chapter [which reviews the
literature on dreams in relation to a number of issues, including the function of
dreams] will only arise as by-products in the course of carrying out my proper
task. (1900, S.E., IV: 96)
Here Freud gives us reason to consider the interpretability claim as the most
important. On another occasion, however, Freud emphasizes a different ele-
ment. In Josef Popper-Lynkeus and the Theory of Dreams, he writes, “Pre-
27
28 Chapter 2
WISH-FULFILLMENT
criminal and that these were the reasons that they had not received professor-
ships. Since Freud was neither a simpleton nor a criminal, he was not subject
to rejection for such reasons. Thus the dream represented the content of his
wish. The representation in this case is complex and can only be understood
by one with the requisite background knowledge—as is true also of allusive
works of art. It is, on Freud’s account, a disguised representation; that is to
say, its indirectness is due mainly to the need to avoid censorship.
Table 2.1.
Rationale for D3
Cs can be considered together, since the only thing that divides them is the
placement of attention: those thoughts on which attention alights are in Cs,
those from which it is currently absent are in Pcs. This is not so with Ucs; the
thoughts in this system differ from the thoughts of Pcs precisely because they
cannot be made conscious simply by an effort of attention. They have been
made unconscious, and are being kept so, by the force of repression (1900,
S.E., V: 541-42). 1
With these theoretical resources in place, we can now return to the prob-
lem that Freud faced. What drives non-straightforward dreams? One thing is
clear, thoughts from the previous day, including non-wishful thoughts, do
find their way into dreams. Such thoughts, on Freud’s model, belong to the
system Pcs. But in finding their way into dreams, they are manifestly not
doing so in anything like the normal way in which such thoughts become
conscious. The thoughts of which we are speaking, such as anxieties, wor-
ries, and problems that need solving, are the kind of thoughts that are apt to
keep us awake. They are thus precisely what need to be diminished if we are
to sleep. So if the thoughts in question are to find their way into conscious-
ness during sleep, they must follow a route different to their normal one
(ibid., 555).
The situation of unconscious thoughts is otherwise. Freud claims that his
work on neuroses indicates that unconscious thoughts are operative through-
out our waking lives. 2 There is no obvious reason why they should not be
equally operative at night. Unlike preconscious thoughts, their modus oper-
andi is not to express themselves in conscious propositional thought, so the
diminution in the normal processes of thinking does not imply a diminution
in their activity. Indeed, the interpretative evidence indicates that they are
highly active during sleep and seeking expression in dreams—which is why
the censor is needed.
There is thus a difference between preconscious and unconscious
thoughts at night: the first are diverted from their normal path, the second are
not. Freud thinks the way sleep works is to “de-energize” thoughts—that is,
to reduce in them that which makes them a force in our mental life. His
explanation of why it is that preconscious thoughts do not follow their nor-
mal path at night is that they are de-energized in this way. This is what needs
to occur if they are not to disturb sleep. But unconscious thoughts are not
(significantly) de-energized. 3 Indeed, unconscious thoughts are more potent
than usual, since the repression that keeps them at bay during wakefulness is
greatly weakened in sleep. They are thus pressing for expression harder than
ever, while the preconscious thoughts less so. This makes all the more plau-
sible that it is unconscious, rather than preconscious, thoughts that are driv-
ing dream expression.
For Freud, since the mind is directed toward de-energizing preconscious
thoughts in order to preserve sleep, it becomes a puzzle as to how precon-
Wishes 33
sciousness in the normal way (via propositional thought) but in the unusual
one of an extended-narrative hallucination; unconscious thoughts are intrinsi-
cally strong and are pressing for expression harder than ever due to the
diminution of the force of repression; unconscious thoughts have, in the
continued presence of repression, no direct means into consciousness. The
hypothesis that the preconscious thoughts of the previous day, unable to enter
consciousness the normal way, do so instead by obtaining reinforcement
from powerful unconscious motives of the same tenor which are already
pressing for expression, fits well with these purported facts. It is therefore not
without merit.
There is perhaps even more merit in a related but weaker hypothesis. This
is that, regardless of whether or not some dreams are driven by preconscious
thoughts, at least many dreams are driven by unconscious thoughts. For this
we need not accept all the supposed facts listed above. It suffices to accept
only that, while unconscious thoughts are pushing for expression throughout
our lives, during sleep, due to the diminution of repression, they are pushing
for expression harder than ever. Taking this a step further, if the force of
repression during wakefulness was just enough to keep them at bay, then
upon its weakening, that is no longer so, hence the unconscious thoughts
must find a form of expression. The only one available to them is dreams.
Thus even if we are not prepared to go all the way with Freud to his general
conclusion that all (non-straightforward) dreams are driven by unconscious
thoughts, we may nevertheless get as far as the still significant result that at
least many dreams are driven by unconscious thoughts.
Let us take stock. The question that Freud is trying to answer is this: What is
driving the expression of thoughts in dreams? Wishes are certainly of a kind
that can do this, and indeed clearly do in the case of straightforward dreams.
But cannot it be that other thoughts can do so too? Freud considers the case
of preconscious thoughts (other than wishes), 5 but finds it wanting. He
argues that such thoughts are on their own too weak to drive dream expres-
sion. On the other hand, unconscious thoughts are strong enough to do so.
Indeed, the interpretative evidence suggests that unconscious thoughts are
always present in distorted dreams. Thus he posits that, other than certain
wishes, it is only unconscious thoughts that can drive dream expression.
This solution is, though, incomplete. So far all we have is that the driving
force for non-straightforward dreams comes from the unconscious. But why
should it be in the form of a wish? Freud informs that, “the reason why
dreams are invariably wish-fulfilments is that they are products of the system
unconscious, whose activity knows no other aim than the fulfilment of
wishes and which has at its command no other forces than wishful impulses”
(1900, S.E., V: 568). This answer depends on some deeper theorizing about
Wishes 35
els and tests them against input. This idea was introduced by Freud’s contem-
porary Hermann von Helmholtz and is known today as the “Bayesian brain.”
On Friston’s conception, the Bayesian brain attempts to predict its sensory
inputs through multiple, hierarchically organized, levels of representation
(Friston 2009, 2010). Predictions are propagated downwards through the
levels of representation, while the error between these predictions and the
inputs is propagated upwards. By way of this feedback process, the brain
seeks either to adjust its representations in line with the input (the basis of
perception), or the input in line with its representations (the basis of action).
Free-energy is, roughly, a measure of the prediction error—that is, the “sur-
prise” in getting certain input given the model used to predict it. According to
Friston’s “free-energy principle,” the brain works so as to minimize free-
energy. This principle sounds similar to Freud’s principle of constancy. Fris-
ton and his co-author, Carhart-Harris, are unequivocal: “the [Freudian] pro-
cess of minimizing ‘the sums of excitation’ is exactly the same as minimiz-
ing the sum of squared prediction-error or free-energy in Helmholtzian
schemes” (Carhart-Harris and Friston 2010, 1270). In particular, they draw
the correspondence between how “large-scale intrinsic networks” suppress
free-energy by top-down prediction and Freud’s conception of how the sec-
ondary system (the Pcs, or, later, ego) “binds” the motile, free energy of the
primary processes. They also draw numerous other correspondences between
neurobiology and Freud’s ideas (Carhart-Harris and Friston 2010; Hopkins
2012).
Aside from raising the prospect of an impressive vindication of Freud’s
ideas, these developments illustrate the point that concepts too readily dis-
missed upon an obvious current interpretation may have validity upon a less
obvious future interpretation. Even if one does not accept Friston’s theory or
its suggested correspondence to Freudian notions, the point stands that one
cannot take for granted one’s understanding of historical concepts and too
readily infer that they have been invalidated by subsequent discoveries.
Returning to the epistemological issue, one should understand that the
speculations that Freud and his contemporaries were engaged in were not
pure speculation. They were constructing explanatory models designed to
account for a broad range of phenomena. These were speculative only in the
relative sense that they were not as closely tied to the facts as theories
developed upon an alternative approach that attempted to carefully construct
theory directly from patterns observed in the data (Ellenberger 1970, 477).
Freud’s account of the mind as a reflex apparatus should therefore not be
dismissed out of hand. That the notions upon which it is based are in line
with sophisticated modern hypotheses suggests that he may have been on the
right track.
Furthermore, whether or not the theory of the mind as a reflex apparatus
is plausible as a whole, those parts of it that relate to Freud’s claim about
Wishes 39
wishes are plausible. These concern the parallel he draws between action and
a certain kind of hallucinatory phenomena, of which dreams are an instance. 6
On his theory, hallucination substitutes for action in circumstances where
action is impossible. A rudimentary analysis of action and wish-fulfillment
shows this to be a reasonable hypothesis.
Consider the action of drinking water. The typical process is as follows. A
bodily need for water gives rise to thirst, and subsequently to a desire to drink
water. This desire is directed toward an experience of satisfaction and sets
the mind in motion toward this end. It does so by combining with beliefs to
form a motivating reason for action, which one then acts on by drinking
water. In drinking water, one both satisfies one’s desire and meets the need
that gave rise to it. The satisfaction of the desire is accompanied by an
experience of satisfaction which pacifies the desire. Since the need that gave
rise to the desire has been met, the desire stays pacified.
Freud’s account of hallucinatory wish-fulfillment is, in part, analogous to
this process. A need gives rise to a wish. The wish, like the desire, is directed
toward an experience of satisfaction and sets the mind in motion toward this
end. Unlike for the desire, however, the route via objective fulfillment is
unavailable. For this reason, and in the absence of reality-testing, the wish
short-circuits directly to a perception of the object wished for. In so doing, it
represents itself as fulfilled, though the need that gave rise to it is not met.
The representation of the fulfillment of the wish is accompanied by an expe-
rience of satisfaction which pacifies the wish. Since the need that gave rise to
it is unmet, this pacification will not last long, but it will at least offer
temporary relief. 7
This account of wish-fulfillment gains support from our familiarity with
wishful imagining and wishful thinking. In wishful imagining, we imagine
what is wished for, and this brings relief. For example, vividly imagining a
romantic encounter can lead to an experience of satisfaction, even if one
knows the encounter is a fantasy. In wishful thinking, the wish directly
causes belief in what is wished for, and this too brings relief. For example,
one’s wish that one’s romantic interest be reciprocated can lead to a belief
that it is reciprocated, even if there is no evidence for this. Both wishful
imagining and wishful thinking are similar to hallucinatory wish-fulfillment:
the first involves a sensory representation much like wish-fulfillment; the
second involves a state that can perform the same function as the sensory
representation. They demonstrate how representation of fulfillment, even if
not accompanied by actual fulfillment, can pacify a wish (at least, in the short
term).
One can therefore see how hallucinatory wish-fulfillment can serve as a
substitute for action. Since in sleep action is (normally) impossible and real-
ity-testing is suspended, the conditions are ripe for hallucinatory wish-fulfill-
ment. This makes all the more plausible Freud’s thesis. His point is that
40 Chapter 2
One can summarize Freud’s reasoning as follows. Why and how do thoughts
find expression in dreams? Perhaps one can continue mulling over the day’s
events semi-consciously, but why should such thinking become vividly con-
scious? And given that the normal path to consciousness is blocked, how
does it do so? What, in other words, drives thoughts to find expression in
consciousness? The question becomes more challenging when one focuses
on the form of consciousness that such thoughts take. The consciousness in
dreams is not the same as the consciousness of thoughts in waking life.
Dreams are not made up of largely verbal ruminations, plans, and judgments.
They are dramatic visual narratives, of the quality of perceived events. Just
as waking hallucinations are startling phenomena, so too are sleeping halluci-
nations. The question then is: What drives thoughts to find expression in
consciousness in the unconventional form of an extended-narrative halluci-
nation?
The question should be considered in the light of what Freud takes him-
self to have established. He takes himself to have found that dreams invari-
ably have a meaning and that these meanings have thus far turned out to be
wishes. He also takes himself to have found that dreams contain both precon-
scious thoughts and unconscious wishes, and that the two are always related
in some way. With his claim that the driving force of dreams is a wish, Freud
is providing an answer to the question that is in line with these “findings” and
also addresses the more general question of why extended-narrative halluci-
nations should occur at all.
Freud integrates this solution into a yet larger picture:
Wishes 41
If we insist, for even a moment longer, upon our right to base such far-
reaching speculations [concerning the theory of mind summarized above]
upon the interpretation of dreams, we are in duty bound to prove that those
speculations have enabled us to insert dreams into a nexus which can include
other psychical structures as well. If such a thing as a system Ucs. exists (or
something analogous to it for the purposes of our discussion), dreams cannot
be its only manifestation; every dream may be a wish-fulfilment, but apart
from dreams there must be other forms of abnormal wish-fulfilments. And it is
a fact that the theory governing all psychoneurotic symptoms culminates in a
single proposition, which asserts that they too are to be regarded as fulfilments
of unconscious wishes. (1900, S.E., V: 568–9)
The obvious objection to Freud’s claim is that there are many dreams that, on
the face, are not wish-fulfillments. These include anxiety dreams, punish-
ment dreams, and counter-wish dreams (dreams in which a wish is presented
as frustrated). Freud’s response to this objection is twofold. First, he consid-
ers some examples of such dreams and shows that upon interpretation they
turn out to be wish-fulfilling after all. Second, he explains in theoretical
terms how the objection can be met.
We need not consider the first part of the reply here, for examples are to
be found throughout The Interpretation of Dreams, in particular Chapter IV.
It is the second part that is of most interest. It consists of explaining how the
two-agency theory described in the previous chapter shows the objection to
be a shallow one. The theory, recall, is that dreams are the products of two
opposed agencies: one, the censor, which acts on the motive to repress, and
the other which acts on the motive to express. There is already within this
picture the reason why many dreams are not straightforwardly wish-fulfill-
ing. The wish is disguised in order to elude the censor.
This disguise may consist of distressing material. Recall from the previ-
ous subsection that Freud argued that the best account of what drives dream
expression is that an unconscious wish attaches to preconscious thoughts that
are of the same tenor and is thus able to make its way past the censor to find
expression in the dream. The preconscious thoughts need not be wishes; they
may indeed be distressing or painful thoughts. What matters is that they have
some connection—not necessarily relating to wishfulness—with thoughts
associated with the unconscious wish. Thus distressing preconscious
42 Chapter 2
thoughts, driven by the unconscious wish, find their way into the dream-
content. The unconscious wish is fulfilled under cover of these distressing
thoughts.
A number of such dreams are recounted by Freud. For example, in one, a
woman dreamt with an indifferent tone of the death of her nephew. Upon
interpretation, the dream turned out to be about the woman’s wish to see
again, after a long absence, a man to whom she was strongly attracted and
who she had seen before under similar circumstances—that is, after a long
absence at the funeral of a nephew. The repressed wish to see the man had
found its way into the dream under cover of the thought of the death of her
nephew, as this thought connected with the wish through its similarity to a
past event that had coincided with what was wished for (1900, S.E., IV:
152–54, 248).
The above account explains the presence of distressing material in
dreams, but it does not yet explain why many dreams are of a negative
emotional tone. Why is it not the case that, though the fulfilled wish is
disguised, the emotion of the dream is nevertheless that of satisfaction? One
explanation is that the feeling of satisfaction on the fulfillment of the wish is
often obscured by the negative emotion of the preconscious thoughts. This
would explain many unpleasant dreams, but it would not explain those
dreams so distressing that they lead to the waking of the dreamer: the all-too
familiar anxiety dreams.
To explain anxiety dreams, the two-agency theory must be invoked in
another way. This is best illustrated by a fairy tale that Freud recounts:
A good fairy promised a poor married couple to grant them the fulfilment of
their first three wishes. They were delighted, and made up their minds to
choose their three wishes carefully. But a smell of sausages being fried in the
cottage next door tempted the woman to wish for a couple of them. They were
there in a flash; and this was the first wish-fulfilment. But the man was furious,
and in his rage wished the sausages were hanging on his wife’s nose. This
happened too; and the sausages were not to be dislodged from their new
position. This was the second wish-fulfilment; but the wish was the man’s and
its fulfilment was most disagreeable for his wife. You know the rest of the
story. Since after all they were in fact one—man and wife—the third wish was
bound to be that the sausages should come away from the woman’s nose.
(1900, S.E., V: 581n1; added 1919) 10
What this fairy tale illustrates is that, where there are two agents, there are
two sources of wish, and the fulfillment of one agent’s wish may be to the
displeasure of the other. This is the case in anxiety dreams. These are dreams
in which a repressed wish is fulfilled but the disguise is inadequate. The
anxiety comes when the censor recognizes that an abhorrent wish has been
fulfilled. 11 As Freud puts it, the “sleeping ego . . . reacts to the satisfying of
Wishes 43
the repressed wish with violent indignation and itself puts an end to the
dream with an outburst of anxiety” (ibid., 557). 12
As the fairy tale alludes, there may also be cases where the censor itself
has a wish fulfilled in the dream. This occurs in punishment dreams. 13 Ac-
cording to Freud, dreams of punishment are usually due to a double wish-
fulfillment: the fulfillment of a repressed wish and of the censor’s wish to
punish the dreamer for having such a wish. The first wish-fulfillment may
have occurred in an immediately preceding dream, thus leaving the punish-
ment wish on its own as motivated purely by the wishes of the censor. Such a
punishment dream would therefore be exceptional, in the context of Freud’s
theory, in not being driven by a wish from the system Ucs. It is, nevertheless,
still a wish-fulfillment (ibid.). 14
Thus far we have seen Freud’s responses to the objection presented by
unpleasant, anxiety, and punishment dreams. Some might consider these
responses ad hoc, but I think not, for they utilize theoretical resources already
inherent in Freud’s theorizing about dreams. In particular, they derive from
the two-agency theory that Freud had developed in order to explain other
facts. There is, however, a further class of dreams that may be thought to be
problematic for Freud. These are counter-wish dreams: dreams in which,
rather than a wish being fulfilled, a wish is frustrated.
Freud has two different ways of responding to counter-wish dreams. The
first is to treat these dreams in the same way as any other apparently non-
wish-fulfilling dream. He simply argues that the manifest content of the
dream, which includes the frustration of a wish, is deceptive, and that upon
analysis it is found that the dream is wish-fulfilling after all. The dream
frustrates a wish but at the same time fulfills a more important one. An
example of this is the supper-party dream that I discuss at length in the next
chapter. 15 The second way Freud deals with counter-wish dreams is to point
out two common kinds of wish that he thinks are often fulfilled in them. One
is a masochistic wish. The other is the wish to prove Freud wrong.
These latter attempts to squeeze counter-wish dreams into his theory
seem manifestly ad hoc, and have engendered much ridicule from Freud’s
critics. They may indeed turn out to be laughable, but one must be careful not
to let the prospect of schadenfreude obscure proper evaluation. What matters
here is context. For in certain contexts either of the explanations Freud offers
may be plausible. Where a patient has otherwise shown a propensity toward
masochism, the masochism explanation may not be so far-fetched. And if it
were the case that a counter-wish dream occurred in a context in which the
dreamer had recently encountered Freud’s theory of dreams and had shown
signs of motivated opposition to it, then the notorious “wish to prove me
wrong” explanation may also not be so far-fetched. 16 As with any purported
interpretation, a proper evaluation depends on the details of the particular
case. But in any case, as we have seen with his first response, Freud has the
44 Chapter 2
The purported counterexamples that we have seen thus far do not falsify
Freud’s theory. But there is a class of dreams that by Freud’s own admission
do. These are repetitive dreams of traumatic events. The most poignant ex-
amples are the dreams of soldiers suffering from “war neuroses,” abundant
cases of which emerged during the first world war. Freud considered these to
be a genuine exception to his wish-fulfillment claim:
This would seem to be the place, then, at which to admit for the first time an
exception to the proposition that dreams are fulfilments of wishes. . . . It is
impossible to classify as wish-fulfilments the dreams we have been discussing
which occur in traumatic neuroses, or the dreams during psycho-analyses
which bring to memory the psychical traumas of childhood. They arise, rather,
in obedience to the compulsion to repeat. (1920, S.E., XVIII: 32)
It was not unusual for Freud to change his mind in the face of countervailing
evidence—he did so on numerous occasions, most famously with his aban-
donment of the theory that neuroses are caused by childhood sexual abuse. In
the current case, he does not abandon the theory completely, but revises it in
light of the evidence. It is, though, a fundamental revision: it applies to his
very understanding of the nature of the unconscious.
Recall that Freud thought that the unconscious “has at its command no
other forces than wishful impulses” (1900, S.E., V: 568). In Beyond the
Pleasure Principle (1920), he comes to admit that there are other driving
forces at play after all. They relate to the phenomenon of repetition. Freud
observed repetition not only in recurrent dreams of traumatic events, but also
in the revival of traumatic childhood experiences by his patients, and in
children’s play. With respect to trauma, he hypothesized that repetition is in
part an attempt to generate the anxiety that, had it preceded the traumatic
event, would have greatly diminished its effect—to “master the stimulus
retrospectively” (1920, S.E., XVIII: 32). The attempt to master a traumatic
stimulus is to be distinguished from a wishful impulse, a technical distinction
having to do with Freud’s notion of psychical energy. Recall that according
to Freud’s view of the mind as a reflex apparatus, the mind is directed toward
keeping levels of energy constant by discharging any excess. The attempt to
deal with trauma is different from this in that it aims, not to discharge energy,
but to bring it under control. This is necessary in situations where there is a
great deal of uncontrolled energy—as happens after a traumatic event. In this
regard, traumatic repetition is an adaptive response. But Freud thinks that in
repeating an earlier experience, the mind is also manifesting a more funda-
mental psychic trend: the drive to divest the psyche of all energy and return
Wishes 45
to an earlier state of being—which Freud calls the “death instinct.” This drive
manifests in a “compulsion to repeat.” This is the ultimate motive force
behind all repetition phenomena, including recurrent traumatic dreams.
In sum, recurrent traumatic dreams neither fulfill wishes nor are moti-
vated by wishes, but are instead failed attempts to master the excessive
excitation generated by the trauma—attempts that draw upon the mind’s
primordial need to rid itself of all disturbance.
Later, however, Freud appears to backtrack on this revision by explaining
how, though the cases of traumatic dreams are indeed exceptions to the rule
that all dreams are wish-fulfillments, they are in keeping with the spirit of the
rule. As he explains:
With the traumatic neuroses things are different. In their case the dreams
regularly end in the generation of anxiety. We should not, I think, be afraid to
admit that here the function of dreams has failed. I will not invoke the saying
that the exception proves the rule: its wisdom seems to me most questionable.
But no doubt the exception does not overturn the rule. . . . We say that a dream
is a fulfilment of a wish; but if you want to take these latter objections into
account, you can say nevertheless that a dream is an attempt at the fulfilment
of a wish. No one who can properly appreciate the dynamics of the mind will
suppose that you have said anything different by this. In certain circumstances
a dream is only able to put its intention into effect very incompletely, or must
abandon it entirely. Unconscious fixation to a trauma seems to be foremost
among obstacles to the function of dreaming. While the dreamer is obliged to
dream, because the relaxation of repression at night allows the upward pres-
sure of the traumatic fixation to become active, there is a failure in the func-
tioning of his dream-work, which would like to transform the memory-traces
of the traumatic event into the fulfilment of a wish. In these circumstances it
will happen that one cannot sleep, that one gives up sleep from dread of the
failure of the function of dreaming. (1933, S.E., XXII: 29)
that this aim has failed in the case of traumatic dreams. This likely relates to
Freud’s view that the compulsion to repeat is rarely if ever pure; it is almost
always mixed with wishful impulses (1920, S.E., XVIII: 23). Children’s play
is an example of this: Freud thinks that games that repeat traumatic situations
are an admixture of the compulsion to repeat and wishes, such as the wish to
be grown-up and the wish for revenge (ibid., 16–17). It seems he assumes, in
his later rendering, that the compulsion to repeat is mixed with a wish, and
that it is the fulfillment of this wish that the dream aims at. Without this
assumption it would be difficult, on Freud’s understanding of the mind, to
explain why the compulsion took the path of a hallucination rather than
directly waking the sleeper. Of what this wish might be Freud gives no hints.
But if this is his supposition, then one can understand why he would speak of
traumatic dreams as failed attempts at wish-fulfillment: the accompanying
wish is that which the dream aimed at fulfilling, but was unsuccessful due to
something like a loss of nerve—as Freud puts it, the “dread of the failure of
the function of dreaming.”
These theoretical meanderings are driven by Freud’s psychological spec-
ulations. He sees traumatic dreams as a challenge to his understanding of
how the mind works. On the one hand, he views hallucination (of this kind)
as a product of the need to fulfill wishes. On the other hand, he views
traumatic dreams as driven by the non-wishful compulsion to repeat. In
Beyond the Pleasure Principle he toys with the idea of relinquishing the first
view, but in his later account he seems to have found a way of reconciling the
two. Though this may be a good solution for Freud, from an epistemological
perspective, neither account can be considered as more than merely plau-
sible. That this is so makes salient the degree to which Freud’s wish-fulfill-
ment claim is based on speculative theory. This is not to say that the claim is
without support, but it is to say that this support falls considerably short of
warranting the degree of confidence that Freud shows in it.
compatible and in some way connected. They are condensed into the dream-
content; the dream fulfills them simultaneously.
The explanation of over-determination lies in the phenomenon of rein-
forcement, as discussed in the rationale for D3. As we saw there, Freud
argued that preconscious thoughts need to find reinforcement from uncon-
scious wishes if they are to make their way into the dream-content. This
occurs most usually when the preconscious thoughts are unfulfilled wishes
that bear some relation to the unconscious wishes pressing for expression.
Similarly, greater reinforcement can be gained where several unconscious
wishes with some kind of affinity to each other combine. Over-determination
is, in this way, a manifestation of psychic opportunism.
Having established for himself the reality of over-determination, Freud is
able to make the following observation:
I have therefore been compelled to ask myself whether this characteristic may
not be a further essential precondition of dreaming. Stated in general terms,
this would imply that every dream was linked in its manifest content with
recent experiences and in its latent content with the most ancient experi-
ences. . . . It is still extremely hard to demonstrate the truth of this suspicion;
and I shall have to return in another connection (Chapter VII) to a considera-
tion of the probable part played by the earliest experiences of childhood in the
formation of dreams. (ibid.)
The reference here is to the general theorizing about the mind that we en-
countered in the rationale of D3. It concerns the earlier-mentioned thesis that
any preconscious thought found in the dream must have attained reinforce-
ment from an unconscious wish. Freud claims that his work on the neuroses
leads to the conclusion that unconscious wishes are ultimately of “infantile
origin.” By this he means a wish that first arose in the child at the age of five
or under and was at that stage repressed so as to become a permanent element
of the system Ucs.
Most controversially, Freud believes that the underlying infantile wish is,
in a broad sense, sexual. It is in The Interpretation of Dreams that Freud first
published an outline of what he would later call the Oedipus complex. 17 He
begins as follows:
48 Chapter 2
In my experience, which is already extensive, the chief part in the mental lives
of all children who later become psycho-neurotics is played by their parents.
Being in love with the one parent and hating the other are among the essential
constituents of the stock of psychical impulses which is formed at the time and
which is of such importance in determining the symptoms of the later neurosis.
(1900, S.E., IV: 260-61)
Table 2.2.
Rationale for D4
It will then appear as though the conscious wish alone had been realized in the
dream; only some small peculiarity in the dream’s configuration will serve as a
finger-post to put us on the track of the powerful ally from the unconscious.
(1900, S.E., V: 553)
This suggests that there is always some small aspect of the dream that has not
been fully accounted for by the initial interpretation, and this is what calls for
another, deeper interpretation. One might object that if there is an aspect of
the dream not accounted for, then that would imply that the initial interpreta-
tion was incorrect. But for Freud, the details of the case are often such that
the initial interpretation is clearly correct—as an explanation it fits too well
with the dream material to not be a true explanation. The point, though, if
accepted, brings another objection: what is needed is not another whole layer
of interpretation, but simply small modifications of the initial interpretation
to account for the peculiarity. But to settle only for the initial interpretation
would not do, since the subsequent, deeper interpretation is also clearly cor-
rect; it too fits too well with the dream material to not be a true explanation. 18
Since both the initial and the subsequent interpretation are, according to
Freud, clearly correct, something else must be invoked to explain the situa-
tion. This is where the notion of over-determination comes in. A full expla-
nation of the dream can be obtained if one admits that there are several
superimposed layers of meaning. We need abandon neither the well-fitting
initial interpretation nor the well-fitting deeper interpretation. By accepting
over-determination, on Freud’s view, we have the best explanation of the
dream-content.
Over-determination is, moreover, not a new notion, designed just for
dreams. It was something postulated by Freud from the beginning of psycho-
analysis to explain hysterical symptoms (1895, S.E., IV: 263). 19 It is also
smoothly consequent on Freud’s theoretical account of the formation of
dreams. As we saw in the rationale of D3, based on his model of the mind,
Freud has a reasonable argument for the idea that preconscious thoughts—
50 Chapter 2
In order to go beyond this bias, Freud considers his own dreams and finds
that these too have sources that go back to childhood. This fuels his suspicion
that the proposition is true in general. However, the evidence is still insuffi-
cient to draw the conclusion, as he admits:
It is still extremely hard to demonstrate the truth of this suspicion; and I shall
have to return in another connection (Chapter VII) to a consideration of the
probable part played by the earliest experiences of childhood in the formation
of dreams. (1900, S.E., IV: 218)
adopted by such a wish. I do not know yet to what extent I shall be able to stick
to this extreme theory, or let it loose in the dream book. (Bonaparte et al. 1954,
246–47)
The “extreme theory” did find its way into the book, but as we have seen, in
an appropriately provisional way.
The proposition that dreams are the fulfillments of infantile wishes is
therefore, on Freud’s view, not adequately supported by actual dream inter-
pretations. It derives rather from his general theorizing about the mind. It is a
consequence of two premises. One is the theoretical conclusion that the
primary motive for dream expression in adults comes from the system Ucs,
which we saw the argument for earlier. The other is that, as his work on
neuroses ostensibly showed, the wishes permanently lodged in this system
are infantile ones (1900, S.E., V: 553). This latter premise is in accord with
the developmental story, and the theory of mind it gives rise to, that I out-
lined earlier. 20 Here though we must draw the line, as a deeper investigation
into the epistemological bases of these ideas would take us too far afield.
In the previous chapter, we saw the process by which the latent content of a
dream is transformed into the manifest content through the operations of the
dream-work. Having examined Freud’s rationale for his claims D3 and D4,
we are now in a position to elaborate on this and see Freud’s full account of
the formation of dreams.
The latent content of the dream derives from both preconscious and un-
conscious sources. The preconscious sources are either “day residues” or
somatic stimuli. Day residues are thoughts from the day before that have not
been fully de-energized (1900, S.E., V: 554–55). Somatic stimuli are either
externally induced sensory impressions, such as those produced by heat,
light, and sound during the night, or stimuli produced by internal bodily
processes (1900, S.E., IV: 220-40). These sources stir up additional precon-
scious thoughts, and with them constitute the preconscious dream-thoughts.
They do not, however, have enough energy on their own to enter conscious-
ness. For this, they need to recruit a wish. 21
Freud identifies several sorts of wish that can serve as a dream-wish. One
sort is preconscious wishes from the day before that have been left unful-
filled. A second sort is preconscious wishes from the day before that have
been repudiated. A third sort is current wishes arising from bodily needs,
such as those relating to hunger, thirst, or sex. A fourth is unconscious
wishes. Wishes of the first two sorts listed above cannot, on Freud’s account,
drive the dream process on their own. They need reinforcement by an uncon-
scious wish. Whether wishes arising from bodily needs can drive a dream on
52 Chapter 2
their own is not entirely clear from Freud’s writing, but considerations of
dreams of convenience suggest they can. Ultimately, though, all non-
straightforward dreams require the energy of an unconscious wish in order to
become conscious. This unconscious wish, as we have seen, is a repressed
infantile wish.
The latent content of a (non-straightforward) dream thus consists of an
unconscious wish—plus various childhood memories associated with it—
and preconscious dream-thoughts. Where the preconscious dream-thoughts
contain a wish, this may, upon reinforcement by the unconscious wish, be-
come the preconscious dream-wish; where they do not contain a wish, they
combine with the unconscious wish to create anew a preconscious dream-
wish (1917, S.E., XIV: 226). There are therefore usually at least two mean-
ings of the dream: a preconscious meaning, consisting of the preconscious
dream-wish, and an unconscious meaning, consisting of the unconscious
dream-wish.
The latent content of the dream has enough energy to drive itself into
consciousness. It is, however, blocked from doing so by the censor, on ac-
count of the repressed material it contains, and by the wish to sleep. It is
therefore forced to undergo a process of regression. 22 On Freud’s model, the
normal path that psychical energy takes is from the perceptual part of the
mind to the motor end, via a series of memory systems of increasing degrees
of organization. We may think of this as the activation of increasingly more
abstract representations, culminating in propositional thought, from whence
action can ensue. Because the censor and the wish to sleep forbid the latent
content from becoming conscious thought, it is forced back toward the per-
ceptual part of the mind. This path takes the latent dream-thoughts through
the portion of the mind governed by primary process thinking. They are thus
subjected to condensation and displacement, as well as the inevitable trans-
formation into more concrete representations that the regression entails. It is
in this way that the latent content is transformed into the manifest content.
Entry into consciousness via perceptual representation is not, however, a
given. The censor patrols all entry points to consciousness, so is present here
also. Only thoughts that have been sufficiently distorted by the dream-work
to disguise their repressed elements (or association with such) will be able to
become conscious. If not, then entry is denied, and they will have to undergo
further distortion by the dream-work before they are ready for admittance. 23
In this way, censorship exercises a continuous pressure on the dream-work to
effect a suitable disguise. As a result, dream disguise is more or less assured.
Thoughts that make it this far therefore represent a compromise between the
needs of the dream-wish (to discharge energy) and the needs of the censor (to
prevent the anxiety that would ensue from the discharge of unconscious
energy). Finally, these thoughts undergo a further operation before, and
Wishes 53
while, becoming conscious: the secondary revision that weaves the thoughts
into an intelligible narrative.
This, then, is Freud’s account of how dreams are formed. The account is not
without problems. 24 It is not, however, my aim to resolve such problems
here. The epistemologically important issues are the ones I have already
discussed in the rationales of the major claims. My current purpose is to
present the wider picture, even if it is not without gaps. As Freud himself
would admit, his theory of dream formation is, in part, highly speculative,
but it nevertheless brings together his theoretical concerns and interpretative
findings into a coherent vision of the dream process.
One feature of the theory that should be emphasized is that all aspects of
dream formation are underpinned by a mechanistic account of how the mind
works. This pertains, in the main, to Freud’s notion of psychical energy. It is
the economics and dynamics of energy that underscore the movements de-
scribed above. Preconscious thoughts cannot enter consciousness because
sleep deprives them of the requisite energy; unconscious thoughts are neces-
sary because they have this energy; the subsequent transformations of these
thoughts, in particular their displacement and condensation, occur due to the
transfer of energy between ideas; the whole process of dream formation is
driven by the requirement to keep the level of energy constant. Thus to
understand more fully why these processes occur, we need to delve further
into the dynamics of psychical energy.
As with the claim that all dreams are wish-fulfillments, Freud uses the exam-
ple of children to motivate his claim about the function of dreams:
Let us observe a mother putting her child to sleep. The child gives vent to an
unceasing stream of desires; he wants one more kiss, he wants to go on play-
ing. His mother satisfies some of these desires, but uses her authority to post-
pone others of them to the next day. It is clear that any wishes or needs that
may arise have an inhibiting effect upon falling asleep. We all know the
amusing story told by Balduin Groller of the bad little boy who woke up in the
middle of the night and shouted across the night-nursery: “I want the rhino!” A
better behaved child, instead of shouting, would have dreamt that he was
playing with the rhino. Since a dream that shows a wish as fulfilled is believed
during sleep, it does away with the wish and makes sleep possible. (1901, S.E.,
V: 678)
In the example, that the good boy’s dream about the rhinoceros serves the
function of preserving sleep is brought out by comparison with the bad boy:
the one who dreams, sleeps, the one who does not, awakes.
54 Chapter 2
One can object that, though dreams often, perhaps even always, serve the
function of preserving sleep in children, the situation with adults is different.
Adults are adept at allaying gratification, so it seems unnecessary to have
dreams serve this function for them. Freud’s response to this depends on D3.
He points out that, in accordance with his rationale for that claim, adults have
unconscious wishes that are represented as fulfilled in dreams. In normal
waking life, these wishes are kept from having a disturbing influence by the
full power of the censor. But in sleep the censor is weakened. As such, these
wishes have the capacity to disturb sleep. On his theory, dreams serve the
function of preventing these unconscious wishes from disturbing sleep.
In sum, in the case of children’s dreams and adult dreams of convenience,
the dream manifestly serves the function of preserving sleep. In non-straight-
forward adult dreams, the dream also serves this function, this time protect-
ing sleep not from conscious and preconscious desires, but from unconscious
wishes.
This claim bears elaboration. As we have seen, according to Freud, much
of the working of our minds can be understood as having to do with the flow
and distribution of psychical energy, or excitation. An unconscious wish, on
this view, carries energy that needs discharging. Repression keeps it from
finding discharge during the day, but when this force is weakened at night it
is able to make more progress. Freud explains, “There are two possible
outcomes for any particular unconscious excitatory process. Either it may be
left to itself, in which case it eventually forces its way through at some point
and on this single occasion finds discharge for its excitation in movement; or
it may come under the influence of the preconscious, and its excitation,
instead of being discharged, may be bound by the preconscious” (1900, S.E.,
V: 578). The first outcome, of motor discharge, would rouse the sleeper.
Thus left to itself, unconscious energy would lead to waking, making pro-
longed sleep impossible. If ordinary sleep is to be possible, then the energy
must follow the second course outlined by Freud—it must be bound by the
preconscious. By “bound,” Freud means that the excitation is brought within
a system of stable energy. Freud says:
The cathexis [that is, investment of energy] from the Pcs. which goes half way
to meet the dream after it has become perceptual, having been directed on to it
by the excitation in consciousness, binds the dream’s unconscious excitation
and makes it powerless to act as a disturbance. (1900, S.E., V: 578)
Thus when the unconscious wish finds conscious expression in the dream, its
excitation is brought under the control of the ego rather than being allowed to
flow unrestrainedly and thence disturb sleep. Dreaming therefore has a clear
function: to provide an alternative course for unconscious excitation to that
which would disturb sleep.
Wishes 55
It was indeed to be expected that dreaming, even though it may originally have
been a process without a useful purpose, would have procured itself some
function in the interplay of mental forces. And we can now see what that
function is. Dreaming has taken on the task of bringing back under control of
the preconscious the excitation in the Ucs. which has been left free; in so
doing, it discharges the Ucs. excitation, serves as a safety-valve and at the
same time preserves the sleep of the preconscious in return for a small expen-
diture of waking activity. (ibid., 579)
Rationale for D5
Freud’s own arguments for the claim that the function of dreams is to pre-
serve sleep rely on his theory of mind. Nevertheless the claim can be sup-
ported without appealing directly to this theory. There are three steps to
doing so. The first is to establish that dreams result in the preservation of
sleep from disturbance—that is, but for the dream, the sleeper would have
woken. The second is to establish that dreams serve the purpose of preserv-
ing sleep from disturbance—that is, that dreams having this result is no
accident. The third is to establish that the function of dreams is to preserve
sleep from disturbance—that is, to serve this purpose is the reason why we
dream.
Children’s and straightforward adult dreams provide a template for this
argument. We have already seen how Freud motivated his claim by consider-
ing how children’s dreams can serve the function of preserving sleep. The
case is similarly clear with many straightforward adult dreams. These include
thirst, morning waking, and sex dreams. For example, if the disturbance
produced by thirst were to follow its normal course, the sleeper would have
to awake. Instead, in thirst dreams, the sleeper is able to represent the fulfill-
ment of his wish, with the result of prolonging sleep. Something similar can
be said for the other cases as well. These dreams manifestly result in prolong-
ing sleep.
56 Chapter 2
time that the wish to sleep is having some influence on the outcome. This is
true of morning waking dreams. For many people these are all too familiar.
One half awakes with the thought that one must soon rise and set about the
day’s duties, while also harboring the strong wish to return to one’s slum-
bers. The matter is all too conveniently resolved by a dream of awaking that
prolongs one’s sleep. The best explanation is that the wish to sleep was the
cause of the potential disturber—the wish to rise—taking the path toward
dream rather than action. This makes more plausible that this is so too in
other cases. The second point is that there appears little evidence of normal
adults being regularly awoken by unconscious thoughts outside the context
of a dream. Awaking from anxiety dreams is common enough, but what we
would expect, on Freud’s assumptions, if dreams were not serving the pur-
pose of preserving sleep from unconscious disturbance, is normal adults
frequently awaking with anxiety from a non-dream state. The best explana-
tion of this not being so is that the wish to sleep is causing unconscious
wishes to take the path toward dreaming rather than the path toward action.
Thus, as long as one buys the assumptions mentioned, it is reasonable to
conclude that the dream’s preservation of sleep is no accident, but instead is
the reason why there was a dream. This is the second step.
The conclusion thus far faces an objection. As touched on above, dreams
are often the causes of waking, rather than of the continuation of sleep. Such
anxiety dreams are all too common. It cannot therefore be true that it is the
purpose of all dreams to preserve sleep. Freud’s rebuttal is that this apparent
failure is due to an intricacy of dream function. As he explains:
This view is not traversed by the fact that there are marginal cases in which the
dream—as happens with anxiety dreams—can no longer perform its function
of preventing an interruption of sleep, but assumes instead the other function
of promptly bringing sleep to an end. In doing so it is merely behaving like a
conscientious night-watchman, who first carries out his duty by suppressing
disturbances so that the townsmen may not be woken up, but afterwards con-
tinues to do his duty by himself waking the townsmen up, if the causes of the
disturbance seem to him serious and of a kind that he cannot cope with alone.
(1901, S.E., V: 680)
This indicates an elaboration of the function of dreams that takes into ac-
count the censor. The censor’s function is to prevent unconscious wishes
coming to consciousness. At nighttime, though weakened by sleep, it is still
active. It has a dual role. First, as a preventative measure, it inspects all
expression-seeking thoughts and forbids those that are openly objectionable
from making progress toward the dream. Second, as a post hoc measure, it
monitors the conscious content of the dream in order to ensure that there is
nothing within it that would require the full force of repression. Normally,
due to its vigilance in the first role, the unconscious wish that is fulfilled in
58 Chapter 2
CONCLUSION
In this chapter and the last I have described Freud’s theory of dreams and
reconstructed his rationale for it. This exposition shows that Freud supports
his theory in two ways. One is by pointing to the evidence from dream
interpretations. The other is through his general theorizing about the mind.
Some commentators on Freud’s dream theory emphasize the one; others, the
other. But both are important.
In the rationale of the first two claims, that dreams are interpretable and
are the products of censorship and disguise (D1 and D2), dream interpreta-
tions predominate. For the wish-fulfillment claim (D3), dream interpretations
and general theorizing matter equally. In particular, the claim depends upon
Freud’s topographical model and his conception of the mind as a reflex
apparatus. In the last two claims, that dreams are the fulfillments of infantile
wishes and that the function of dreams is to preserve sleep (D4 and D5),
general theorizing predominates. D4 depends on D3 and also on Freud’s
conclusion, derived from his work on neuroses, that the Ucs consists pre-
dominantly of infantile wishes, and D5 depends on D3. I will in Chapters 3
and 4 have much to say about the validity of the interpretative evidence. But
before we get to that, a few comments are in order about Freud’s general
theorizing about the mind.
I have already observed that Freud himself considered his various models
of the mind as speculative, but that they are not pure speculation. They are
intended as broad explanatory frameworks whose value lies in their ability to
generate solutions to explanatory problems. To the extent that they generate
fitting solutions to these problems, they are supported by evidence. Freud’s
regarding his models as speculative is an acknowledgment that the evidential
support is not as strong as, say, that which he takes to underlie his dream
interpretations. On Freud’s view, the evidence supplied by free association
more or less compels the interpretations he gives. There is a tight fit between
the explanation (that is, the dream interpretation) and the data it explains (the
dream-content and the dream-thoughts uncovered by free association). Each
part of the interpretation is doing explanatory work, accounting for some of
the data. This is not so with Freud’s models of the mind. They nevertheless
gain credence by being able to unify disparate phenomena within a conceptu-
ally tidy construction. Freud treats these constructions as works in progress
(this being a utility of seeing them as speculative). Their worth lies in their
ability to generate explanations, so where they fail to do so satisfactorily,
they are to be elaborated on, revised, or, as a last resort, replaced by another
model. 25
Freud’s reliance on his general theorizing about the mind in arguing for
significant parts of his theory of dreams does not disqualify these parts from
serious consideration. It does, however, make their evaluation difficult. This
60 Chapter 2
The relation this argument bears to Freud’s own rationale should be clear
enough to anyone who has engaged with the reasoning recounted in the
previous sections. Steps 1 to 4 are based on arguments for the wish-fulfill-
ment claim (D3), steps 5 to 9 on arguments for the sleep-preserving claim
(D5), and step 10 on Freud’s theoretical account of censorship-disguise, by
which he explains dream-distortion (D2).
The conclusion falls short of the full version of Freud’s theory. It does so
in a couple of ways. One is that Freud’s core claims are expressed as univer-
sal, but the conclusion of the argument is not universal. In particular, it
allows that dreams can serve purposes other than the expression of impulses.
Another way it falls short of the full theory is that it does not specify the
nature of the unconscious impulses in question. As we have seen, for reasons
that had to do with both his psychological theorizing and his interpretative
findings, Freud thought that unconscious impulses seeking expression were
62 Chapter 2
ultimately infantile wishes. This is why he makes the specific claims D3 and
D4. But these details notwithstanding, one can see that the argument captures
the thrust of what Freud wishes to convey. As I observed in the introduction
to Chapter 1, a theory can be wrong in detail, but right in certain fundamental
respects, as was Copernicus’s. What is important is that we correctly grasp
the spirit. This is what the default argument brings out.
The argument is not compelling as it stands. Not only are its premises as
yet unsupported, but also its conclusion is not a strict consequence. Step 9
leaves room for an alternative way of dealing with unconscious impulses,
and the posit of censorship and disguise is offered merely as a possible
explanation of the fact that unconscious thoughts cannot be directly dis-
cerned in dreams. Thus even if one accepts its premises, the argument needs
supplementing. This can be done through dream interpretations. Such could
provide the needed support for the weaker steps in the argument by offering
evidence that unconscious impulses do make their way into dreams, and do
so, as expected, in a disguised way. The default argument and the evidence
from dream interpretations, that is, could mutually support each other in
justifying the theory. This brings our attention back to Freud’s method of
interpretation.
NOTES
1. Later Freud replaces this topographical model with the “structural” model of id, ego, and
superego. His main reason for this is that processes that he was attributing to the system Pcs are
in fact unconscious, thus belying their descriptive labels. Nevertheless, one can, roughly, treat
the dual system Cs-Pcs as corresponding to the ego, and the system Ucs as corresponding to the
id.
2. This is a fundamental claim, alluded to often in Freud’s writings (e.g., 1900, S.E., V:
554; 1924, S.E., XIX: 152–53; 1933, S.E., XXII: 18; 1940, S.E., XXIII: 183–84).
Wishes 63
3. Freud thinks they are in fact de-energized to some extent, but much less than precon-
scious thoughts (1917, S.E., XIV: 224–25).
4. This involves the unconscious thought transferring its energy to the preconscious
thought, whereby a link between the two thoughts is established. On account of this, the censor
will prevent the reinforced preconscious thought from entering consciousness (1900, S.E., V:
594).
5. In The Interpretation of Dreams Freud argues that even preconscious wishes are not on
their own enough (1900, S.E., V: 553). This, though, would presumably exclude wishes related
to basic drives such as hunger and thirst. Also, later Freud admitted that ego—in particular,
superego—wishes can drive dreams (see discussion of punishment dreams in next subsection).
6. But not all hallucinatory phenomena. The kind he refers to consists of dreams and
psychotic hallucinations. He points out that there are other kinds of hallucination that may
operate on entirely different principles (1917, S.E., XIV: 230).
7. The structural analyses of this and the preceding paragraph are based on Gardner (2006,
125–26).
8. On Freud’s view (1900, S.E., V: 544), the explanation for this and other kinds of
hallucination is that they are thoughts turned into images by regression—an hypothesis sup-
ported by hypnagogic hallucinations (cf. Chapter 1) and by Freud’s and Breuer’s clinical
experience (ibid., 545–46). The issue, then, is what kinds of thought would drive the extended-
narrative hallucinations of dreams.
9. See Kumar et al. (2009) for a concise review of general theories of hallucination.
Aleman and Larøi (2008) present a unified account that, in emphasizing the role of emotion and
motivation in psychiatric hallucinations, offers some encouragement for Freud’s explanation.
In Chapter 7, I discuss the activation-synthesis model of dreaming, which is the most prominent
competing explanation of dream hallucination.
10. This fairy tale comes in different versions. It was originally published in Charles Per-
rault’s Histoires ou contes du temps passé, avec des moralités: Contes de ma mère l’Oye
(Stories or Fairy Tales from Past Times, with Morals: Mother Goose Tales) of 1697.
11. This is not to say that the dreamer upon waking would recognize the wish-fulfillment,
for, first, not all ego-awareness corresponds to the person’s awareness (otherwise censorship
would be an explicit process), and second, upon waking, repression returns in full force. It is,
however, to say that the disguise of the wish-fulfillment in anxiety dreams is thin.
12. This quote and the ideas related to it are relatively late revisions of Freud’s thinking
about anxiety dreams. It comes from paragraphs added in 1919. See Chapter 5 for details about
the changes.
13. Freud’s account of punishment dreams is also from the 1919 additions.
14. This account of punishment dreams is an initial step toward a new development in
Freud’s thinking: the acknowledgment of a much larger role for the ego in the formation of
dreams. The account admits that the dream can be the fulfillment not just of an infantile sexual
wish, but also an ego-wish (cf. Weiss 1992).
15. In the example, the wish frustrated is one that arises only within the dream, but in other
dreams the wish frustrated may be one from waking life.
16. It should also be borne in mind that the “wish to prove me wrong” is preconscious, and
hence according to Freud’s theory conceals an unconscious wish that should emerge upon
deeper analysis.
17. The name “Oedipus complex” first occurs in Freud’s work in 1910 (S.E., XI: 171).
18. The dream material in this case would likely have been expanded by further associations
by the dreamer on the recalcitrant aspects of the dream.
19. Laplanche and Pontalis (2006, 292) summarize the idea well: “The hysterical symptom
is said to be over-determined in that it is the outcome . . . of a number of traumatic events: one
of these factors on its own is not enough to produce or sustain the symptom.”
20. According to Freud, that the Ucs exists at all is due to the relatively late development of
the secondary processes (1900, S.E., V: 603), meaning that unconscious wishes are inevitably
infantile.
21. Preconscious stimuli can be the overall instigators of a dream (1900, S.E., V: 560).
Freud compared the relationship between these and the unconscious wish that drives the dream
64 Chapter 2
to that between entrepreneur and capitalist (ibid., 561). But the overall instigator—the entre-
preneur—can also be an unconscious wish aroused by events from the day before.
22. Freud’s account of why regression occurs is not straightforward—he postulates other
factors besides censorship and sleep (1900, S.E., V: 547, 574).
23. One should not interpret the order of presentation up to this point strictly. As Freud puts
it, “I have been obliged to adopt this order in my description; but what happens in reality is no
doubt a simultaneous exploring of one path and another, a swinging of the excitation now this
way and now that, until at last it accumulates in the direction that is most opportune” (1900,
S.E., V: 576).
24. One problem has to do with the fact that dreams are not purely perceptual, but some-
times contain propositional thoughts (Macmillan 1997, 266–67). This presents a difficulty
because, on Freud’s account, dreams arrive into consciousness via the perceptual system. A
solution to this problem may be attainable through the following considerations: (i) Freud’s
point that there “is no doubt a simultaneous exploring of one path and another” (1900, S.E., V:
576); (ii) acceptance that some elements of the latent content may break through to conscious-
ness via propositional thought (cf. 1900, S.E., V: 535); (iii) allowing that the ego has the ability
to weave together related parts of the dream that have become conscious in different ways.
25. Freud had a modest view of the metaphysical status of these models. At times he
described them as “myths.” It is unlikely that he meant by this that they were mere fictions, for
that would not be in keeping with his regarding them as speculative (a mere fiction is not the
kind of thing that can be speculative). Rather, if one considers a myth as a more or less
metaphorical expression of some truth, one can see his models similarly as more or less
metaphorical expressions of truth—ones given in terms of concepts appropriate to the scientific
sensibilities of the day. One can therefore see how a “myth” can nevertheless be speculative,
for, in essential respects, it may or may not reflect the truth. It is a view compatible with a semi-
realist metaphysics.
26. This is, of course, the fundamental tenet of psychoanalysis. I use the term “impulse”
rather than “thought” to denote its highly directed nature, in line with Freud’s own terminologi-
cal preference (1933, S.E., XXII: 18).
Part II
Jigsaw Pieces
Wittgenstein was ambivalent about Freud. On the one hand, he took him
seriously—as he recounts, “I happened to read something by Freud, and I sat
up in surprise. Here was someone who had something to say” (1966, 50). On
the other hand, he regarded his style of thinking as something that needs
“combating.” This was especially so with regard to his method of dream
interpretation, which he argued against as follows:
The fact is that whenever you are preoccupied with something, with some
trouble or with some problem which is a big thing in your life—as sex is, for
instance—then no matter what you start from, the association will lead finally
and inevitably back to the same theme. Freud remarks on how, after the analy-
sis of it, the dream appears so very logical. And of course it does.
You could start with any of the objects on this table—which certainly are
not put there through your dream activity—and you could find that they all
could be connected in a pattern like that; and the pattern would be logical in
the same way.
One may be able to discover certain things about oneself by this sort of
free association, but it does not explain why the dream occurred. (ibid., 50–51)
The objection, in short, is that the fact that one has come up with a set of
thoughts that “make sense” of the dream and its associations is not good
evidence that those thoughts caused the dream. This point has been raised, in
various forms, by numerous critics, going back to the very inception of the
theory. 1
Freud said many things in defense of his method, but the most intriguing
of these is an analogy. Writing of the dream interpreter’s conviction, he
explains:
67
68 Chapter 3
What makes him certain in the end is precisely the complication of the prob-
lem before him, which is like the solution of a jig-saw puzzle. A coloured
picture, pasted upon a thin sheet of wood and fitting exactly into a wooden
frame, is cut into a large number of pieces of the most irregular and crooked
shapes. If one succeeds in arranging the confused heap of fragments, each of
which bears upon it an unintelligible piece of drawing, so that the picture
acquires a meaning, so that there is no gap anywhere in the design and so that
the whole fits into the frame—if all these conditions are fulfilled, then one
knows that one has solved the puzzle and there is no alternative solution.
(1923, S.E., XIX: 116)
The case against Freud’s method is most thoroughly developed by the philos-
opher Clark Glymour. His main objection has been endorsed by a number of
critics, most of whom take it to be decisive. The tone Glymour adopts indi-
cates that he himself does not think the problems are resolvable: “The meth-
od itself is worthless . . . [and] the objections to it are obvious ones. . . . It is
hard to believe that they did not occur to Freud himself. The whole business
seems the cheapest of rhetorical tricks” (Glymour 1983, 65).
There are three problems. First, the method that Freud describes does not
correspond with the method he practices. Second, there is an obvious objec-
tion to his method to which Freud fails to respond. Third, Freud’s inferences
are in fact fallacious.
As we saw in Chapter 1, Freud’s method of dream interpretation involves
free associating with elements of one’s dream. This means reporting whatev-
er comes to mind in relation to a particular element, on the condition that one
does not critically appraise or hold back any thoughts. According to Freud,
Jigsaw Pieces 69
“If [the analysand] succeeds in doing that, innumerable ideas come into his
consciousness of which he could otherwise never have got hold. The material
which is in this way freshly obtained for his self-perception makes it possible
to interpret . . . his dream-structures” (1900, S.E., IV: 102). Glymour objects
that this description does not correspond to how Freud’s actual dream inter-
pretations proceed. Particularly in his self-analyses, the process is more com-
plicated, involving Freud free associating not only with elements of the
dream, but also with conclusions he draws about the meanings of dream
elements, questions he raises about such conclusions, answers he gives to
such questions, and so on (Glymour 1983, 63). Glymour adds that the way
the associations generated by these processes are selected and sorted seems
even more complicated, and nowhere does Freud help his reader understand
how this works.
Glymour is right that in practice the process of interpretation is more
complicated than Freud’s brief description suggests. It is, however, unclear
how much of a problem this is. Freud himself elsewhere describes several
variations on his method, including simply asking the patient to recount
events from the previous day or putting before him some plausible symbol
interpretations (1923, S.E., XIX: 109–10). Also, as we saw in the quote
above, Freud claimed only that associations “make possible” an interpreta-
tion, indicating that there may be much work still to do before an interpreta-
tion can be teased out. Thus Freud may well have been prepared to accept
Glymour’s above charges. There is nothing, though, to indicate that these
advert to anything other than an expository failing on Freud’s part.
Glymour, however, thinks there are deeper problems. Freud fails to re-
spond to the following “obvious” objection:
This is similar to Wittgenstein’s objection in that both are to the effect that,
using free associations, one can end up with a plausible-looking “interpreta-
tion” that nevertheless has nothing to do with how the dream-content actually
arose. It presents a serious challenge to the validity of Freud’s method. It is
70 Chapter 3
clear, though, that Glymour does not regard it as just a challenge to which
Freud needs to respond, but as one to which he cannot respond. As he goes
on to remark about Freud’s alleged interpretations of counter-wish dreams:
If his patients were convinced of his thesis by the interpretations Freud offers
of their recalcitrant dreams, then they cannot have been very clever people.
For by Freud’s method every dream could be made out to be an expression of a
wish just as every dream could, with almost equal ease, be made an expression
of disgust or regret or fear or. . . . (66; his ellipsis)
Freud’s failings are thus, on his view, not merely expository; they are ele-
mental.
Glymour takes the point a step further by offering an account of wherein
the fallacy lies. He does this in the context of an example of one of Freud’s
interpretations. The dream in question, by a woman patient, is as follows:
I wanted to give a supper-party, but I had nothing in the house but a little
smoked salmon. I thought I would go out and buy something, but remembered
then that it was Sunday afternoon and all the shops would be shut. Next I tried
to ring up some caterers, but the telephone was out of order. So I had to
abandon my wish to give a supper-party. (1900, S.E., IV: 147)
The woman was then asked to produce associations based on this dream, and
eventually she came up with the following:
She went on to tell me that the day before she had visited a woman friend of
whom she confessed she felt jealous because her (my patient’s) husband was
constantly singing her praises. Fortunately this friend of hers is very skinny
and thin and her husband admires a plumper figure. I asked her what she had
talked about to her thin friend. Naturally, she replied, of that lady’s wish to
grow a little stouter. Her friend had enquired, too: “When are you going to ask
us to another meal? You always feed one so well.”
The meaning of the dream was now clear, and I was able to say to my
patient: “It is just as though when she made this suggestion you said to your-
self: ‘A likely thing! I’m to ask you to come and eat in my house so that you
may get stout and attract my husband still more! I’d rather never give another
supper-party.’ . . . The fact that what people eat at parties makes them stout
had been brought home to you by your husband’s decision not to accept any
more invitations to supper in the interests of his plan to reduce his weight.” All
that was now lacking was some coincidence to confirm the solution. The
smoked salmon in the dream had not yet been accounted for. “How,” I asked,
“did you arrive at the salmon that came into your dream?” “Oh,” she replied,
“smoked salmon is my friend’s favourite dish.” I happen to be acquainted with
the lady in question myself, and I can confirm the fact that she grudges herself
salmon no less than my patient grudges herself caviar. (1900, S.E., IV: 148)
Jigsaw Pieces 71
Glymour takes Freud’s report to imply that the association the woman
makes with the dream, the thought of her skinny friend, was caused by two
elements of the dream: the supper-party and the smoked salmon. That is, the
causal links are as follows:
He thinks that Freud then “transposes the causal relation” in concluding that
the thought of the skinny friend caused the supper-party and smoked salmon
elements of the dream. That is, he switches the causal links to the following:
“What,” Glymour asks, “are the grounds of the transposition? Evidence for
the first causal model is not necessarily evidence for the second” (1983, 68).
Glymour goes on to argue that the first causal model is indeed not evi-
dence for the second:
If we think simply in terms of the second causal picture, the fact that the dream
both contains a failed attempt at a supper-party and mentions smoked salmon,
and both of these elements lead to the remembrance of features of the patient’s
friend, seems an amazing coincidence that demands explanation. The best
explanation seems to be that these elements of the dream have a common
cause, and that cause has to do with a thought about the friend in question. But
if we stand back for a moment we see that this coincidence is manufactured:
one associates, at Freud’s direction, until one thinks of something which has
connection with several elements in one’s dream; the several elements cause
the common thought, not vice versa, and the coincidence requires no further
explanation. (ibid., 68)
I will refer to this . . . as the “free association fallacy.” . . . The fallacy occurs,
for instance, when it is assumed that something—a thought, a feeling, a mo-
tive—which turns up during free association to a dream, merely because it
turns up, must be a background thought of the dream. In other words . . . if I
associate from A to B, then B must have been a determinant of A. (Sand 1993,
531)
EPISTEMOLOGICAL INTERLUDE
I think the critics are wrong. Freud’s method is not based on a fallacy. But
though I think the case for the method is stronger than the critics claim, it
nevertheless falls short of a justification. This is because the kind of evidence
needed to justify the conclusion reached in any particular dream interpreta-
tion is not unambiguously available to those outside the analytical setting.
This means that a defense on the basis of dream interpretations alone cannot
take the form of a justification, but of something less, namely, a validation.
Let me explain the distinction I have in mind. To justify a theory or claim
is to show that it is right to believe that theory or claim. One can do this by
providing a valid argument based on premises that our audience would be
justified in believing. 3 There are two parts to this. First, the argument needs
to be valid. An argument can be valid in two ways: it can be deductively
valid, that is, have premises that logically entail the conclusion; or it can be
inductively valid, that is, have premises that provide strong inductive support
Jigsaw Pieces 73
for the conclusion. The second part is that the premises need to be such as the
audience would be justified in believing.
The problem with Freud’s interpretative claims is that, often, the best
argument for them is based on premises that others may not be justified in
believing. In which case such claims cannot be justified in the above sense.
They can, however, be defended in a more limited way. This way is that to
which the first part of our equation applies: there is a valid argument for
them.
Let me expand on this. A method of inference is a means by which one
can draw a conclusion (the output) from some premises (the input). I define a
method of inference as methodologically valid just in case, for any particular
application of it, if the input is of the right kind then the output will be of the
right kind. The model for this is deductive validity. An argument is deduc-
tively valid just in case, if the premises are true then the conclusion must be
true (this is the standard definition used in logic textbooks). The notion is
defined in such a way that its application is independent of the truth of the
premises. That is, whether an argument is deductively valid or not does not
depend on whether the premises are true or not. Validity tells us something
positive about an argument, but is insufficient to establish that it is a sound
argument. It is nevertheless a useful notion, much used in logic, since estab-
lishing the validity of an argument is an important step toward establishing
its soundness. Similarly with inductive validity. An argument is inductively
valid just in case if the premises are true then the conclusion is very probably
true. 4 An argument’s being inductively valid is insufficient for it to be a
cogent argument, but nevertheless the notion is useful because establishing
the inductive validity of an argument is an important step toward establishing
its cogency.
In my definition of methodological validity, I am extending this model in
one important respect: it need not simply be that the input consists of true
propositions that matters for methodological validity, but that it (perhaps) has
additional properties. Consider, for example, the kind of inductive reasoning
called enumerative-inductive generalization, whereby one infers from a sam-
ple in which each member has the property in question to the conclusion that
each member of the target population has the property in question. Such a
generalization is cogent as long as the data on which it is based are not only
veridical but also constitute a large enough and representative enough sam-
ple. We can say of a particular such generalization that is based on a large
enough and representative enough sample that it is inductively valid, mean-
ing, if the data are veridical then the conclusion is probably true. But we can
say more generally that the method of enumerative-inductive generalization
is valid, meaning, for any particular application of it, if the data are veridical
and constitute a large enough and representative enough sample then the
conclusion is probably true. The cogency of any application of a valid meth-
74 Chapter 3
heuristic inference and justificatory inference are not the same, it is specifi-
cally justificatory inference I am concerned about.
Armed with this distinction, let us first examine Sand’s criticism. The
accusation is that Freud merely assumes that items associated with elements
of the dream are causes of the dream. That this is not an accurate description
of what Freud is doing is suggested by Glymour’s observation that Freud is
selecting and sorting among the associations made. Something more seems
to be going on than the mere assumption that an item associated with the
dream-content was a cause of the dream-content. 5 The more pertinent point,
however, is that even if the accusation were true, it need only be so of his
method of heuristic inference, not of his method of justificatory inference.
At least in the case of his specimen dream interpretations, it is not true
that Freud considers his positing of associations as causes as warranted mere-
ly by the fact of their having been freely associated with elements of that
content. This much I think should be held on the grounds of the principle of
charity, as explained above. But Freud also provides us with plenty of clues
as to what his method of justificatory inference is. Such are contained, for
example, in his jigsaw puzzle analogy, in which he explains that confidence
in an interpretation has to do with something analogous to the “absence of
gaps” and the “fit into the frame” of a completed puzzle. Such are also
contained in his specimen dream analyses. Consider, for example, what
Freud says in rejecting a potential interpretation of one of his dreams:
Anyone who interprets this dream without regard for my rules will conclude
that I was worried about my friend’s health and that this worry was realized in
the dream. . . . But I should be glad if anyone interpreting the dream in this
way would be good enough to explain to me why my fears on Otto’s behalf
should have lighted on Basedow’s disease—a diagnosis for which his actual
appearance gives not the slightest ground. (1900, S.E., IV: 269)
The real coincidence is that on the one hand, on one day the dreamer was
visited by a friend, whose favourite dish is smoked salmon, and who had asked
her when she, the dreamer, planned to have another supper-party, and, on the
other hand, that night she dreamed of a supper-party and smoked salmon. Is
this coincidence evidence of a causal connection between the encounter with
her skinny friend in the dream? Perhaps it is, in the case at hand, for the real
event and the dream are proximate in time and share a number of independent
features, and one doubts that any event so proximate would share those fea-
tures. (Glymour 1983, 69)
Thus Glymour concedes that when the nature of the connections is taken into
account, the Freudian explanation may be superior to his explanation. But if
it is conceded that it is likely that the thought of the skinny friend caused the
dream-content, then there is no reason to maintain that the coincidence that
the thought of the skinny friend connects two distinct elements of the dream
is manufactured. Hence Glymour seems to be undermining both his analysis
of how the interpretation proceeded and his argument against Freud.
Glymour’s point, however, now changes. The problem for Freud is that
the case was meant to be illustrative of his method, yet the features which
indicate the success of his interpretation do not seem generalizable. Glymour
takes this as showing that Freud has failed in his purpose of demonstrating
the general trustworthiness of his method.
This is an odd point to make given the context in which the case is
presented. Freud’s purpose in giving his interpretation of the supper-party
dream was not to demonstrate the general trustworthiness of his method but
to address the objection from counter-wish dreams. He presents the case as
an example of a dream in which the content seems the frustration of a wish
and yet turns out to be wish-fulfilling after all. If, as Glymour seems prepared
to concede, the interpretation of the dream is correct, then Freud has fulfilled
his purpose.
What, nevertheless, does this case show about the general validity of
Freud’s method? Though I agree that Freud has not demonstrated that his
method of interpretation will always yield successful interpretations—which,
as we have seen, he was not trying to do—if this interpretation is successful,
he has nevertheless shown that the method of interpretation can yield suc-
cessful interpretations. More importantly, the case is—contra-Glymour—ge-
neralizable. It indicates a method of justificatory inference. One can infer that
a particular interpretation is correct on the basis that it explains both the
dream-content and the existence of a coherent set of concerns of the agent
that happen to have several fitting connections with the dream-content—
something the expectation of which would have been low prior to the process
80 Chapter 3
of interpretation. It is, in other words, the quality of the connections that can
clinch the case: where the association fits the dream well, Freud’s explana-
tion becomes more plausible than Glymour’s. In the case of the supper-party
dream, the facts that Glymour alludes to, the proximity of events and the
shared independent features, are part of what renders the connections in
question as particularly fitting. In other cases what renders the connections as
fitting may be different, but as long as there is something that does so, the
proposed interpretation can be deemed successful on the basis of the same
pattern of inference.
Let me elaborate on this idea. There are two problems we need to address: (1)
how to establish that an association is a dream-thought; (2) how to establish
that the overall interpretation of the dream is correct.
The first problem can be addressed through two considerations. The first
consideration is the fittingness of an association’s connections with the
dream-content. The second is the number of fitting connections it has with
the dream-content—clearly, the more such connections, the more probable it
is that the association is indeed a cause. Both fittingness and number of
connections are of course a matter of degree, and that degree is important to
the validity of the inference. But the key observation is this: There is surely a
point where, when the fittingness of the connections is high enough and/or
there are enough such connections, it is reasonable to judge that the expecta-
tion there would be a thought with such connections to the dream-content is
low—that is, that it is improbable that it is just a coincidence there is a
thought with such connections. People may differ as to where they think that
point would occur, but this is a difference of judgment, not of principle.
Assuming the judgment to be well-placed, the inference that the element thus
connected was a cause of the dream-content is reasonable. The method of
justificatory inference is therefore valid, even if any specific application of it
is not.
The second problem is with establishing that the overall interpretation of
the dream is correct. Uncovering dream-thoughts would not necessarily un-
cover a coherent meaning for the dream, for these thoughts may be just as
arbitrary as the dream-content appears to be. So we need an account of how,
in general, Freud arrives at the meaning of the dream.
What Freud proceeds to do in his interpretations is to uncover a coherent
set of concerns of the dreamer that are related to all, or almost all, the
elements of the dream via the dream-thoughts thus far uncovered—that is,
via the associations that have been established as causes of the dream. The
concerns are, more specifically, what render the dream-thoughts emotionally
salient. The method of justificatory inference here is not essentially different
from that which we saw before. The set of concerns may be posited as a
cause of the dream-content in that this posit provides a unified explanation of
Jigsaw Pieces 81
the following: (1) the content of the dream in its entirety, and (2) the fact
there happens to be a coherent set of concerns that fittingly relates to all, or
almost all, the elements of the dream. The best explanation of why there
should be a set of concerns, coherent among themselves, that are connected
to the dream-content in the way described is that these concerns are the
meaning of the dream.
An alternative to this explanation compatible with the thesis that dreams
do not have meaning is that the dream-content was generated arbitrarily and
it is just a coincidence that there exist concerns that happen to so fittingly
relate to the dream-content. The question is: Which is the better of these two
explanations? The latter explanation cannot be more probable than the prior
probability of the purported coincidence. Estimating this prior probability
will, of course, depend on subtle judgments about the connections. But the
point is this: Depending on how such judgments go, one may be justified in
concluding that the first explanation is significantly better than the second. In
such a case, one will be justified in drawing the conclusion that the concerns
identified caused the dream-content—that is, they are the meaning of the
dream.
This provides an answer to Glymour’s earlier criticism, that from one’s
associations with random arrangements—such as ink blots or rock forma-
tions—one can foist just about any meaning one wants onto these arrange-
ments. The objection makes out Freud’s procedure to be gratuitous, whereas
on the above account it is not. Firstly, there is, as we have seen, a principled
method by which associations may be judged causes of the dream-content—
namely, by that being the best explanation of why there should be emotional-
ly salient thoughts with such fitting connections to the dream. Second, there
is a principled method by which a meaning may be derived from these
associations—namely, by being those concerns that render the associations
in question emotionally salient. The supper-party dream exemplifies this, for,
as we have seen, the conversation with the skinny friend to which the patient
associated was emotionally salient precisely because of the patient’s jealousy
about her husband’s high opinion of that friend and her consequent wish that
the friend should not grow stouter so as not to attract her husband more. The
interpretation that Freud gives of this dream is hence far more plausible than
one would expect if Freud’s procedure were as gratuitous as Glymour makes
out. There are therefore subtle differences between gratuitously assigning a
plausible meaning to a set of associations and what Freud is doing, but ones
that nevertheless make all the difference with regard to the validity of his
method.
This also provides an answer to Wittgenstein’s objection that associating
with dream elements would inevitably lead back to a unified set of concerns
even if these were not the source of the dream. For this does not take into
account the fittingness of the connections. As I have argued, if the connec-
82 Chapter 3
tions are fitting enough, then the prior probability that there would exist a set
of concerns with such connections to the dream-content would be relatively
low. Much depends on just how fitting the connections are, but it is surely
possible that the degree is high enough to make it an unlikely coincidence
that the concerns the dreamer is “preoccupied with” happen to have such
connections with the dream-content. In such cases, the better explanation is
that those concerns were causes of the dream-content, as is always plausible
(since we know that concerns that one is preoccupied with can be, and often
are, causes of dream-content).
cable to dreams. The principles have to do with what Blass calls “networks of
meaning”—integrated sets of causally related psychical elements within an
agent with respect to which we can interpret his or her intentions. The as-
sumptions that underlie psychoanalysis, she argues, are reasonable with re-
gard to those networks operative when we are awake, for then we have
introspective and intuitive evidence for them. But the same evidence does not
apply to dreams. It follows that we cannot know that there are networks of
meaning for dreams in the same way as we can know this for wakeful states,
and even if there were, we cannot know that they are continuous with those
that are operative at the time of analysis in the same way as we do for
wakeful states. Unless there is any other evidence available, we would have
no reason to apply the principles of psychoanalysis to dreams, hence Freud’s
claim would remain unjustified even for those who accept the general princi-
ples of psychoanalysis.
Blass, however, goes on to argue that there is alternative evidence avail-
able. It comes from a particular quality of experience, that of “a feeling of
connectedness of psychic entities or ideas” (2002, 171). Blass calls this “the
experiential quality of meaningfulness” (henceforth EQM). She emphasizes
that this quality does not derive from a judgment, implying that an agent can
experience this quality without being committed to the truth of any proposi-
tion. The quality has three aspects. The first is “immediacy.” This means that
it falls upon the agent—that is, is not mediated by reflection or other similar
processes of thought. The second is “contentlessness.” This means that it is
not about anything in the way that, for example, the states of belief and
desire are about something, but rather is simply a particular quality of experi-
ence. Thus what we are concerned with is not, strictly speaking, the experi-
ence of meaningfulness, but the way in which something is experienced. The
third aspect is “judgementalness.” Though the quality does not derive from a
judgment, it nevertheless bears some similarity to the experience of making a
judgment. Blass compares it in this respect to the experience of déjà vu—the
“judgmental” feeling that one’s current experience has occurred before. As
such, EQM differs from the “immediate and contentless” experience of anx-
iety, which does not have the “judgmental” aspect to it.
There are three kinds of experience that can have this quality. The first is
the experience of a meaningful connection between two known psychical
entities, A and B. Formally, an experience of this kind which has the quality
is represented as EM[A M B], standing for the MEANINGFUL EXPERI-
ENCE THAT [A is MEANINGFULLY CONNECTED TO B]. The second is
the experience that “psychic entity A has some meaningful connection to
something,” though the agent does not know to what. Formally, an experi-
ence of this kind which has the quality is represented as EM[A M x], where x
is a variable standing for some unknown entity. The third is the experience
that “something meaningful has happened or been stated,” but the agent
84 Chapter 3
knows not what. Formally, an experience of this kind which has the quality is
represented as EM[x M y], where x and y are both variables standing for an
unknown entity (Blass 2002, 170–71). 8
Blass argues that the “experiential quality of meaningfulness” provides
evidence that the general principles of psychoanalysis are applicable to
dreams as follows. She first argues that the fact that a person experiences a
connection meaningfully is good reason for believing that there is a meaning-
ful connection. This is so because (a) it would be difficult to explain why
someone would have an illusion of meaningfulness, and (b) clinical experi-
ence indicates that “contents that are felt to be connected for reasons un-
known to the patient emerge as significantly tied to each other” (2002, 182).
She then observes that it is common for people to experience EQM in rela-
tion to their dreams—that is, to have an unmediated feeling that the dream is
meaningfully connected to something. Since EQM is a reliable indicator of
meaning, we may conclude that, in such instances, the dream does have
meaning. Moreover, Blass argues, just as one can experience only one’s own
pain, so one can experience only one’s own meanings—that is, those that
relate to one’s currently operating network of meaning. It follows that the
network of meaning of a dream must be the same as the network of meaning
of an awake individual at the time of this experience, thereby justifying the
application of the principles of psychoanalysis to dreams. Thus we have a
conditional justification of Freud’s claim.
There are three manifestations of EQM that can serve as such evidence.
The first is the one that Blass refers to in her argument: that people often
have the immediate experience that their dream, or some part of it, is mean-
ingful. This is the manifestation of EQM formally denoted as EM[A M x],
where A here is either the dream as a whole or some element within it. This
also seems to be the phenomenon that Wittgenstein observes when he re-
marks, “It is characteristic of dreams that often they seem to the dreamer to
call for an interpretation” (1966, 45; my italics). Since the analyst provides a
meaning for the dream through her interpretation, she also thereby provides a
particularly apposite explanation of why this characteristic should be present:
we have the sense that our dream, or some part of it, is meaningfully con-
nected to something because it is meaningfully connected to something,
namely, the dream meaning. It is, however, difficult to see how any explana-
tion of the dream-content that holds it to be meaningless could explain this
phenomenon.
The second manifestation is that the analysand often experiences an ele-
ment that emerges during free association as meaningfully connected to the
dream-content. This is the experience of EQM denoted by EM[A M B],
where A is an element of the dream-content and B is an association. When
this experience is present it either constitutes or contributes to the fittingness
of a connection. Any alternative explanation to that in which B is posited as
the cause of A is, in line with my argument of the previous section, limited in
likelihood by the probability that it is just a coincidence that there happens to
be an element with such a fitting connection to the dream-content. It is also
limited by the need to explain why the analysand’s experience of the connec-
tion would have that particular quality. To the extent that the experience of
EQM is relatively rare, the probability of such an alternative explanation is
low.
The third manifestation is that, when provided with an interpretation of
her dream, the analysand often comes to experience the concerns thus uncov-
ered as meaningfully connected to the dream-content as a whole. That is, the
interpretation strikes her as being the meaning of the dream. This is the
experience of EQM denoted by EM[A M B], where A is the dream as a whole
and B is the meaning of it given by the interpretation. Freud provides hints
that this is the case with at least some of his dream interpretations. For
example:
I was able to give her the correct interpretation of the dream, which she
afterwards confirmed. (1900, S.E., IV: 152)
I differ from Glymour in taking these, not as evidence that the patients in
question were not clever, but rather as evidence that the patients experienced
an immediate conviction that many people who have had their dreams inter-
preted (either by themselves or others) have also experienced and which
seems independent of any case that has been made for that particular conclu-
sion. If we suppose that a dream interpretation is accompanied in this way by
the distinctive experience of meaningfulness, then we have another datum
that is explained by the correctness of the interpretation. Here, though, there
is an obvious alternative explanation—namely, that which Glymour offers,
that the analysand is convinced by the analyst that her interpretation is indeed
the meaning of the dream. This explanation, however, does not do justice to
the phenomenology of the experience. As remarked, the experience has an
immediacy that is absent from that of becoming convinced by the case made
for it by the analyst. The explanation offered by the correctness of the inter-
pretation, on the other hand, does do justice to the phenomenology. So once
again, when present, this is evidence in favor of the analyst’s interpretation.
Overall, the experience of EQM in any one or more of these three mani-
festations considerably strengthens the argument that the analyst’s interpreta-
tion is correct because its being so is the best explanation of the facts under
view. Of course, we are dealing here with evidence not available to anyone
outside the analytical setting. For this reason I think we can only speak of this
evidence as pertinent to the validity of Freud’s method of justificatory infer-
ence, rather than to a general justification.
NOTES
Interconnections
You know a jigsaw puzzle is complete when you have used all the pieces,
there are no gaps, the picture has meaning, and everything falls within a
frame. But it matters also that the pieces fit together snugly. In a jigsaw
puzzle, each piece has tabs and slots so it fits uniquely only with the right
neighboring pieces. It would be too much of a coincidence that pieces fit so
well had they not been constructed for the purpose. Thus even before we
have completed the puzzle, we know we are making progress when we have
a unit of fitting pieces. The completed puzzle then both confirms our earlier
work and is confirmed by it. The justification is both piecemeal and holistic.
According to the argument of the previous chapter, dream interpretations
are like this. They are built step-by-step through fitting connections between
psychical elements, and then confirmed by an overall interpretation that pro-
vides meaning and unifies all that has preceded it. This interpretation is
justified as the best explanation of the dream-content and the associations
made with it—it is the simplest, most unified, most plausible causal model of
how the dream arose.
The key to this explication of Freud’s jigsaw puzzle analogy is the quality
of the connections between psychical elements. Freud posits that almost all
associations are dream-thoughts. What evidence, though, can we have, in any
particular dream, that that is indeed the case? As I argued in the previous
chapter, the evidence lies in the fittingness of the connections. In this chapter
I will present examples of the qualities at issue. There are three main kinds
(though this list is not intended as exhaustive): commonsense motive-match-
ing, distinctive similarity, and felt significance.
87
88 Chapter 4
Commonsense Motive-Matching
Our first quality is well illustrated by the supper-party dream discussed in the
previous chapter. Glymour indicates that in this dream it is the proximity of
events and their sharing of independent features that makes the connection
fitting. But I think there is more to it than this. The thought of the skinny
friend carried with it certain emotions and motives: the patient’s jealousy
toward her friend and her wish that her friend should not grow stouter in
order to attract her husband still more. Here is the crucial point: This motive,
that the skinny friend should not grow stouter, fits most appropriately to the
content of the dream. Recall that the dream was about the patient’s inability
to give a supper-party, that the theme of supper-parties had been evoked by
the friend herself, and that the relevance of supper-parties to stoutness had
been brought home to the patient by her husband’s remarks of the previous
day. We therefore have a motive—to prevent the friend from growing stout-
er—and a dream scenario in which a supper-party, which would help the
friend grow stouter, cannot occur. Given that we know that wishes can be
represented as fulfilled in dreams in this way, it seems a most apposite
explanation that this is what occurred in this case.
A big part of what makes this an apposite explanation is that the motive
matches the content in the same way as would occur in a commonsense-
psychological explanation of an action. I call this commonsense motive-
matching. Commonsense psychology involves making inferences about the
motives on which people act. When we know that somebody has a certain
motive and they exhibit behavior of a kind that one acting on that motive
would exhibit, we often infer that that was the motive out of which the person
acted and was thereby the psychological cause of their behavior. Indeed, we
need not even know that the person had the motive in question. It may be that
we infer the motive from the behavior. To give a trivial example, when you
observe somebody pouring herself a glass of water, you may infer that she is
acting on her motive to quench her thirst. Such is typical of commonsense
psychological inference. It is based on our understanding of human motives
and behavior.
We know that this pattern of inference can apply to dreams as well, for
that is the case with straightforward dreams. An example of such is Freud’s
own: having eaten salted anchovies just before he slept, he dreamt of drink-
ing a refreshing glass of water (1900, S.E., IV: 123). The cause of the dream-
content can be easily inferred from the match between the motive—his want-
ing to quench his thirst—and the content of the dream—his drinking water.
Freud offers another interesting example when he describes the dream of a
medical student who had difficulty waking early. One morning, in response
to his landlady’s calling through the door to wake him, the student dreamt
that he was lying in bed in the hospital where he was expected. His desire to
Interconnections 89
fulfill his duty without sacrificing his comfort was neatly fulfilled by this
dream, for as he was already in the hospital, he did not need to leave his bed
(ibid., 125). Since we know that dream-content can represent the satisfaction
of a motive in this way, it is reasonable to suppose that this happens more
generally. Thus when an association is a motive and the dream scenario
includes an event that would satisfy that motive, it can be reasonable to infer
that the motive was a cause of the dream-content.
Philosopher Jim Hopkins (1991, 1999) gives a number of examples of
this pattern of inference from Freud’s dream of Irma’s injection. For exam-
ple, as we saw in the introduction to this book, one of Freud’s associations
with the dream is his wish to absolve himself of responsibility for Irma’s
suffering. This connects with several elements of the dream, such as Freud’s
saying to Irma, “If you still get pains, it’s really only your fault,” and his
depicting Irma as suffering from an organic complaint, for which he cannot
be responsible. Here the match between the motive and the dream elements is
close enough to support the commonsense psychological inference that the
motive was a cause of these elements. Similarly, another associated thought,
Freud’s wish to blame Otto, connects with the part of the dream in which
“Otto had given her an injection . . . and probably the syringe had not been
clean.” Once again, the match between motive and dream element is close
enough to support the commonsense psychological inference that the motive
was a cause of the element.
Thus one kind of fittingness is commonsense motive-matching. It is ulti-
mately the most important for establishing the meaning of the dream, as, if
Freud’s wish-fulfilment claim is to be believed, this meaning will always be
a motive that needs to be matched with the content of the dream. Here the
matter might proceed as follows. Upon free association, the analyst discovers
within the dreamer a powerful motive. She then discerns a match between
this motive and the dream-content according to the pattern outlined above.
On this basis she infers that the motive was indeed a cause of the dream-
content. Insofar as the motive fittingly relates to all, or almost all, elements of
the dream, and thereby unifies them, she can infer that the motive is the
meaning of the dream.
In the examples considered above, the match between motive and content
is easily discernible. One may object, however, that the purported motive of a
dream does not always match the dream-content so transparently. Indeed,
unconscious, as opposed to preconscious, motives will match the content
only in a very indirect way. If Freud’s claim that most dreams are ultimately
motivated by unconscious wishes is to stand, then we need to go beyond
simple cases of commonsense motive-matching.
To do so requires an extension of commonsense psychology. According
to our normal commonsense psychological understanding, beliefs naturally
amplify desires. For example, if I have a desire to drink water and the belief
90 Chapter 4
that drinking from that glass would be drinking water, this amplifies my
desire to drink water, so I come to desire to drink from that glass (Hopkins
2012). On the relevant extension of commonsense psychology, a motive may
be amplified by something other than a belief. It may be amplified by an
association. Thus if I have a motive to x and I associate with x, y, then it can
happen that I form the motive to y, even though I do not believe that y-ing is
x-ing. It is in this way that a dream may represent the disguised fulfillment of
a wish: the wish is related to the dream only via one or more associations.
The rationale for this extension of commonsense psychology lies in its ex-
planatory potential, not just for dreams, but also for symptomatic behavior.
Freud believes he has ample evidence for making this extension from his
clinical experience, whereby otherwise inexplicable behavior is rendered
comprehensible by invoking just such explanatory resources. 1
Though a single instance of a motive being related to the dream-content
via associations may on its own be insufficient to sanction the inference that
the motive was a cause of the dream-content, that there should be numerous
such cases increases the plausibility of each such inference. Placed in the
context of a wider experience of the potential efficacy of this kind of expla-
nation, this form of inference becomes all the more reasonable. So even
without independent evidence that intermediary associations are dream-
thoughts, one may be able to validly infer that the motive is a cause of the
dream-content (and thereby that the intermediary associations are dream-
thoughts). But the strength of the inference would increase if one could
adduce additional evidence that the associations (at least some) are indeed
dream-thoughts. This requires appeal to other kinds of fittingness.
Distinctive Similarity
would be that the two items are similar in uncommon detail and with high
precision. The more uncommon the detail and the higher the precision, the
more distinctive the similarity. Elements thus associated with dream-content
are likely to be sources of it, for the distinctiveness of the details makes it
implausible that there are other such elements in memory, and it is also
implausible that dream-content so similar to the associated element was con-
structed anew from other psychical elements.
An analogy will illustrate. Two texts are distinctively similar insofar as
they show a precise match in the words used and the arrangements of these
words. The more precise the match, and the more unusual the words and
arrangements where there is such a match, the more distinctive the similarity.
On the basis of this distinctive similarity, knowing one of the texts to be
original, one can conclude that the other derives from it.
If one looks carefully, one will find that Freud’s analyses are replete with
this kind of fittingness. To give a couple more examples, recall those we
noted in the introduction to this book in relation to the Irma dream. Of the
woman he wished to replace Irma with, Freud had remarked, “I had found
her by a window in the situation reproduced in the dream;” and of the
inspection of Irma’s body, he observed, “The dull area low down on the left
seemed to agree in every detail with one particular case in which Leopold
had struck me by his thoroughness” (1900, S.E., IV: 110, 113; my italics). In
these cases, the language Freud uses strongly suggests distinctive similarity
between the dream element and the associated memory.
This feature, distinctive similarity, occurs most poignantly in visual im-
ages, but it need not be restricted to them. For instance, a passage of speech
heard in a dream might match exactly one in the agent’s memory, not just in
terms of the words spoken, but also in the manner and mood with which they
are spoken. This is perhaps the case in the Irma dream, where the words of
Dr. M.—“It’s an infection, but no matter. Dysentery will supervene and the
toxin will be eliminated”—may have sounded distinctly similar to the words
of the foolish doctor in Dr. M.’s anecdote (either as Dr. M had spoken them
or as Freud had imagined them)—“No matter . . . the albumen will soon be
eliminated” (ibid., 115). Also, more subtly, but perhaps more consequential-
ly, a feeling experienced might be of an emotional tone precisely similar to
that of an associated thought. 2
Felt Significance
sees in the night is a fox, and one has reason not to prefer the leading
alternative explanations (too big to be a cat, too lithe to be a dog, too vivid to
be an illusion, etc.), then it is reasonable to infer that it is a fox.
What often underlies this feeling of significance is best expressed by
Freud himself: “In the example of the abandoned supper-party the connection
was given at once: ‘smoked salmon,’ being the friend’s favourite dish, was
an immediate constituent of the group of ideas which were likely to be
aroused in the dreamer’s mind by the personality of her friend” (1900, S.E.,
IV: 175). More generally, we can speak of ideas drawn together by pre-
existing lines of subjective significance—that is, relations of significance for
the dreamer that predate the dream—such that the one idea is likely to arouse
the other in the dreamer’s mind. In the smoked salmon case, the lines of
significance were relatively long-term and based on objective relations, so
easily grasped by a third party. In other cases, however, they may be more
recent and idiosyncratic. Such grounds give rise to a sense of relatedness that
is the main feature of the felt significance of a connection, and also captures
much of what Blass means by the experiential quality of meaningfulness (see
previous chapter).
There is though usually more to felt significance than just this sense of
relatedness. Three additional features are also often present. The first is emo-
tional resonance. This is a sense that the connection relates to matters of deep
emotional significance for the dreamer. The second is immediacy. This is the
fact, mentioned also by Blass in relation to EQM, that the feeling falls upon
the dreamer as soon as the connection is made. The third is certainty. Though
the manifest similarity between the two items connected may be only super-
ficial, one nevertheless often has an instant conviction that the associated
element is the meaning of the starting element. What is most noteworthy
about this third feature is the surprising degree of the conviction: it exceeds
any objective case that can be made for the causal hypothesis and seems
more akin to the self-knowledge one has when experiencing one’s own con-
scious psychological states. Bleuler expresses it well when he remarks,
“Whenever an interpretation or part of it is unclear, in our experience, it is
the dreamer who mostly has the definite feeling of correctness” (in Marinelli
and Mayer 2003, 163–64). As noted in the previous chapter, there are clues
in Freud’s account that this is the kind of experience his patients sometimes
had, and when so, this seems to have played a role in confirming to him the
accuracy of his interpretation. 3
This kind of fittingness is also important to Freud’s claim that his method
can uncover a deeper-level interpretation of the dream: not just a precon-
scious meaning, but an unconscious one. Distinctive similarity and simple
commonsense motive-matching on their own are unlikely to take one this far
in an interpretation, for, on Freud’s account, one must push beyond recent
memories and the rational amplification of motives toward the kind of asso-
94 Chapter 4
Conclusion
Besides the qualities we have seen, there are also numerous other features
that can contribute to the fittingness of a connection. One such is the recency
of an association. By this I mean that the associated thought relates to some-
thing that had recently (though prior to the dream) appeared in the dreamer’s
life. The thought could, for example, be a memory from the previous day’s
events, as was the case in the supper-party dream; or it could be an older
memory that had recently been revived; or, not an episodic memory at all,
but another recently occurring thought. Such proximity in time enhances the
fittingness of a connection due to the reasonable background assumption that
thoughts that have been active recently are more likely to be sources of the
dream than ones less recent.
Another feature that enhances the fittingness of a connection is the emo-
tional salience of the association. By this I mean that the associated thought
Interconnections 95
is one that stands out due to its emotionality. This feature enhances the
fittingness of a connection due to the reasonable background assumption that
thoughts of higher emotional salience are more likely to be sources of the
dream than ones less so.
Putting all this together, we have a tenable account of what constitutes
fittingness. Thus we make good Freud’s jigsaw puzzle analogy. Recall that
the problem with the analogy was that Freud failed to specify what corre-
sponds to the flush fit we get between puzzle pieces. My account offers a
solution. The fit between jigsaw pieces corresponds to fitting connections
between psychical elements. A thought may have several such connections,
of varying degrees of fittingness, with the dream-content and/or with other
parts of the dream interpretation. As such it may fit into the dream interpreta-
tion in the same way that a jigsaw piece fits into a portion of a puzzle.
It should also be remembered that the evidence for the overall interpreta-
tion is not just that it connects to the dream through such fitting connections.
It is that it does so in a way that accounts for all, or nearly all, of the elements
of the dream, and that it is moreover continuous with the rest of the dream-
er’s psychical life. As Freud puts it, “If one succeeds in arranging the con-
fused heap of fragments . . . so that the picture acquires a meaning, so that
there is no gap anywhere in the design and so that the whole fits into the
frame—if all these conditions are fulfilled, then one knows that one has
solved the puzzle and there is no alternative solution” (1923, S.E., XIX: 116).
The evidence, therefore, does not rest simply on a case-by-case assessment of
individual connections, but also holistically, on how the thought fits within
the overall picture.
One thereby gets something like the following sketch of how a well-
evidenced dream interpretation unfolds. Through free associating with ele-
ments of the dream, the analysand brings up numerous thoughts. Some of
these are linked to the content of the dream by connections of distinctive
similarity of varying degrees. Some are also linked with the content of the
dream by connections of felt significance of varying degrees. As further
thoughts emerge, a common theme becomes apparent. This takes the form of
a single motive (or a cluster of related feelings centered on a particular
motive). The motive is one of deep emotional significance for the analysand
and fits intimately with those concerns that have been, consciously or uncon-
sciously, most preoccupying her. This motive can then be seen to match the
content of the dream according to the pattern of commonsense motive-
matching that I described earlier, though often only via several interlinking
thoughts from those previously brought up. The manner of the connection is
such that all, or almost all, elements of the dream are accounted for. Based on
this the analyst concludes—justifiably on the account I have given—that this
motive is the meaning of the dream.
96 Chapter 4
right kind. For example, during analysis, one might make numerous associa-
tions with connections to the dream-content without experiencing a sense of
significance, and then for one connection experience it in a strong way, with
no clue from the context as to why suggestion should have alighted on this
rather than any other connection. Indeed there may be no evident source of
suggestive cues at all (as is so, most particularly, in self-analysis). Further-
more, the connection in question may be unusual; its phenomenology may be
distinctive—meaningful, emotionally resonant, and of the character of
psychological self-knowledge; and the interpretation it leads to may be such
as one would strongly wish to deny. All these facts would point to the
significance explanation, rather than the suggestion explanation. Thus it
seems at least possible to be personally justified in believing that one has the
right kind of experiential state.
To put the point another way, one should not give the suggestion hypothe-
sis more credence than it is due. It is only one possible explanation of felt
significance. In this it competes with the significance explanation: that the
thought so strikingly associated is so because it picks out a significant inter-
nal connection (that is, one not due to suggestion). There is no reason to give
a priori preference to the suggestion explanation. One should compare the
hypotheses as one would any competing explanations, by looking at how
well they explain the evidence. The evidence includes the subtle phenomeno-
logical features of the experience and the circumstances in which it arose.
There seems no good reason why these details might not favor the signifi-
cance over the suggestion hypothesis. Thus there seems no good reason for
supposing that one cannot be justified in believing that one has the right kind
of experiential state.
How does one know that felt significance is relatively rare? Might it not be
much more common than you envisage?
This could be ascertained by personal experiment. The best test is to see
how striking one finds a connection that somebody else finds striking for
idiosyncratic reasons. An example of such is the peculiar connection I once
happened to draw between “baboon” and “barbarian” on the basis of the
alliterations between these words. The words are clearly related in sound, but
this is not a connection that would strike most people as particularly signifi-
cant. It did so to me, however, when I drew this association some years ago. 6
The felt significance this connection had for me is, in my estimation, rela-
tively rare—I would not have experienced just any connection as significant
in this way. Yet for others it offers a contrast to their own experience: it
would appear to them, I venture, a most tenuous connection.
Maybe the evidence from felt significance applies in the case of self-analysis.
But how would an analyst gain access to a subjective state like this?
Interconnections 99
The analyst has access to the dreamer’s subjective states in the same way
that we all have access to others’ subjective states. We can, of course, never
experience such states for ourselves, but we can often know that a person is
having an identifiable experience.
One way of knowing this is through behavioral observation. This is well-
illustrated by the example of how we know of another’s emotions. We often
know what a person is feeling through our observing the typical behavioral
indicators of that emotion. This is so even for subtle feelings. For example, a
flicker of irritation often does not go unnoticed. What applies for emotion
can apply just as well for disbelief or for sudden realization. Even when the
movements are slight, they are seldom missed by an intimate acquaintance.
One could, of course, also ascertain another’s state simply by asking
them. As with behavioral cues, this is not foolproof, but it is usually reliable.
There are, however, also other ways for the analyst to draw relevant
conclusions. One is that sometimes a connection can be striking in its felt
significance for the analyst. This is so when the basis of the felt significance
consists in culturally shared ideas—that is, when it is such as would strike
anyone from the same culture as significant. When this occurs, other people
of the same culture, including the analyst, can experience the connection as
fitting in the same way the analysand does.
Consider the following analogy. Talented comedians have an acute sense
of what their audience will find funny. This is so even though they have no
specific knowledge of their thoughts and feelings. Rather, the thoughts and
feelings comedians work from are ones they can reasonably expect most of
their audience to share. Given that much humor is culture-specific, these will
often have to do with culturally shared beliefs and values. Through his under-
standing of such normal lines of thought and response, a comedian can
construct a routine that most of the time gets it right as to what the audience
would find funny. In a similar way, through his understanding of culturally
normal lines of thought and response, a talented analyst can often get it right
as to what the analysand would feel as significant.
The analogy can be extended, for cultures can be defined in increasingly
narrower ways. A small locale, a group of people, even a pair of friends can
have their own “culture.” Correspondingly, a comedian may have in-jokes
that work within each of these increasingly smaller circles, to the point where
a joke works only between him and another individual. It works in part by his
knowing just what makes this other individual tick—his having an intimate
sense of how she would respond to various ideas. He knows what she alone
would find funny. In a similar way, an analyst who has worked with a patient
over a considerable period of time and has come to know the patient’s inner
life better than anyone can develop an intimate sense of how she would
respond to various thoughts. Through this sense, he may acquire the ability to
predict reliably what the patient would experience as significant. This means,
100 Chapter 4
crucially, that he can know what she would have experienced as significant
regardless of the interpretative process.
When these facts pertain, one would be hard-pressed to find a unified expla-
nation of them that is better than the one on offer, namely, that the interpreta-
tion given is a meaning—and thereby a cause—of the dream.
The above relies on free association being genuinely free. But Freud himself
asserts that this is not the case. He admits that the process often fails to
arrive at a dream-thought.
This objection takes its cue from Freud’s own comments. For example, in
his 1932 lectures he explained:
Interconnections 101
The associations to the dream are not yet the latent dream-thoughts. The latter
are contained in the associations like an alkali in the mother-liquor, but yet not
quite completely contained in them. . . . An association often comes to a stop
precisely before the genuine dream-thought: it has only come near to it and has
only had contact with it through allusions. At this point we intervene on our
own; we fill in the hints, draw undeniable conclusions, and give explicit utter-
ance to what the patient has only touched on in his associations. (1933, S.E.,
XXII: 12)
The reason that an association might stop before the genuine dream-thought
is resistance. Under pressure from the force of repression, the dreamer en-
gaged in free association fails to arrive at the dream-thought and instead
arrives at a compromise—a halfway house between the dream element and
the dream-thought. It therefore requires additional work to infer the dream-
thought itself.
This, however, does not always apply. In practice, Freud often takes
associations with dream elements to be dream-thoughts. The associations in
the supper-party dream are a case in point, as are most of the associations in
the dream of Irma’s injection, and those in many other dreams besides.
Even when it does apply, it is still possible to validly infer a dream-
thought. Recall how earlier I argued that, once one has inferred that certain
associations are dream-thoughts, it may be possible to infer an overall inter-
pretation of the dream. This would occur if there is a coherent set of con-
cerns, plausibly attributable to the dreamer, that has fitting connections with
all, or most, of the dream-thoughts thus far uncovered. My new point is this:
what applies for the whole, can also apply for the part. When the analysand
gives a series of associations to a single dream element, none of which
themselves are dream-thoughts, these associations can be treated like the
dream-thoughts in the inference just described. This would occur if there is a
thought, plausibly attributable to the dreamer, that has fitting connections
with the associations in the series.
As an illustration, consider Freud’s example of retrieving a forgotten
name (1916, S.E., XV: 111). He had forgotten the name of the principality
that has Monte Carlo as its capital. He then proceeded to make a series of
associations with Monte Carlo, none of which were themselves what he was
looking for. But there were enough clues in this series for the recovery of the
forgotten name. The series was “Piedmont, Albania (which was then quickly
associated with Montenegro, on the basis that ‘Albus’ is Latin for ‘white’ and
‘negro’ is Italian for black), Montevideo, Colico.” Each of these words
shares a syllable with “Monaco.” Freud was thus able to recover the source
thought on the basis of its clear connections with each of the associations in
the series. In a similar way, when the analysand produces a series of associa-
tions with a dream element, none of which are the dream-thought, the analyst
may be able to infer the thought on the basis that it has fitting connections to
102 Chapter 4
all (or almost all) of the associations in the series. Indeed, it is often the case
that the same type of connection occurs more than once (as was the case with
the syllable “mon” in the above example), raising the likelihood of its signifi-
cance. In this way the analyst may be justified in inferring a thought as the
dream-thought even though it was not directly associated with the dream
element by the analysand.
Though the case with dream interpretations is different from the recall of
proper names, in that we have no independent way of confirming the source
element, there are, nevertheless, a number of considerations that can count in
favor of a particular interpretation: that it unifies the associations through
fitting connections; that it is, as predicted, an objectionable thought; that it
fits with the pattern of the rest of the dream interpretation; and that it fits with
the general theme of the dreamer’s psychical life (a helpful analogy here is
solving crossword puzzles). In this way one can have a strong inference to
the best explanation that a particular thought is a dream-thought.
Allowing this kind of inferential procedure means that one can have
interpretations that are more complex and penetrate deeper, but are no less
principled than in the straightforward cases we have seen. Clearly, the more
inferential steps required the less probable the conclusion. But also, the more
complexity, the more possibility of internal confirmation. In this way, such
an analysis may be justified.
Many of the connections that Freud picks out in his dream interpretations do
not seem fitting.
A proper judgment about the fittingness of any particular connection can
often only be made by one with intimate acquaintance with the dreamer. This
is because a judgment of the felt significance of a connection often has to be
made relative to what they would feel significant, not what we would. This
requires a close familiarity with the dreamer’s inner world. What seems
tenuous to us might not be tenuous for them (cf. “baboon”-“barbarian” exam-
ple).
Michael’s use of the dinner party dream to illustrate various points obscures
his defense of Freud’s method because the method he describes is not based on
a traditional analysis. One wishes that he had picked a dream whose interpreta-
tion did not lie on the surface, one of the many analyses which have been
challenged by critics over the years—the dream about his friend Otto looking
as though he had Basedow’s disease, for instance. (Sand 2012b, 99)
Sand is apparently unaware that I have in fact analyzed in detail exactly the
dream she recommends, in a paper published considerably before her above
comments (Michael 2008, 59–62). I chose this dream because it is the one
she had used to exemplify her objections to Freud’s method, to which I was
responding (Sand 1993). I similarly chose the supper-party dream because it
is the one Glymour uses in his influential argument against Freud’s method.
It is a little unfair to criticize me for using precisely the dreams that the critics
themselves have used to advance their point.
I am confident that my defense of Freud’s method applies to many of the
dream interpretations Freud gives in The Interpretation of Dreams, including
the supper-party dream, the dream of Otto’s illness, the dream of Irma’s
injection, and the dream of the yellow beard, to name only ones we have
already encountered. These I think are all cases in which Freud takes the
interpretation to be justified because it offers a fitting explanation of the
dream-content and associations with it along the lines of the principles I have
outlined.
It is, however, possible that many of Freud’s interpretations, in particular,
the more intricate ones from his clinical work, are not valid according to the
104 Chapter 4
So far I have defended Freud’s method against the accusation that it is based
on a fallacy. But what did Freud himself have to say against this criticism?
And how does my argument relate to this?
As mentioned earlier, Freud reported a version of the criticism we have
been considering in The Interpretation of Dreams. Here it is:
Our critics argue . . . along the following lines. There is nothing wonderful in
the fact that a single element of the dream should lead us somewhere; every
idea can be associated with something. What is remarkable is that such an
aimless and arbitrary train of thoughts should happen to bring us to the dream-
thoughts. The probability is that we are deceiving ourselves. We follow a
chain of associations from one element, till, for one reason or another, it seems
to break off. If we then take up a second element, it is only to be expected that
the originally unrestricted character of our associations will be narrowed. For
we still have the earlier chain of thoughts in our memory, and for that reason,
in analyzing the second dream-idea, we are more likely to hit upon associa-
tions from the first chain. We then delude ourselves into thinking that we have
discovered a thought which is a connecting point between two elements of the
dream. Since we give ourselves complete liberty to connect thoughts as we
please . . . we shall find no difficulty in the long run in concocting out of a
number of “intermediate thoughts” something which we describe as the
dream-thoughts and which—though without any guarantee, since we have no
other knowledge of what the dream-thoughts are—we allege to be the psychi-
cal substitute for the dream. But the whole thing is completely arbitrary; we
are merely exploiting chance connections in a manner which gives an effect of
ingenuity. In this way anyone who cares to take such useless pains can worry
out any interpretation he pleases from any dream. (1900, S.E., V: 527)
Sand raises the following objection. Even if Freud is right in his assumption
that the mind is purposive, that still does not show that associated ideas are
causes of the dream, for one can set up a plausible alternative to Freud’s
account. This alternative, suggested also by Wittgenstein (see the introduc-
tion to the previous chapter), is that free associations are drawn toward
themes that are of central preoccupation to the dreamer. The dreamer will
tend to associate with any element an idea that is close to a matter of deep
concern for him; hence as the process of association proceeds, he will con-
verge upon this matter. This holds regardless of his starting point, and so his
ending up in the same place says nothing about the origins of the elements
from which he began. I call this the thematic-guidance hypothesis. It is a
plausible alternative to Freud’s view that associations inevitably lead back to
origins.
Freud, in his response to the imaginary critic in The Interpretation of
Dreams, does not address this point. This is because the specific criticism he
Interconnections 107
was focusing on was not this, but one that rested on the assumption that free
associations were arbitrary and aimless. Nevertheless, by not addressing the
point it would seem that he has failed to discharge his burden of proof. As it
happens, though, in a later work he tackles precisely this objection. In his
Introductory Lectures of 1915–1916, Freud actually welcomes the thematic-
guidance hypothesis. Indeed, he considers it to be indisputable. Specifically,
he thinks it the case that if one associates with a random item, such as a
random number, one will eventually be led to thoughts of psychological
importance: “The associations to numbers chosen at random are perhaps the
most convincing; they run off so quickly and proceed with such incredible
certainty to a hidden goal that the effect is really staggering” (1916, S.E.,
XV: 107).
It is ironic that Freud welcomes the thematic-guidance hypothesis, where-
as Sand treats it as an objection. She recounts the example of an experiment
from the 1920s of a patient who associated with a randomly chosen number
to the wish that her period would start soon, as she feared she was pregnant
(Sand 2012a, 89). This, she thinks, refutes Freud’s view that associated ideas
will always lead back to originating thoughts.
The difference in Sand’s and Freud’s attitudes to the thematic-guidance
hypothesis can be explained thus. For Sand it shows that Freud’s glass is
half-empty: free associations can be purposive without taking one back to the
origins of the starting thought. For Freud it shows that his glass is half-full: it
is confirmation of his idea that free association is not arbitrary and aimless,
but is in fact guided by matters of psychological importance. But Freud is
also aware that this is not enough to justify his use of free associations in
dream interpretation. He puts the objection in the words of another imaginary
critic:
We acknowledge now that thoughts that occur to one freely are determined
and not arbitrary as we supposed. We admit that this is also true of thought in
response to the elements of dreams. But that is not what we are concerned
with. You assert that what occurs to the dreamer in response to the dream-
element will be determined by the psychical background (unknown to us) of
that particular element. This does not seem to us to be proved. We quite expect
that what occurs to the dreamer in response to the dream-element will turn out
to be determined by one of the dreamer’s complexes, but what good does that
do us? This does not lead us to an understanding of dreams but, like the
association-experiment, to a knowledge of the so-called complexes. But what
have they got to do with dreams? (1916, S.E., XV: 109)
Freud’s response is to point out that in the case of dream interpretation the
starting item is not random, but is an element of the dream. This he thinks
makes a crucial difference. At least, it does if we accept the additional prem-
ise that there are unconscious elements behind the dream. For then it is
108 Chapter 4
plausible that the associations the dreamer makes will be determined by these
particular unconscious elements, rather than unrelated elements. As Freud
puts it, it is “therefore not precisely fantastic to suppose that the further
associations linked to the dream elements will be determined by the same
complex as that of the element itself and will lead to its discovery” (ibid.,
110).
What is important here is not that this argument is watertight. What he is
offering is a plausibility argument. He is claiming, not that he has justified
the view that free associations will lead to the origins of the dream, but that
this is a reasonable enough view to entertain.
Freud then proceeds to argue that this view—that free associations from a
stimulus that originates in an unconscious element will lead back to that
element—receives confirmation from his experiences of trying to remember
proper names. An example is that of the case where he forgot the name of
Monaco, which I discussed in the previous section. In this he was led back to
the forgotten name through free associations with Monte Carlo (ibid., 111).
Freud thinks this confirms the following principle: where there is an uncon-
scious element that is behind a linked substitute element and one associates
with the substitute element, the associations will be guided both by the sub-
stitute element and the unconscious element it is a substitute for, so that
eventually they will offer enough clues to enable the recovery of the uncon-
scious element. If this principle is general, then it will apply also to dream
elements. So assuming there are unconscious elements behind the dream, it
gives us confidence that the associations will eventually provide enough
clues as to what this is.
This argument should be considered in conjunction with the ones made
earlier, in Freud’s responses to the first imaginary critic. These are to the
effect that his clinical work offers significant confirmation of the principle
enunciated above. For example, his work on hysterical symptoms apparently
shows that when patients free associate with a symptom, they are eventually
led back, by a chain of associations, to a traumatic event that bears thematic
relation to the symptom and which closely preceded the first appearance of it.
According to Freud, on expression of this event and release of the affect that
accompanies it, the symptom disappears. It seems then that the unconscious
idea behind the symptom was guiding the free associations, to the point of
their eventually leading back to it. As such, the principle that where there are
unconscious ideas at play they will inevitably guide one’s free associations
can be seen as a generalization of that which underpins his clinical work.
What are we to make of Freud’s argument? It seems to fall short of the
“proof” that he promised. He himself concludes by saying that “we seem to
have produced some justification of our technique” (ibid., 112; my italics).
This indicates that he does not think that he has fully justified the method.
Freud is aware that he is taking liberties, but feels in his right to do so by the
Interconnections 109
nature of the subject matter. “If anyone finds the whole thing too laborious
and too insecure, or if anyone is accustomed to higher certainties and more
elegant deductions, he need go no further with us. I think, however, he
should leave psychological problems entirely alone, for it is to be feared that
in this quarter he will find impassable the precise and secure paths which he
is prepared to follow” (ibid., 102). But here it is easy to misunderstand what
is going on. It is, I think, best to interpret Freud’s argument—that is, both the
initial plausibility argument and the subsequent alleged confirmations—as
another, stronger, plausibility argument. He reasons from what he takes to be
reasonable, if unproven, premises to the tentative conclusion that free associ-
ations will take us to the unconscious origin of a dream, and then offers some
degree of confirmation of this conclusion through his examples. He is right
that this is some justification of the technique, but it is not enough for out-
right acceptance of it. His putative confirmations at best merely ramp up the
plausibility; they are not decisive confirmations. That is acceptable for the
purpose of his argument. What he needed to show was that his technique was
worth serious consideration, was worth trying. But—pardon the cliché—the
proof of the pudding is in the eating. The final justification must surely lie in
whether the method actually works—whether it gives the results it is sup-
posed to give. In this case, this would be that it takes one to the unconscious
material behind the dream. That can only be ascertained upon application of
the method.
Let me pause to reiterate the point. Sand thinks that Freud uses free
association because he has a prejudice that it would work. In particular, she
thinks that even if free associations are purposive, that does not mean they
would take one to the origins of the dream. Freud offers a response to this
that shows that there is some reason to think they would take one to the
origins of the dream. Sand is unlikely to be satisfied. The reasons are insuffi-
cient for full conviction. My point, however, is that if Sand is looking for
Freud to justify fully his method with such reasoning, then she is misunder-
standing the justificatory situation. It would surely be asking too much to
justify the reliability of any method in this way. Whatever one’s reasons for
thinking that a method might work, what matters is whether it does work.
Freud, despite some misleading remarks, does appreciate this. This is why he
advises his listener/reader to first try the method on his own dreams, rather
than on someone else’s, because then “the process carries more conviction”
(ibid., 114). It is why he later writes “now perhaps is the time to take a dream
and try our technique upon it and see whether our expectations are con-
firmed” (117; my italics). It is why, even later, he writes “Only the outcome
of our experiment can show whether we are right” (1933, S.E., XXII: 9). It is
why so much of his writing on dreams is devoted to giving actual examples
of dream interpretations. He appreciates that the full justification of the meth-
od can only come through its application.
110 Chapter 4
This is where my own defense of Freud’s method comes in. It validates the
reasoning that lies behind Freud’s belief that his method works in practice.
All the other reasoning that Freud uses can be considered as just a plausibil-
ity argument. What matters is that, when he applies his method, he can see
that it actually works—that it really does uncover the sources of the dream.
Freud clearly believes this is the case. Thus he writes, “Whenever I began to
have doubts of the correctness of my wavering conclusions, the successful
transformation of a senseless and muddled dream into a logical and intelli-
gible mental process in the dreamer would renew my confidence of being on
the right track” (1933, S.E., XXII: 7). He is impressed by the results of the
application of his method; he thinks the method manifestly works. The speci-
men-dream interpretations he gives in his book are intended to show this.
They do so by being impressively fitting explanations of the dream elements
and their associations—that is, by being the best explanation of the evidence
at hand in just the way I have argued that they can be.
What I am asking of the critics is to step back a little from Freud’s own
arguments and rhetoric and try to understand the rational process at work
here. Freud may well, as a result of various prejudices and assumptions, have
come to have confidence that the method of free association will work for
dreams. But none of this would have mattered if, in practice, the associations
led nowhere. As it happens, though, they do lead somewhere. They lead to
material that allows Freud to offer some impressively fitting interpreta-
tions—ones that well explain the dream elements in light of the associations,
whilst also fitting into the dreamer’s psychical life in general. They allow a
completed jigsaw puzzle. That they do so offers confirmation of the assump-
tions that Freud based his initial confidence in the method on. From an
epistemological point of view, it is the confirmation that is important.
Freud was, from this point of view, wrong, in The Interpretation of
Dreams, to emphasize the long response to his imaginary critic over the first
short response. The first short response is that of “appealing to the impres-
sion made by our interpretation, to the surprising connections with other
elements of the dream which emerge in the course of our pursuing a single
one of its ideas, and to the improbability that anything which gives such an
exhaustive account of the dream could have been arrived at except by follow-
ing up psychical connections which had already been laid down” (1900, S.E.,
V: 528). This response has the elements of what is epistemologically impor-
tant. Freud at the time underplayed these elements, perhaps because his way
of thinking was not that of an epistemologist but of a psychologist. He was
guided by the need to establish how the mind works. In order to defend his
own view of how the mind works, he felt the need to refute the alternative
view contained in the criticism he was responding to. But when we look at it
Interconnections 111
CONCLUSION
Since its inception, Freud’s method of dream interpretation has been strongly
and repeatedly criticized. Critics allege that Freud assumes that free associa-
tions are causes of the dream without good reason for thinking so. In this
chapter I have defended Freud’s interpretative claims against this criticism.
My defense is not a general justification, since the relevant evidence is not
unambiguously accessible to those outside the analytical setting. It is, rather,
a validation. I argue that there are certain kinds of evidence that, if present
during the interpretative process, support the conclusion that thoughts uncov-
ered by free association are causes of the dream. This undermines the charge
112 Chapter 4
that Freud’s method is based on a fallacy and establishes that one can be
personally justified in believing in a Freudian-style interpretation.
Freud aspired to something more than I have been able to establish in this
part of the book. He thought his theory and method were scientific, hence
acceptable to any rational person, not just those with personal experience of
dream interpretation. Whether or not this is so will be the topic of the next
two chapters. For now we must settle for something less, but nevertheless
something that refutes the most serious objection raised by critics.
NOTES
1. The following example illustrates. One of Freud’s patients, nicknamed by him the Rat
Man, engaged in the following odd behavior when on holiday in the mountains: after lunch he
would immediately, in the blazing heat of the day, go for a run up the mountain to the point of
exhaustion, ostensibly in order to lose weight. Freud was able to give an interpretation of this as
follows. The Rat Man’s companions on holiday included “his lady” and her English cousin
Richard, and the Rat Man was jealous of Richard and wanted to rid himself of him so he could
have his lady to himself. On Freud’s interpretation, he wished to murder Richard, but was not
prepared to act on this wish. Because Richard’s nickname was Dick, and dick in German means
fat, the Rat Man’s wish to rid himself of Richard was displaced to the wish to make himself less
fat. This manifested in the compulsive urge to lose weight, which was enacted in the bizarre
ritual described above. Freud further explains that the Rat Man’s wanting to lose weight in this
punishing manner was due to his wanting to punish himself for the murderous wish (1909, S.E.,
X: 187–89; Wollheim 1993, 96–97).
2. There are two kinds of case. The first kind is where one can point out exactly to what the
distinctive similarity pertains. This is most like the examples given above. Depending on just
how distinctive the similarity is, there may be little doubt that the elements are causally related.
A second case is where there is an apprehension of distinctive similarity but one cannot point
out exactly wherein it lies—it eludes articulation. There is a sense that the elements are distinct-
ly similar, but one cannot quite put one’s finger on what this consists in. This too can be an
instance of fittingness, though it blends into the kind I call “felt significance.”
3. If Freud’s theory is true, then we may suppose that these four features of felt signifi-
cance correspond to aspects of a dream’s formation. Relatedness would correspond to the fact
that the dream-work is opportunistic and so will take advantage of any pre-existing relatedness
between psychical entities. Emotional resonance would correspond to the fact that the direction
the dream-work takes is often guided by the deep emotions related to the meaning of the dream.
Immediacy and certainty may be consequences of one’s somehow “perceiving” the trace of a
recent causal path between the two elements—presumably the one taken in the formation of the
dream.
4. Cf. also my own example of the connection between “baboon” and “barbarian” that I
describe in a later endnote.
5. This distinction has been emphasized by the philosopher John Searle (1998, 42–45).
6. As it happens, in this case I know why the connection struck me as significant. It picked
out a recurring element from my stream of consciousness—something that appears insignifi-
cant, a mere flotsam of thought, but that, as it turns out, had to do with matters of importance
for me. It relates to the etymological theory that the word “barbarian” contains a parody of the
speech of foreigners. The Greeks had used this word to demarcate themselves from others, as in
the derogatory expression, “whoever is not Greek is a barbarian.” In idle moments I had often
played around with the sound “bar-bar” or variants of it, wondering how exactly the speech of
foreigners would have sounded to Greek ears. As a person of Greek origins, though born and
raised in a foreign land, the demarcation was significant for me. The sound with which I was so
often preoccupied represented the boundary between Greekness and foreignness, and I knew
Interconnections 113
not which side of this boundary I lay. One morning I found myself pondering a dream, part of
which involved an encounter with a troop of baboons. I associated with this the word “barbar-
ian,” precisely on account of the kind of sound that had preoccupied me in the past. This
connection between “baboon” and “barbarian” struck me instantly as meaningful. The conclu-
sion that it was indeed meaningful was supported by additional considerations: other parts of
the dream also had to do with issues of ethnic identity, and the night before I had had a
conversation precisely on this topic, one that had left me with a sense of loss. The image of
baboons also brought to mind a negative stereotype of Greek men. It seems that I was, in the
dream, distancing myself from a Greek identity, while ironically (and spitefully) equating such
with barbarism and brutishness.
7. This relates to his “prejudice” that “mental processes are determined” (1910, S.E., XI:
29).
8. See Chapter 5 for a description of such cures.
Part III
False Negatives
117
118 Chapter 5
POPPER
Popper is most famous for his demarcation criterion, the purported way of
distinguishing science from non-science. This is tied in with his rejection of
psychoanalysis as a science, for as he recounts in his autobiographical ac-
count of the development of his thought, it was the contrast that he saw
120 Chapter 5
between psychoanalysis and physics that led him to the criterion. According
to Popper, genuine sciences like physics attempt to test their theories, while
pseudosciences like psychoanalysis do not. Popper observed that advocates
of psychoanalysis support it by pointing to its explanatory successes, but
physical theories are not accepted because of their successes but because of
their ability to withstand rigorous testing. The example that most impressed
him was Einstein’s theory of general relativity. The theory has the conse-
quence that light should be affected by gravity. This is something that can be
tested: the observed incidence of light from a known star during an eclipse
would establish whether the sun’s gravity has the predicted influence. In
1919 the British physicist Sir Arthur Eddington carried out the necessary
observations for such a test. If Eddington’s observations had deviated signifi-
cantly from Einstein’s prediction, then, according to Popper, the theory
would have been refuted. But they did not. For Popper, it was not that the
theory had passed the test that rendered it scientific, but that it could be
subjected to such a test at all. It was precisely this—the ability to specify
what would refute it—that was missing from Freud’s theory. With this in-
sight emerged Popper’s famous demarcation criterion: the difference be-
tween a scientific and a non-scientific theory is that the former but not the
latter is empirically falsifiable.
Popper’s criterion is, appropriately enough given his views, startlingly
bold. It says that all and only genuinely scientific hypotheses are empirically
falsifiable. In formal terms, a hypothesis is empirically falsifiable if there
exists a logically possible observation statement (or set of such) that logically
contradicts it. For a trivial example (given by Popper), the hypothesis “all
swans are white” is inconsistent with the observation statement “this swan is
not white,” and hence is falsifiable. This purely formal criterion, however,
needs supplementing. For it is all too easy to take a falsifiable hypothesis and
defend it against some actual observation statement by criticizing the state-
ment. Indeed, on Popper’s view, such criticism is always logically possible,
for all observation statements are fallible. Thus, to extend the trivial example,
one could always find some ground for claiming that the observed swan was
not really non-white or was not really a swan. In fact, the attribution of
whiteness or swanhood depends on other scientific theories, and since on
Popper’s view all theories are fallible, one could always defend one’s
hypothesis by arguing another theory as false. That, for Popper, would be a
mistake, for if such were allowed, then any hypothesis could be retained, thus
negating the value of the criterion. Falsifiability is not simply a descriptive
criterion, it is that which guarantees the credibility of science and ensures its
progress. As such, Popper stipulates that attempting to save a hypothesis in
the light of disconfirming evidence is forbidden. As long as the falsifying
observation statement is endorsed, the statement should be accepted and the
False Negatives 121
There are numerous problems with Popper’s argument against the scientific-
ity of Freud’s theory of dreams. The most fundamental pertain to the demar-
cation criterion itself. As Popper would presumably want it to be, it has been
subjected to considerable critical scrutiny by historians and philosophers of
science. Most have concluded that, in the form in which Popper expressed it,
it fails to stand up to this criticism.
History is relevant because Popper intended his theory to be descriptive
as well as prescriptive. Understandably so, for if it turned out that the most
successful scientific theories in history—those that may reasonably be taken
as paradigms of scientificity—were unfalsifiable on Popper’s terms, then his
criterion would appear chimerical. But that is indeed what historians of sci-
ence have claimed. Lakatos summed it up with the memorable phrase that
scientific theory “floats in an ocean of anomalies.” That is to say, accepted
scientific theories always face numerous apparent falsifiers, yet are main-
tained despite this. To give just one prominent example, Copernicus’s theory
was “falsified” by astronomers’ continued failure to observe stellar parallax:
the theory predicts that the relative position of stars as observed from the
earth will change as the earth moves in its orbit around the sun, but even the
most sophisticated instruments of the day could not detect this. This was as
clear a “falsification” as one is likely to get, yet the most brilliant scientists of
that time, Galileo and Kepler, stuck by the theory. Were they behaving
“unscientifically” and thereby rendering the theory “unfalsifiable?” Histo-
rians of science suggest not—it is typical that scientific theories be retained
in the face of apparent falsifiers.
This ties in with the major theoretical criticism of Popper. The criticism
pertains to the philosophical underpinning of the demarcation criterion—
Popper’s theory of falsificationism. According to this, though the predictive
successes of a scientific theory entitle it to be retained, they have no bearing
on its epistemic value. That a theory has passed even a strict test and has
122 Chapter 5
been successful even in a precise prediction does not make the theory any
more probable than it would otherwise have been. This is a counterintuitive
claim. To take Popper’s own favorite example, one would think that Ein-
stein’s theory passing the test of the eclipse observations at least raises the
probability of its truth, something that should count in its favour if in the
future the theory were to face a challenge. Such weighing of evidence for and
against, however, is contrary to Popper’s view.
This view in turn depends on Popper’s rejection of inductive reasoning.
Having taken to heart philosophical arguments, known since the days of
David Hume, that induction cannot be defended against reasonable scepti-
cism, he insisted that science can only be rational on deductive terms. In line
with this, he sought to build a philosophy of science upon a logical asymme-
try. According to deductive logic, nothing about the truth value of a theory
can be inferred from the truth of its consequences, but something can be
inferred from the falsity of its consequences—namely, that the theory is
false. This is for many the beauty of Popper’s theory: it provides a logical
rationale for making predictive failure, rather than predictive success, the
touchstone of science.
There is, however, a problem. The logic only applies in an ideal situation
where a theory strictly entails a prediction and the falsifying observation
statement is known to be true. But by Popper’s own lights, such an ideal
situation can never arise. Scientific theories only ever entail a testable predic-
tion in conjunction with other statements, 3 and these, on Popper’s view, we
can never know to be true or even probable. Moreover, the observation
statements by which any prediction is to be evaluated are also such as we can
never know to be true or probable. This means that, strictly speaking, no
theory is falsifiable. Popper knew this, but he nevertheless thought the ideal
situation could be replicated by fiat. Scientists, he demanded, must make
methodological decisions to accept the background and observation state-
ments as true, even though they can never know that they are. Only if they do
so will the logic of falsificationism apply.
Such methodological decisions, however, are both unjustifiable and
risky. 4 The statements that scientists decide to hold true—simply in order, as
we have seen, to facilitate the logic of falsificationism—might not be true.
Accepting that they are means scientists run the risk of dismissing a true
theory as false. Does science really need to be so precarious?
The criticism is well-illustrated by a historical example (given by the
philosopher of science Hilary Putnam [1974]). When the planet Uranus was
discovered in the eighteenth century, astronomers quickly set about applying
Newton’s theory of universal gravitation to predict its orbit. But their obser-
vations did not match the predictions. On Popper’s account, then, it seems
Newton’s theory should have been rejected. That was not what happened. By
that time, astronomers were convinced that Newton’s theory was true, so if
False Negatives 123
there was a mismatch between prediction and observation, then the fault
must lie elsewhere. The most obvious candidate was an auxiliary assumption
used to deduce the prediction, the assumption that Uranus was the last planet
in our solar system. They therefore took the failed prediction as showing that
it was this, rather than Newton’s theory, that was false. Hence, they con-
cluded, there must be another planet. The French astronomer Urban Le Verri-
er and the English John Couch Adams used Newton’s theory to calculate
where this other planet might be. On the basis of these calculations, the
planet Neptune was discovered. Thus, far from taking the failed prediction as
falsifying Newton’s theory, they used it to make an important new discovery.
This illustrates the problem with Popper’s falsificationism. Newton’s the-
ory, like almost any other scientific theory, can only entail a testable predic-
tion in the light of other assumptions. In which case, if there is a predictive
failure, it is always possible that it is one of these other assumptions that is
false rather than the theory under test. Popper nevertheless holds that if
scientists are to test a theory, they need to decide to hold the auxiliary
assumptions as true. Once they make this decision, the logic of falsification-
ism dictates that a failed prediction should result in the rejection of the
theory. To put the blame of the failed prediction on the auxiliary assumption
is forbidden. It constitutes an unacceptable attempt to explain away negative
results. But, of course, it might be that the decision was wrong, and that it
really is one of the background assumptions that is to blame for the failed
prediction, as was the case with Uranus’s orbit.
Popper, aware of the problem, was in his later writings willing to relax the
methodological strictures he had earlier insisted on and allow an “appeal
procedure” against an apparent falsification. He did not, however, provide
details about how this would work. That is unsurprising, since the problem
remains that the decisions that scientists are required to make remain un-
grounded in anything other than purely methodological considerations. Pop-
per is, as his most prominent student Lakatos described him, a “methodologi-
cal conventionalist,” someone who holds that the process of science rests
ultimately on unjustifiable decisions. This, to many, does not seem a satisfac-
tory account of scientific rationality.
It leads one to wonder, moreover, about the wisdom of abandoning the
basic intuition that predictive success should count for something, epistemi-
cally speaking. The most commonsensical interpretation of our earlier exam-
ple is that the scientists were unwilling to take the anomaly of Uranus’s orbit
as a reason to reject Newton’s theory because they assigned a high degree of
credence to the theory as a result of its impressive track record—that is, its
extraordinary predictive and explanatory successes. This was why they re-
examined the background assumptions rather than blame the theory itself.
And this seems a perfectly rational way of thinking about the matter.
124 Chapter 5
It may be objected that all the above examples pertain to extra-clinical test-
ing. Freud, however, intended his theory to be scientific on clinical grounds.
So at least with regard to Freud’s conception, it matters (to a Popperian) that
his theory be amenable to falsification according to his own methods. Pop-
per, as we saw earlier, admitted that Freud’s analyses of dreams in The
Interpretation of Dreams are fundamentally correct, so he would have been
prepared to grant that numerous individual dreams can be, and have been,
correctly interpreted according to Freud’s method. The problem, however, is
whether the method of interpretation could ever be taken to yield negative
evidence. It seems all too easy to explain away an inconvenient interpreta-
tion, or non-interpretation, on the grounds that the method was not carried
through far enough, the analyst lacked the requisite skill, or the analysand
was resisting excessively. As such, the theory seems unfalsifiable if we re-
strict its evaluation to purely psychoanalytic evidence.
There is, however, reason to resist this conclusion. Freud himself was
willing to accept falsification of his claims. He did this, most noticeably,
when he admitted that repetitive traumatic dreams contradicted his wish-
fulfilment claim. This necessitated a revision of his theory, so he was clearly
treating such dreams as falsifiers. Less noticeably, Freud admitted to not
being able to interpret typical dreams using his method of free association
(1900, S.E., IV: 241). Though he claimed to nevertheless be able to interpret
these dreams symbolically, that he was prepared to infer that these dreams
were not interpretable by his normal method suggests there can be grounds
for accepting that interpretation is unable to yield the results expected of it.
This is surely correct: there will be occasions when the excuses listed
above—lack of thoroughness, lack of skill, excessive resistance—are simply
not credible. In which case, a negative result can count as a falsifier. Thus it
is reasonable to regard Freud’s theory as “falsifiable” by psychoanalytic
evidence, just as it is by non-psychoanalytic evidence.
Popper himself did not assert that Freud’s theory of dreams was formally
unfalsifiable. Rather, he argued that Freud’s reluctance to face up to falsifiers
renders the theory methodologically unfalsifiable. As we have seen, howev-
er, there is reason to be wary of this version of falsifiability, for it would
render most, if not all, major scientific theories unfalsifiable, since scientists
often vigorously defend against alleged falsifiers. The most common reason
they do so is that the theory does not strictly entail the alleged false predic-
tion. Likewise, this would be Freud’s reason.
Popper’s charge centers on the objection that anxiety dreams refute the
wish-fulfillment claim. He maintains that Freud’s only response is to promise
a “programme” of analyzing anxiety dreams to show that they are disguised
wish-fulfillments, but that he never follows through with this promise. I think
he misunderstands Freud. Freud does believe that anxiety dreams, when
128 Chapter 5
properly analyzed, turn out to be wish-fulfilling, like all dreams. In this they
are no different than other dreams that do not on the face of it seem to be
wish-fulfilling, evidence for which Freud offers numerous examples
throughout his book. This leaves the explanatory challenge of accounting for
the presence of the anxiety. But to this Freud gives an answer: The anxiety
arises due to repressed thoughts breaking through to consciousness (1900,
S.E., IV: 267; V: 580–81).
Freud also fusses about a deeper theoretical issue, to do with the affective
origin of the anxiety, an issue about which he changed his mind. His first
(1900) theory was that anxiety was libidinal excitation made unpleasurable
by repression (i.e., anxiety originates in the id). His later (1926) theory was
that the anxiety came from the ego’s reaction to its becoming aware that a
wish it forbids is making substantial progress toward fulfillment (i.e., anxiety
originates in the ego). About this theoretical issue he says, “If it were not for
the fact that our topic [the theory of dreams] is connected with the subject of
the generation of anxiety by the single factor of the liberation of the Ucs.
during sleep, I should be able to omit any discussion of anxiety-dreams and
avoid the necessity for entering in these pages into all the obscurities sur-
rounding them” (1900, S.E., V: 582; my italics)—an expression that Popper
both misquotes (he omits, tellingly, “the subject of the generation of”) and
misconstrues as the repudiation of his “programme” (1983, 166).
Popper misinterprets Freud as repudiating his program because he is con-
fusing Freud’s theoretical preoccupations with the matter of evidence. In
complaining of wanting to “omit any discussion of anxiety-dreams,” Freud is
concerned with how anxiety is generated. Popper, on the other hand, is
concerned with the prima facie contradiction that anxiety dreams present to
Freud’s wish-fulfillment claim. These concerns need to be distinguished.
Freud’s vacillation on the theoretical issue should not obscure his answer to
the evidential one. For Freud, the apparent contradiction that anxiety dreams
present is not real. First, that anxiety dreams are not wish-fulfilling on their
face is no more of a problem than any other apparently non-wish-fulfilling
dream. Freud’s claim is not that dreams are wish-fulfilling on their face, but
that they are so at least disguisedly. So anxiety dreams do not falsify his
theory in this respect. Second, though the presence of anxiety in dreams
presents an explanatory challenge, Freud’s theory does not entail that dreams
should not be accompanied by such an emotion. The explanatory challenge
can be met by reflecting on the conflict that underlies all disguised dreams,
and which also underlies neurotic symptoms: pre-dream censorship in the
first instance, and symptom-construction in the second, exist precisely in
order to prevent anxiety 7—a purpose that, alas, they can serve only imper-
fectly. So as long as one pays attention to the subtleties of the theory, one can
see that anxiety dreams are not falsifiers because the theory does not pre-
clude them. 8
False Negatives 129
Those inclined to think that this elucidation of the theory represents ex-
actly the kind of maneuver that renders a theory unscientific—and the kind
of explanatory flexibility that Popper was so suspicious of—should perhaps
be sensitive to what happens in the case of other scientific theories. They
could, for example, review what Darwin had to say in the Origin of Species,
particularly in Chapter 6, “Difficulties on Theory,” and Chapter 9, “On the
Imperfection of the Geological Record.” Darwin was prepared to elaborate
on how his theory could deal with apparent falsifiers like “organs of extreme
perfection” such as the eye—in this case, just as Freud does with anxiety
dreams, he admits the appearance of a contradiction, but then goes on to
elucidate why it is not actual. 9 Popperian critics should also perhaps reflect
on the fact that Freud was prepared to accept falsification of his theory, as he
did when he admitted that the dreams of those suffering from war neuroses
implied that the wish-fulfillment claim, as originally conceived, was in fact
false. This would be surprising if Freud were as unscrupulous in his desire to
uphold his view as many claim, and unnecessary if the theory were as flex-
ible as some suspect. Rather, as with any scientific theory, one needs to pay
careful attention to the theory in order to understand just what it does and
does not entail.
Conclusion
For the reasons outlined above, Popper is wrong that Freud’s theory of
dreams fails his criterion and that the criterion adequately demarcates be-
tween science and non-science. On the latter point, most contemporary phi-
losophers of science agree. In my view, Bayesianism and explanationism are
more promising theories of how science works than falsificationism, and
neither of these requires the demarcation Popper advocates.
Nevertheless, there remains something of value in Popper’s stance. For
though falsificationism may be too strict to apply in practice, the attitude of
seeking to falsify has much to recommend it. It can serve as a healthy correc-
tive to an insidious tendency that threatens all theoretical inquiry, that of
confirmation bias. As psychologists have shown, people have a tendency to
favor evidence that confirms a theory over evidence that disconfirms it. The
concern is that Freud, and psychoanalysts in general, in failing to adopt the
corrective attitude, are guilty of just this kind of bias. This is a legitimate
worry, independent to what one thinks of Popper’s philosophy as a whole.
Though Freud did not dodge apparent refutations and was prepared to
change his theory when necessary, the suspicion remains that his attitude
toward favorable and unfavorable evidence was not equitable. This does not
automatically disqualify the theory from being scientific, but it does have
implications for how we should regard it. First, to the extent that the attitude
prevails within psychoanalysis, it raises questions about the reliability of
130 Chapter 5
GRÜNBAUM
something similar—I will grant the evidence of the report in order to evalu-
ate Grünbaum’s criticism of Freud’s reasoning. My aim is to examine
Grünbaum’s critique, not defend Freud’s conclusion.
What is relevant about this case in the present context is what Freud
inferred from the fact that symptoms were removed by this technique. This is
where, according to Grünbaum, the fallacy occurs. On his account, Freud
inferred from the fact of the removal of the symptom on application of the
technique the following hypothesis:
Grünbaum observes that the etiologic hypothesis deductively entails the ther-
apeutic hypothesis. He then interprets Freud and Breuer as having reasoned
from this fact to the acceptance of the etiologic hypothesis. This, Grünbaum
points out, would be a crude hypothetico-deductive inference. So crude, in-
deed, that it is fallacious. The mere fact that a hypothesis deductively entails
evidence is not sufficient warrant for accepting that hypothesis. 12
His alleged avowals aside, I doubt Freud would have accepted this inter-
pretation of his reasoning. In any case, as I observed earlier, whether he
would or not is beside the point. What matters is how he was actually reason-
ing, not how he thought he was. To gain a sense of this we need a broader
view of the evidence.
According to Grünbaum the evidence for the therapeutic hypothesis was
the following:
that Freud and Breuer thought incompatible with the placebo hypothesis. But
as Grünbaum points out, it could well be that the doctor’s own concerns with
a particular symptom could have been transmitted to the patient during hyp-
nosis, thus resulting in the specificity of the result in question. Thus particu-
larity is not a good enough reason for dismissing the placebo hypothesis.
So far, I think, we must concur with Grünbaum. But what I think he
misses in his interpretation are a number of other facts that are not in accord
with the placebo hypothesis. One of these may have been touched on by
Grünbaum himself, though in an erroneous way. Freud writes of his method
of dream interpretation that it “is identical with the procedure by which we
resolve hysterical symptoms; and there the correctness of our method is
warranted by the coincident emergence and disappearance of the symptoms”
(1900, S.E., V: 528). Grünbaum comments on this that he believes that Freud
made a “confusing slip of the pen” in writing emergence (German: Auftau-
chen) rather than resolution (German: Auflösung). This may be, though there
is an alternative explanation, that Freud made no slip here, but by emergence
meant a more emphatic emergence of the symptom. For as Breuer writes of
Anna O., there was “a feature that was always observable when a symptom
was being ‘talked away’: the particular symptom emerged with greater force
while she was discussing it” (1893–1895, S.E., II: 37). In any case, what I
want to highlight is not Grünbaum’s potentially erroneous correction of
Freud, but the fact that Breuer speaks of. That the symptom emerged more
forcefully as it was being “talked away” is a datum that calls for explanation.
Related to this is a fact that Grünbaum is very much aware of. According
to both Breuer and Freud’s reported experience, recall of the forgotten trau-
matic memory that preceded symptom removal was usually accompanied by
a strong expression of affect—such as Anna O.’s expression of disgust and
anger when she was recalling the scene with the dog. That there should be a
spontaneous release of emotion occurring simultaneously with the emer-
gence of the memory is another datum calling for explanation. 13 This should
be considered in conjunction with two other facts. One is that the traumatic
event recalled always involved the patient suppressing a strong emotional
reaction. The other is that the memory that eventually emerged to conscious-
ness was always surprisingly “fresh”—that is, vivid and detailed
(1893–1895, S.E., II: 8–9).
Another observation that Breuer makes, and which Freud endorses with
regard to his own cases, is this: “We often noticed that her [Anna O.’s] dread
of a memory . . . inhibited its emergence, and this had to be brought about
forcibly by the patient or physician” (1893–1895, S.E., II: 37). There was, in
other words, resistance to the emergence of the traumatic memory. This
point is most important. Freud writes, “It was on this idea of resistance, then,
that I based my view of the course of psychical events in hysteria” (1910,
False Negatives 135
S.E., XI: 23; 1893–1895, S.E., II: 268). That resistance was crucial to Freud’s
reasoning is something that is omitted in Grünbaum’s account.
The above facts are, I believe, pertinent to the conclusion drawn by Freud,
despite their not being mentioned by Grünbaum as part of the evidence base
of Freud’s inference. One reason why they are significant is that they can
help adjudicate between the hypotheses that, on Grünbaum’s account, are at
stake. The key question is this: Which of the hypotheses better explains the
facts, the repression hypothesis or the placebo hypothesis?
The repression hypothesis, in a little more detail than Grünbaum gives, is
as follows. By suppressing her emotional reaction to the traumatic event, the
patient had failed to sufficiently “abreact” the trauma. The memory of this
event therefore remained charged with strong negative affect. Due to this, it
was repressed from consciousness. In so being, the memory would “act like a
foreign body which long after its entry must continue to be regarded as an
agent still at work” (ibid., 6). This “agent’s work” involved causing the
symptom. Only when the patient was able to bring the memory back to
consciousness and give full expression to the affect could she de-energize the
memory and thereby bring an end to the symptom.
The facts, recall, are as follows: attempts to uncover the traumatic memo-
ry were typically accompanied by resistance; as the memory would begin to
surface, the symptom would often re-emerge more forcefully; the recovery of
the memory would be accompanied by a release of affect; the recovered
memory was surprisingly vivid and detailed; it was of an event that had
closely preceded the first appearance of the symptom and was thematically
related to the symptom; as soon as this memory was recovered and the affect
released, the symptom would disappear. Freud’s repression hypothesis ex-
plains these facts well. One can understand the resistance as a manifestation
of repression. One can get a sense of why the symptom would re-emerge
more forcefully as the memory comes closer to consciousness, in that, if the
memory was what was causing the symptom, then the expression of the
memory could have a bearing on the expression of the symptom. One can
understand why there should be a release of affect, in that this was affect that
had been suppressed when the originating event had occurred and therefore
in need of abreaction. One can understand why the recovered memory should
be vivid and detailed in that it had remained energized despite its being
unconscious (and hence, one may assume, requiring an active force to keep it
unconscious). One can understand, if the repressed memory were a cause,
why the symptom would have first emerged closely after the remembered
event had occurred and why it should be thematically related to the memory.
One can understand, if the repressed memory were a cause, why the lifting of
the repression would immediately lead to the disappearance of the symptom.
Freud’s hypothesis explains the facts by indicating (though not yet fully
specifying) plausible causal mechanisms, it explains all the facts, and is a
136 Chapter 5
This brings me to what is, I believe, one of the most telling paragraphs in
Grünbaum’s recent writing. Here it is in full:
Yet some apologists offer a facile excuse for the fallacious H-D confirmation
of a causal hypothesis. We are told that the hypothesis is warranted by an
“inference to the best explanation” (Harman 1965). But in a careful new study,
Salmon (2001) has argued that “the characterization of nondemonstrative in-
ference as inference to the best explanation serves to muddy the waters . . . by
fostering confusion” between two sorts of why-questions that Hempel had
distinguished: Explanation-seeking questions as to why something is the case,
and confirmation-seeking questions as to why a hypothesis is credible. Thus, a
hypothesis that is pseudoconfirmed by some data cannot be warranted qua
being “the only [explanatory] game in town.” Alas, “best explanation”-sanc-
tion was claimed for psychological etiologies to explain and treat the destruc-
tive behavior of sociopaths to no avail for years. (2005, 124)
False Negatives 137
CONCLUSION
As with Copernicus’s and Darwin’s theories at their time, many of our con-
temporaries see psychoanalysis as beyond the pale in its claims to scientific-
ity. They see it as a pseudoscience. However, the arguments for such a
conclusion are questionable. The term “pseudoscience” connotes more than
just being a non-science (or a non-science whose proponents claim it to be a
science). It connotes that the target is in some sense antithetical to science;
140 Chapter 5
NOTES
False Positives
143
144 Chapter 6
trust in firm scientific knowledge claims. They do so because they see sci-
ence as a reliable epistemic enterprise—that is, an enterprise that tends to
produce truth or approximate truth. An affirmative answer to the question of
scientificity would qualify psychoanalysis as worthy of this trust, while a
negative answer would mean it has no automatic right to it. Psychoanalysis
therefore needs to earn the credibility that comes with being a science by
showing itself to be epistemically reliable. This demands more than just
employing a legitimate form of reasoning.
I think there are two approaches to establishing psychoanalysis as epis-
temically reliable. One is by showing it to be based on rigorous epistemic
practices. A rigorous practice is one that incorporates procedures to counter-
act sufficiently the errors and biases that it would otherwise be prey to. The
other approach is by showing it to be based on a reliable technique. This
differs from the first approach in that the technique can be considered purely
instrumentally: its nature need not matter, what does is that it works. I will
consider each of these approaches in turn to see whether it is reasonable to
judge psychoanalysis, and in particular dream interpretation, as reliable ac-
cording to it.
psychoanalytic claims are generated and justified through the study of indi-
vidual cases, usually of an encounter between analyst and analysand. This is
true also of Freud’s claims about dreams. Their empirical grounding lies in
the individual analyses of dreams—including his own—that Freud reports in
his writings. As many commentators have observed, there are numerous
sources of error and bias within this process (Wallerstein and Sampson 1971;
Midgley 2006; Kächele et al. 2009). These fall into two broad categories:
problems with the data and problems with inferences from the data.
My purpose in the forthcoming discussion is not to offer a comprehensive
review, but to present a sample of problems and solutions that shows the
nature of the challenge faced if psychoanalysis is to be rigorous in its epis-
temic practices.
The inferences that psychoanalysts make from the data are inferences to the
best explanation, which as I have argued at length, are valid and ubiquitous
in science. The problem with such inferences in psychoanalysis, however, is
the ease with which one can be misled into making a weak inference to the
best explanation.
One well-known cause of weak inferences is the seductiveness of narra-
tive. David Tuckett expresses the problem as follows:
There is the possibility that a good, well-told, and coherent story creates the
risk of seduction, which in the context of communication to others can be
summed up thus: the more a narrative is intellectually, emotionally and aes-
thetically satisfying, the better it incorporates clinical events into rich and
sophisticated patterns, the less space is left to the audience to notice alternative
patterns and to elaborate alternative narratives. (Tuckett 1993, 1183; cited in
Midgley 2006)
(1) The study should clearly and prominently state the hypothesis being tested
in the case;
False Positives 147
The problem of weak inferences would not be as great if the inferences could
be subjected to intersubjective scrutiny. For though individual psychoana-
lysts may go astray, appropriate critical evaluation of their inferences by
others would serve as a corrective. This solution, however, runs up against
another problem: the data upon which dream interpretations are based are not
public. Since the encounter between analyst and analysand from which the
data are generated is private, others cannot observe the data for themselves.
This thereby limits the ability of outside observers to make reliable judg-
ments about the conclusions reached or offer effective criticism of them.
148 Chapter 6
tion. She may, for instance, have guided the patient toward generating data
more favorable to a preferred interpretation. Suppose, for example, in the
case of the supper-party dream, Freud, knowing of his patient’s jealousy of
her skinny friend, already had an interpretation in mind while his patient was
associating. Through subtle—and unreported—cues, he may have guided her
toward that interpretation. Without the possibility of checking the data at
source and the manner in which they have been generated, it is difficult to see
how one can eliminate the possibility that the analyst has inordinately influ-
enced the analysand. 5
The problem of private data is highlighted by a contrast with Darwin’s
theory. As in psychoanalysis, Darwin used inference to the best explanation
to establish his claims. However, the data he was explaining were publicly
available. To use an earlier example, Darwin’s theory of common descent
well explains the presence in many organisms of vestigial organs. In this
case, the data—the presence of vestigial organs in various organisms—are
readily available. They can be checked and scrutinized by others. This is not
so, however, with psychoanalytic data. For example, Freud’s inference to the
conclusion that repression was a cause of Anna O.’s hysterical symptoms is
valid, as I argued in the previous chapter, but the acceptability of the conclu-
sion is conditional on the evidence being just as reported, and that may be
reasonably doubted. The data in Freud’s theory are not of the same status as
the data in Darwin’s theory. 6
These problems are not insurmountable. One solution is to use video or
audio recordings. This would help overcome the problems of misreporting,
selection bias, and guidance by the analyst, as these could be detected by a
careful analysis of the recording or its transcript. It also has another obvious
advantage in that much more of the data can be made available to others,
including, with video recordings, subtle behavioral cues.
It should be noted that a careful analysis of recordings may be sufficient
to counteract the problem of suggestion. The problem of suggestion was one
that Grünbaum was much concerned with, regarding it as an important rea-
son for rejecting psychoanalytic claims (1984, 1993). He thought that the
way to counteract it was through the use of appropriate controls. This is not, I
think, the only way. Close and skilled critical scrutiny of a recording could
offer evidence for or against the suggestion hypothesis. As I argued in Chap-
ter 4, there surely are circumstances in which the influence of suggestion is
implausible—for example, when there are no identifiable suggestive cues
from the analyst. So though such critical analysis of recordings is not fool-
proof, it can go some way to addressing the problem, and perhaps far enough
to establish the reliability of the data.
Despite the misgivings of an older generation of analysts, 7 it is clear that
increasing use of video and audio recordings is important to putting the case
study method on a secure scientific footing. Psychoanalysts have used such
150 Chapter 6
Conclusion
You will perhaps be surprised to hear that in Europe we have heard a large
number of judgements on psychoanalysis from people who know nothing of
this technique and do not employ it; and who go on to demand with apparent
scorn that we shall prove to them the correctness of our findings. Among these
adversaries there are no doubt some to whom a scientific mode of thought is
not as a rule alien, who, for instance, would not reject the results of a micro-
scope examination because it could not be confirmed on the anatomical prep-
aration with the naked eye, but who would first form a judgement on the
matter themselves with the help of a microscope. (1910, S.E., XI: 39)
But you will always find it difficult to persuade other people of the correctness
of your ideas. Others do not have your vision & are therefore not in a position
to form their own judgment. Psychoanalysis is [neither] a science nor a craft; it
cannot be learned in the usual sense. It is an art, which must be innate and can
only be developed. (28 November 1905; in Marinelli and Mayer 2003, 173)
Hooke traced the cause of some of the disagreements [in observation] to dif-
ferent kinds of illumination. He pointed out that the eye of a fly appears like a
lattice covered with holes in one kind of light . . . like a surface covered with
False Positives 153
cones in another and in yet another light like a surface covered with pyramids.
Hooke proceeded to make practical interventions designed to clear up the
problem. He endeavoured to eliminate spurious information arising from daz-
zle and complicated reflections by illuminating specimens uniformly. He did
this by using for illumination the light of a candle diffused through a solution
of brine. He also illuminated his specimens from various directions to deter-
mine which features remained invariant under such changes. Some of the
insects needed to be thoroughly intoxicated with brandy to render them both
motionless and undamaged. (1999, 20)
issue was whether evidence from the telescope is permissible in science. That
turned on the question of whether the telescope was reliable.
As we know, the telescope did come to be seen as reliable and has been
an indispensable instrument of astronomical research ever since. We may
therefore ask what, if any, are the differences between the telescope and
psychoanalytic technique that might be relevant to instrumental reliability.
Was it that the observations made by the telescope were particularly clear?
No. The lenses used were of poor quality and it was difficult to see clear
images. As a result, a number of opponents complained that the telescope
generated illusions. Giovanni Magini thought he saw three suns through it
(Brown 1985, 489). Jacob Christmann claimed that, depending on the tele-
scope used, he saw Saturn with “two, three, or four companion stars” and that
“Jupiter appeared in three distinct scintillations and exhibited two shimmer-
ing diameters” (van Helden 1974, 52). Martin Horky observed that “Below,
it works wonderfully; in the sky it deceives one, as some fixed stars are seen
double” (Geymonat 1965, 44–45). Was it that it could be used with ease,
requiring no particular skill? No. “Galileo even suggested that . . . the gentle-
men of the Medici court should, perhaps, not attempt to observe Jupiter’s
moons with the telescope until he could be present himself” (Brown 1985,
489). Was it that the observation, once made, could not be contested? No.
The group of Jesuit astronomers, whose confirmation of Galileo’s observa-
tions had much influence on eventual acceptance of those observations, in
their otherwise favorable report, cautioned that “there were some points on
which there was not yet sufficient evidence to accept Galileo’s interpreta-
tions of his observations” (ibid., 489–90). Was it that the theory on which it
rested was well-understood? No. Knowledge of optics at the time may have
been too incomplete to justify its reliability. In any case, even if Galileo had a
good enough grasp of optics, many of those whom he needed to persuade did
not.
Nevertheless, the reliability of the telescope was soon accepted, even by
many of those opposed to Galileo. Van Helden (1974, 51) says, “The time it
took Galileo to convince all reasonable men was astonishingly short.” What
was the basis of this? First, its reliability in terrestrial observation could not
be reasonably disputed, as it was easily checked. One could observe a known
object from afar and see the image through the telescope as that of the object
magnified. The disputable issue was whether the reliability of the telescope
in terrestrial observation could be extended to astronomical observation.
To a large extent, objections to this extension were overcome by more
careful and patient observation. But the issue soon came to focus on a specif-
ic objection. This was that the magnification of stars and planets was not of
the same degree as other objects. “Stars, whether fixed or wandering, appear
not to be enlarged by the telescope in the same proportion as that in which it
magnifies other objects. . . . [A] telescope which is sufficiently powerful to
False Positives 155
enlarge other objects a hundredfold is scarcely able to enlarge the stars four
or five times” (Galileo 1957, 46). To this objection, Galileo responded with a
counterargument. The problem was not with the telescope, but with naked
eye observation. “Bright distant objects” seen at night seem larger to the
naked eye because of the phenomenon of irradiation, the presence of a
“fringe of sparkling rays” (ibid.). The telescope removes this irradiation, and
thus the image of the stars is smaller than expected. To make this persuasive,
he pointed out that one could observe stars and planets without irradiation
with the naked eye, in daylight and through a thin cloud. For example, Venus
in daylight seems “so small that it takes sharp eyesight to see it, though in the
following night it appears like a great torch” (ibid., 131). He also listed a
number of other ways of seeing stars without irradiation, for example, by
viewing them through a black veil, a gap between the fingers, or a pinhole in
a card (Chalmers 1999, 165). In this way, Galileo mounted a powerful case
for the reliability of telescopic as against naked eye observation.
The examples of the telescope and the microscope suggest a strategy for the
acceptance of the reliability of an instrument or technique. This involves
three conditions: (1) a basic check, (2) a smooth extension of this check, and
(3) consistency of outcome henceforth. The basic check consists of ascertain-
ing that the technique does work in at least a number of cases. The smooth
extension of this check consists in (a) the blurring of boundaries between the
basic cases (i.e., those of the basic check) and other cases, and (b) convincing
arguments against any objections to the claim that the technique holds in all
cases in the same way as it holds in the basic ones. Consistency in outcome
consists in other “credible persons”—experts in the field who are not already
biased in favor of the technique—being able, upon correct application of the
technique, to evince the same outcome as has been proclaimed by advocates.
It need not be the case that the outcome is found on every application, but
that it is found on the majority of careful applications by fair and reasonable
people.
The telescope and microscope both meet these conditions.
(1) We have already seen how the basic check works for the telescope. It
is equally trivial for the microscope. One can see structures in the microscope
that are clearly enlarged images of structures that one can see with the naked
eye. That is enough to establish that the technique can work.
(2) The smoothness of the extension for the telescope was, as we have
seen, a major issue, requiring a strong counterargument to a specific objec-
tion, but the blurring of boundaries between basic cases (terrestrial observa-
tions) and other cases (celestial observations) was more straightforward. One
aspect of this blurring of the boundaries was that celestial objects seen with
the naked eye could be seen, enlarged, with the telescope. Another aspect
was that celestial objects seen only by observers with keen eyesight could be
156 Chapter 6
A little girl nineteen months old had been kept without food all day because
she had had an attack of vomiting in the morning; her nurse declared that she
had been upset by eating strawberries. During the night after this day of starva-
tion she was heard saying her own name in her sleep and adding: ‘Swawbew-
wies, wild stwawbewwies, omblet, pudden!’ She was thus dreaming of eating a
meal, and she laid special stress in her menu on the particular delicacy of
which, as she had reason to expect, she would only be allowed scanty quan-
tities in the near future. (1901, S.E., V: 643–44)
CONCLUSION
Until recently I could only apprehend the speculative power of your train of
thought, together with its enormous influence on the Weltanschauung of the
present era, without being in a position to form a definite opinion about the
amount of truth it contains. Not long ago, however, I had the opportunity of
hearing about a few instances, not very important in themselves, which in my
judgment exclude any other interpretation than that provided by the theory of
repression. (In Jones 1957, 217)
Though a third party need not take Einstein’s assertions as grounds for be-
lieving in the said theory, Einstein himself may be personally justified in his
conviction.
A second reason for optimism is that psychoanalysis might in the future
become scientific. The problems that prevent it from being scientific as it
stands are not insurmountable. In the first half of the chapter I described
some of the procedures that psychoanalysis could adopt to bring it closer to
being a science. The second half of the chapter suggests another approach,
that of establishing psychoanalytic method as an effective technique. This
can happen in a number of ways. One is by building on the case that already
exists for regarding psychoanalytic method as an effective technique. In par-
ticular, if a suitable basic check along the lines suggested can be carried out
using scientifically acceptable procedures, then this can provide the basis for
establishing the technique as instrumentally reliable (I explore one such way
in Chapter 8). Alternatively, technological developments in neuroscience
may provide a different way of checking the effectiveness of the technique.
A third reason for optimism is that, though Freud’s theory of dreams is
not scientifically acceptable on Freud’s terms, it might be so on other terms.
The theory is not, contra-Popper, inherently unscientific. It is a candidate for
scientific acceptance. So even if the means by which Freud sought to justify
it are deemed unacceptable to science, it may nevertheless gain acceptance
False Positives 161
NOTES
Brainwaves
Can a once dead theory flourish again? After a presentation by the dream
researcher J. Allan Hobson at the 1976 annual meeting of the American
Psychiatric Association, a vote was taken as to whether Freud’s theory of
dreams was scientifically tenable. Despite the still powerful presence of
psychoanalysts among the members, the vote went overwhelmingly against
Freud (Solms and Turnbull 2002, 190). In the years to come, it was Hobson’s
theory that dominated the headlines, while Freud’s, apparently disproven by
science, fell by the wayside. Fast forward thirty years, and a remarkable
reversal seems to have occurred. In 2006, Hobson had a formal debate with
the neuroscientist and psychoanalyst Mark Solms on the question “Should
Freud’s Dream Theory Be Abandoned?” The debate took place at the April
2006 “Toward a Science of Consciousness” conference in Tucson, Arizona,
attended by psychologists, psychiatrists, neuroscientists, and philosophers. A
vote was taken among the audience of approximately 200 academics, and the
outcome was about two to one in favor of Solms’s position that Freud’s
theory should not be abandoned.
Perhaps we should not read too much into such events. As we all know,
the vagaries of debate can lead to outcomes that do not correspond to the
shape of the evidence. In an article in the UK’s Guardian newspaper, the
prominent psychologist and writer Sue Blackmore, who was present at the
2006 debate, lamented, “A century old, [Freud’s] theory of dreams, which
ought to have slipped quietly into oblivion, still provokes even serious scien-
tists into heated debate” (2006). She concluded with relief, “Happily, unlike
politicians, scientists don't take votes seriously.” Yet the symbolic value of
such votes can be significant, not in determining people’s views, but in
reflecting them. Freud’s theory was for a time, if not dead, then on life
support. But that is no longer so.
165
166 Chapter 7
The dream researcher, Claudio Colace, whom we shall encounter later on,
provides a good overview of developments in the science of sleep and
dreams (2010, xiv–xvii). He analyzes the history of modern dream research
into four trends. The first begins with the discovery of rapid-eye-movement
(REM) sleep by the sleep researchers Aserinsky and Kleitman in the early
1950s and continues to the end of the 1960s. It involves the attempt by
researchers to integrate Freud’s theory with the REM findings. The second
trend begins in the early 1970s and continues in part until today. It involves a
turning away from Freud, and from “psychological-motivational” theories of
dreaming in general, toward purely physiological accounts. Its apotheosis is
Hobson and McCarley’s activation-synthesis model of dreaming, which we
shall consider at some length. The third trend began near the end of the
1970s. The cognitive revolution that had been taking place for some time in
psychology began to manifest itself in dream research. The revolutionaries
had little time for motivational theories of dreaming, and purported instead to
explain dreams in cognitive terms. This trend continues to the present, and is
exemplified by the work of William Domhoff, whose arguments against
Freud we shall also consider at length. The fourth trend, which began in the
1990s, saw a turn back to Freud. It received its main impetus from the new
technologies of brain imaging, which led to fresh interest in affect and moti-
vation. Jaak Panksepp’s Affective Neuroscience: The Foundation of Human
and Animal Emotions (1998) encapsulated the emerging view. It was in this
context that Mark Solms created neuro-psychoanalysis, with the ultimate aim
of realizing Freud’s dream of placing psychoanalysis on a firm neuroscientif-
ic footing. This fourth trend, which is continuing to gain momentum, has led
to renewed emphasis on the motivational aspects of dreaming—and the re-
suscitation of serious scientific interest in Freud’s theory.
It is against this backdrop that we address the question of this part of the
book, of whether Freud’s theory is still relevant in the light of scientific
research on dreams. The outcome of the Arizona debate and the historical
sketch I gave above suggest it is, but there remain many prominent critics
who, like Sue Blackmore, are exasperated by the fact that the question is still
even being posed. These critics have, in the last few decades, been the more
vociferous, and have fostered the impression among lay audiences that
Freud’s theory has been disproven by science. For instance, Hobson, who has
had much media attention, began his opening speech in the Arizona debate
with the assertion, “Freud was fifty percent right and a hundred percent
wrong.” This is in keeping with the tone of his public pronouncements on the
subject in general, as exemplified by his popular science book, Dreaming: A
Very Short Introduction (2005), in which he has a section titled, “What
caused the failure of psychoanalytic dream theory?” Domhoff is similarly
dismissive of Freud’s theory in his book, The Scientific Study of Dreams
(2003), the arguments of which he reprises in a paper titled, “Why Did
Brainwaves 167
HOBSON’S CHOICE
Since Kleitman and Aserinsky’s discovery of REM sleep in the early 1950s,
sleep researchers have added considerably to the picture. REM is a highly
distinct phase of sleep, characterized by levels of brain activity similar to
those of an awake person. It occurs several times during sleep, interspersed
with a phase of low brain activation, called non-REM (NREM) sleep. Most
relevant to our concerns, REM is highly correlated with dreaming. People
report dreams considerably more often when woken from REM sleep than
they do when woken from NREM sleep. 1 Also, REM dreams tend to be
longer and more authentically dream-like—more vivid, lively, and emo-
tive—than NREM dreams (Hobson et al. 2003).
In a now famous paper, published in the American Journal of Psychiatry
in 1977, J. Allan Hobson and Robert McCarley presented a hypothesis of
168 Chapter 7
dreaming that drew heavily on research on REM sleep. The main thrust of
this hypothesis was that much of dreaming can be explained by the physiolo-
gy of this sleep phase. In particular, since evidence points strongly to the
brainstem as the trigger for REM sleep, they inferred that this was therefore
the driver of dreams. They developed these ideas into the hypothesis that
dreams are the result of brainstem-generated activation of the brain and a
subsequent synthesis into a dream narrative of the random stimuli thereby
produced. Thus they sought to explain many of the characteristics of dream-
ing, such as bizarreness and dream forgetting, on this physiological basis, as
opposed to the psychological explanations that Freud had proposed.
Hobson and McCarley’s theory has been seen by many, not least the authors,
as superseding Freud’s theory of dreams. According to them, their theory
“contrast[s] sharply with many tenets of the dream theory provided by
psychoanalysis” (1977, 1346). Later, I shall question just how sharp this
contrast really is. For now, I will outline the main differences they discuss.
There are three. The first has to do with what drives the dream process.
According to Freud’s theory, it is impulses seeking expression. On their
theory, it is brainstem activation. The second has to do with the specific
stimuli for the dream-content. According to Freud’s theory, these are pre-
dominantly thoughts. On their theory, they are the direct physiological conse-
quences of brainstem activation, such as eye movements and random activa-
tion of the visual system. The third area is the process of synthesis. Accord-
ing to Freud’s theory, the dream is formed by the dream-work, which primar-
ily serves the purpose of disguise. On their theory, the dream is formed by a
constructive process—the brain’s attempt to make sense of the random stim-
uli generated by brainstem activation. A corollary of this last contrast is that
the bizarreness of dreams has a different explanation from Freud’s. It is not
due to disguise, but to limitations on the degree of sense that can be made
from random material generated during REM sleep.
If Freud were alive to read of Hobson and McCarley’s theory, he may
have a feeling of déjà vu, for it is, in part, a throwback to many nineteenth-
century scientific accounts, in which dreams were viewed as side effects of
brain physiology (1900, S.E., IV: 76–82). It is perhaps a little more sophisti-
cated than such accounts, in that it considers the contents of dreams to be
shaped by psychological factors as the brain attempts to make sense of the
physiological stimuli. But its main claim to superiority is that, unlike earlier
such accounts, it is firmly based on evidence. Herein lies its challenge to
Freud’s theory: the authors think the evidence significantly favors their theo-
ry over Freud’s in each of the above three areas of difference.
Brainwaves 169
They refer in their paper to several lines of evidence. Perhaps the most
crucial with respect to their opposition to Freud’s theory comes from studies
on cats. The sleep researcher Jouvet demonstrated that lesions of a certain
part of the brainstem, the pontine reticular formation, “prevent the occur-
rence of [REM] sleep for several weeks in cats” (1977, 1338), thereby pro-
viding evidence that this region is necessary for the normal occurrence of
REM sleep. He also demonstrated that lesions in other parts of the brain,
including the forebrain, do not have significant effects on REM, thereby
providing evidence that the activity of the pontine reticular formation is also
sufficient for this sleep phase. From such evidence Hobson and McCarley
draw the provocative conclusion that the facts “completely eliminate any
possible contribution of ideas (or their neural substrate) to the primary driv-
ing force of the dream process” (ibid.). If true, this would knock out Freud’s
theory, as its central theme is that ideas are involved in driving dreams.
Hobson and McCarley’s claim, however, rests on a conflation of REM
sleep with dreaming. If REM sleep and dreaming are distinct, then evidence
that pontine activation is sufficient for the former is not evidence that it is
sufficient for the latter. Thus if their argument against Freud is to go through,
they need to show that treating REM sleep and dreaming synonymously, as
they clearly do here, is justified.
The authors’ inclination to equate REM sleep with dreaming has much to
do with the strong correlation between REM and dreaming. But they also
make additional arguments. These have to do with apparent “isomorphisms”
between the physiological characteristics of REM sleep and dreams. For
example, during REM sleep the visual sensorimotor regions of the brain are
particularly highly activated, and this corresponds to the fact that dreams are
highly visual and contain much action. The authors interpret this correspon-
dence as evidence in favor of their hypothesis.
More compellingly, they advance claims of specific isomorphisms in the
cases of chase and flying dreams. Such dreams are common. Many of us are
all too familiar with “classic” chase dreams—with the increasing panic as we
attempt to elude a pursuer rapidly gaining ground and the subsequent relief
of finding out it was but a dream. The authors consider that such dreams
correspond well with “the feeling of constrained motor action” consequent
on the fact that, though the motor cortex is activated during REM sleep,
motor behavior is inhibited. As they put it, the dream experience of finding it
difficult to flee from one’s chaser “is so universal and the feeling of con-
strained motor action so impressive as to make its physiological basis in the
descending inhibition of motor neurons seem to us inescapable” (ibid.,
1339). Many of us are also familiar with flying dreams—with the exhilara-
tion of gliding unaided through the air and the subsequent disappointment of
finding out it was but a dream. The authors consider that such dreams are
adequately explained by the “central, automatic activation during sleep of the
170 Chapter 7
vestibular system”—that is, the part of the brain responsible for information
about head and eye position and posture. They appear to believe that the
mere plausibility of this hypothesis is sufficient to undermine the traditional
Freudian account. “In view of this reasonable and direct explanation, it seems
gratuitous to “interpret” the sensual flying dream as sexual” (ibid.). With
these claims, they take themselves not only to have undermined Freud’s
psychological interpretations of such dreams, but also to have provided evi-
dence for treating REM sleep and dreaming as synonymous.
Another argument for their view appeals to two items of evidence. The
first is that studies suggest that the eye movements typical of REM sleep
provide information that finds its way into the content of a dream. The
second is that the onset of these eye movements follows brainstem activation
but precedes activation of any cortical region—that is, the areas of the brain
associated with ideas. In other words, brainstem activation occurs first, then
the eye movements of REM sleep, and then activation of the cortex. Since, as
the first study indicates, eye movements are directly linked to dream-content,
these results seem to suggest that the dream process begins before ideas are
invoked. Thus the authors take these studies as supporting their conclusion
that it is brainstem activation rather than ideas that is the driving force of
dreams.
The above evidence is intended to support both the claim that pontine
activation drives dreams and the claim that pontine-generated physiological
events are the main specific stimuli for dream-content. An additional argu-
ment connects these claims with the authors’ third one—that a dream is
formed by the brain’s attempt to make sense of the random stimuli generated
by pontine activation. This is that activation-synthesis adequately explains
certain formal characteristics of dreams, in particular bizarreness and “dream
amnesia” (the difficulty in recalling dreams). If dreams are the product of the
brain’s best effort to make sense of random stimuli under the special condi-
tions of the REM state, then one can understand why they turn out not to
make complete sense. In particular, the authors speculate that specific fea-
tures of dream bizarreness, such as “scene shifts, time compression, personal
condensations, splitting, and symbol formation” may be isomorphic to
physiological features of the brain during REM. They also speculate that
dream amnesia may be explained by REM brain physiology. Again, the
authors think the mere plausibility of their purported explanations supports
their theory and undermines Freud’s.
information that finds its way into the dream-content. Supposing this to be
correct, and supposing Hobson and McCarley’s argument that the onset of
eye movements precedes cortical activation also to be correct, can this be
good evidence that pontine activation is the driver of dreams? I think not. It
would be good evidence that eye movements are stimuli for dreams, but not
that pontine activation drives dreams. It is perfectly possible that eye move-
ments begin before a dream occurs, that a dream is brought about by separate
means, and that subsequently eye movement information is incorporated into
the dream-content. The point applies generally to isomorphisms. If they
could be established, the most that such isomorphisms would show is that
pontine activation provides stimuli for dreams, not that it drives dreams.
This is further illustrated by the example of classic chase dreams. Recall
that Hobson and McCarley think that these can be explained by “the feeling
of constrained motor action” consequent on the fact that, though the motor
cortex is activated, motor behavior is inhibited. However, this pattern of
activation and inhibition is ordinarily always present in REM sleep, yet clas-
sic chase dreams are relatively rare. At least, the authors have failed to
provide evidence that they occur anywhere near as frequently as the feeling
that is purported to explain them. Thus the explanation, even if true, would,
as it stands, be almost certainly insufficient. Such isomorphisms therefore do
not support the claim that pontine activation drives dreams.
Hobson and McCarley might counter that the facts that they cite, the
strong correlation of REM with dreaming and the apparent isomorphisms,
are evidence insofar as they are well-explained by the hypothesis that pontine
activation drives dreams. In my view, however, the explanatory value of the
purported facts is weak. First, there are problems with the purported facts.
Many of the alleged isomorphisms have not yet been adequately established,
and the correlation, if it is to support the explanation, would need to be near-
perfect, but it is not. Second, such facts, even if they were unproblematic, do
not serve to discriminate between the explanations, for precisely the reasons I
gave above. A version of Freudian theory that allows that a high degree of
brain activation is a precondition for dreaming and allows, as the classical
version does, that physiological events can be stimuli for dreams has no
problems accounting for the facts. Admittedly, this explanation is perhaps
not as simple as Hobson and McCarley’s, but that, on its own, is not enough.
The authors, I think, have greatly exaggerated the difference in the quality of
the explanations.
Though isomorphisms, if established, would not support the claim that
pontine activation drives dreams, they would support the claim that it pro-
vides stimuli for dreams. However, this falls short of what the authors need
to distinguish their theory from Freud’s. For Freud’s theory is compatible
with the claim that physiological events are stimuli for dreams. This is shown
in Freud’s discussion of the somatic sources of dreams in The Interpretation
Brainwaves 173
bol formation are part of the dream-work that Freud describes—indeed these
terms are almost certainly taken from him. Thus we have a claim about a
process that is admittedly mysterious, and for which those parts of it that are
described are distinctly Freudian. This is clearly not a promising area for
contrasting their theory to Freud’s.
It should also be noted that Freud does not preclude a “constructive syn-
thesis” of dream stimuli. As observed earlier, he allows that external sources
can find their way into dreams. A Freudian process of incorporation can
reasonably be described as a “constructive synthetic” process, much in line
with what Hobson and McCarley mean by this term. In short, there is signifi-
cant overlap between the two theories in this area.
The principal reason the authors’ think their “synthesis” is different to
Freud’s dream process is that they deny a central role to censorship and
disguise. It should be noted that Freud himself did not claim that all aspects
of the dream-work were due to censorship. The only operation entirely due to
it is displacement. In particular, he allowed that much symbol formation and
dramatization were due simply to the difficulty of representing abstract con-
cepts. Censorship for Freud is a pressure that predominantly leads to an
intensification of already existing processes. As such, the difference between
Freud and Hobson and McCarley seems to be mainly quantitative. While
Freud supposes that the censor provides a constant and strong pressure on the
process, the other authors deny this, while allowing that at times there might
be some such pressure.
What is the evidence they have for their view? It seems the best they can
offer is that their account provides a better explanation of dream bizarreness
than Freud’s. Evaluating their account is complicated by the fact that this
account has changed substantially over the years. Originally, they held that it
is the difficulties of finding a best fit for the random and changing physiolog-
ical activation they think constitutes the primary stimuli that is behind the
strangeness of dreams. More recently, they defend the view that such
strangeness is due to defects in the synthetic process, in particular defects
caused by the special physiology of REM sleep.
Whichever version of their account one considers, there are several rea-
sons for doubting that it provides a better explanation of strangeness than
Freud’s account, which is formulated in terms of efforts to disguise the
dream-content from the censor. First, their explanations depend on prior
claims already shown to be suspect: that the main stimuli for dream-content
come from physiological activation, which as we saw earlier has yet to be
established, and/or that bizarre dreams only occur during REM activation,
which subsequent research has shown to be false. Second, once again the
authors are relying far too much on the mere plausibility of their explana-
tions. To mount a stronger case they would need to show not just that their
explanations are plausible but that they can account in detail for a wide range
176 Chapter 7
of facts about dream bizarreness. Third, and relatedly, one of the most strik-
ing facts is that dream bizarreness increases as children grow, even though
the duration of REM sleep is at the same time decreasing. This seems to
contradict directly the authors’ claim that bizarreness is due to REM physiol-
ogy (see later and next chapter).
I have an additional objection, of a different kind. The authors give the
impression that Freud’s sole argument in favor of the censorship-disguise
theory is that it explains the strangeness of dreams, hence that if they offer a
better explanation of this strangeness, Freud’s theory would be shown false.
But that is not so. As I explained in Chapter 1, Freud had numerous reasons
for putting forward the theory—for example, that the recall of bizarre dreams
is often accompanied by resistance. It goes without saying that if one were to
disregard Freud’s main evidence then his theory would look much weaker.
Since the publication of the 1977 paper, Hobson in particular has continued
to champion the activation-synthesis model and to denounce Freud’s theory.
But his later writings have not advanced his initial arguments. Moreover,
little of the scientific evidence that has subsequently emerged adds weight to
them. Indeed, there has been mounting evidence against the activation-syn-
thesis hypothesis.
Even at the time of the original paper, the evidence in favor of activation-
synthesis was challenged. Objections were raised most forcibly by the sleep
physiologist Vogel (1978), who argued that there was neurophysiological
evidence that the forebrain is involved in the genesis of dreams after all. He
also argued that if the findings that the authors refer to are to serve as
evidence against Freud’s theory, then they need to show empirically that the
physiological processes of which they speak do not correspond to the
psychological processes of which Freud speaks, but they fail to do so. These
arguments prompted even Adolf Grünbaum to admit that “in the light of
Vogel’s critique of Hobson and McCarley it would seem at best premature to
suppose that their significant elucidation of the neurophysiology of dreaming
has obviated appraisal of Freud’s clinical arguments for his psychological
dream theory” (Grünbaum 1984, 219).
As it turns out, there is now considerably more evidence against activa-
tion-synthesis. As we have seen, much of Hobson and McCarley’s case rests
on the assumption that dreaming and REM sleep can be equated. But studies
have shown this false. Sleep-lab subjects have consistently reported dreams
upon awakening from NREM sleep. These include dreams indistinguishable
from REM dreams. Moreover, patients with damage to the REM-generating
regions identified by Hobson and McCarley reported they still have dreams,
whereas patients with damage to specific areas of the forebrain reported a
Brainwaves 177
cessation of dreaming (Solms and Turnbull 2002, 193–96; Solms 2000, 844).
Thus REM and dreaming are doubly dissociable: dreaming can occur without
REM and REM without dreaming.
Hobson has since broadened his theory to account for NREM dreams. His
claim now is that dreaming is correlated to periods of REM-like high activa-
tion, as occurs in REM, sleep onset, and late night NREM. He remains,
however, committed to the view that brainstem activation, under the right
concomitant conditions, 5 is the driver of dreams. This fails to address the
evidence that brainstem activation, even under the right concomitant condi-
tions, is insufficient for dreams—in particular the numerous cases of patients
with forebrain lesions reporting a loss of dreaming despite no effect on REM
(Solms 2000, 846). 6 Since in many of these cases the lesions only affected a
small area of the forebrain unrelated to visual imagery, it is difficult to see
why, if brainstem activation is the driver of dreams, it could not drive them
for these patients.
Solms, argues that the evidence points to forebrain mechanisms rather
than the brainstem as the driver of dreams (Solms and Turnball 2002, 197).
In particular, evidence from both lesioning and imaging studies suggests that
the two areas essential to dreaming are one “heavily implicated in the genera-
tion of visuospatial imagery” (the occipito-tempero-parietal junction) 7 and,
more tellingly as far as Freud’s theory is concerned, one that has to do with
emotion or motivation (limbic frontal white matter). This suggests that emo-
tion or motivation is more constitutive of dreaming than REM activation.
That the limbic system is crucial to dreaming is now widely accepted by
neurophysiologists. But Solms makes a more specific, although more conten-
tious, claim. He argues that, alongside his lesion evidence, pharmacological
evidence implicates the dopamine-fuelled basic emotional system that Pank-
sepp (1998, 52) calls the “appetitive motivational SEEKING system.” This is
“a nonspecific motivational system engaged in looking for something to
satisfy needs . . . linked to the pleasure/lust subsystem” (Solms and Turnball
2002, 201). He describes the following evidence for this:
Conclusion
The balance of the evidence seems not to favor Hobson. Unless he can
account for the contrary evidence in a detailed and consistent way, it may be
that activation-synthesis is untenable as a comprehensive account of dream-
ing. Should that be the case, then what would remain of activation-synthesis?
I think a number of things. One is the claim that extensive brain activation is
either a precondition, or at least a near necessary condition, for dreaming.
Another is that it is distinctly possible that internal physiological events—
such as those generated by pontine activation during REM—provide stimuli
for dreams. Another is that some constructive synthesis of random stimuli
may be involved in dreaming. All these can be important contributions to our
understanding of dreaming, though they require more evidence to confirm
them, and all are compatible with Freud’s theory of dreams.
Nevertheless, even a weak version of activation-synthesis seems to have
some negative implications for Freud’s theory. Freud thought that wishes
must be the driving force of dreams because other thoughts lacked sufficient
“energy” to drive the dream process. As we saw in Chapter 2, this was an
important part of his argument that all dreams are wish-fulfillments. Activa-
tion-synthesis suggests this may not be so. Even if it is the case, as the
evidence indicates, that forebrain mechanisms are the generators of dreams,
it need not be the case that only wishes have enough energy to drive the
process. The energy required for dreaming may be physiologically generated.
Thus, supposing the brain to have been activated by the brainstem, it could
be that non-wishful thoughts can drive the dream. This seems to undercut
Freud’s argument for his key claim that dreams are wish-fulfillments.
It may be thought that this is not a major problem, since Solms has given
evidence suggesting that the dopaminergic SEEKING system is the forebrain
mechanism responsible for generating dreams. But a problem remains. Even
if Solms’s neurophysiological conclusions are correct, this would be despite
the failure of Freud’s own argument for this conclusion. Freud had no knowl-
edge of the neuroscientific evidence for the generation of dreams. His argu-
ment depended on his assumption that non-wishful thoughts did not have
enough energy, which is cast into doubt by activation-synthesis. Thus it
seems that the correctness of his conclusion would be entirely serendipitous.
I think this challenge to Freud’s claim about wish-fulfillment can be met.
First, Freud’s rationale was based at least in part on his interpretative find-
ings, so even if his theoretical argument fails, he could point to this inductive
evidence in his favor. Second, it is not clear that his argument does fail, for it
180 Chapter 7
has not been established that non-wishful thoughts could drive the dream
process. Freud would say that simply being physically energized is not
enough. It is not energy levels per se that matters. What matters is that there
is a force that can drive an entire dream episode. 9 Dreams are usually vivid-
ly-conscious prolonged narratives with a high degree of internal continuity
despite the occasional scene shift. 10 It takes some kind of directed force to
sustain this, in the same way that it takes some kind of directed force to
sustain intentional behavior. Thus Freud would be right to ask not just where
the physical energy for this process comes from, but what the motive force
for it is. His judgment would be that non-wishful thoughts on their own
simply lack the motive force to drive this extended process. The challenge to
his argument posed by activation-synthesis is thus based on a confusion: it
conflates his notion of psychical energy—which pertains to motive force—
with mere physical energy. 11
This relates to a third reply, which dispenses with Freud’s particular
argument and looks at the consequences of his key premises. This is the
default argument I gave in Chapter 2. The most important premise is that
which is central to psychoanalysis as a whole—the claim that, throughout our
lives, we have unconscious impulses that seek expression but are blocked
from this by repression. The idea is that these unconscious impulses are
influencing much of our behavior, including our ordinary waking thoughts,
which take on some of the psychical energy of the unconscious because the
force of repression is never sufficient to block this off completely. If this
general framework is correct, then it has implications for what happens dur-
ing sleep. Freud assumed that sleep robs the force of repression of much of
its power, while the unconscious impulses remain as powerful as ever. This
change in the balance of power would mean that there are now impulses that
must find expression. Most of the normal routes to expression—ordinary
thoughts, speech, physical movement, and so on—are cut off during sleep.
The only route left, it appears, is through dreams. Hence, if the premises of
the argument are correct, the likely conclusion is that unconscious impulses
are finding their expression in dreams.
Neuroscientific evidence contributes to the plausibility of some of the
premises of this argument. For example, brain activation during sleep is
selective: it activates some parts of the brain more than others. In particular,
the limbic system is more highly activated than the prefrontal cortex, the part
of the brain responsible for executive control and planning (Hobson et al.
2003, 17). Since unconscious wishes are likely associated with limbic struc-
tures, and repression—which requires monitoring one’s thoughts—with the
prefrontal cortex, this supports the claim that during sleep (at least during the
phases when dreams are seen to most occur) the force of unconscious wishes
is relatively more powerful than the force of repression is. This does not, of
course, prove the conclusion of the argument, but it does show that, as long
Brainwaves 181
tion tendencies that are normally suppressed and controlled by the ‘executive
ego’” (2013, 205). That sounds remarkably like Freud’s theory.
There are other points of potential convergence. These include Hobson’s
allowing that dreams are interpretable through free association, that they are
in fact “motivated” in much the way Solms describes, that they may well be
motivated by repressed thoughts, and that such thoughts may reflect early
experiences:
The unfettered play of dopamine in REM sleep is in keeping with the assump-
tion that dreaming is “motivated” and that important motivational goals may
be revealed in dreams. (41)
Dynamically repressed (or actively forced down) mental content may well
emerge in the process of dream image creation and plot selection processes
that activation-synthesis credits with dream production. (69)
Sometimes dreams do reveal that earlier life issues, long believed dead, are
still very hot in our non-conscious brain-minds. Freud does deserve credit for
insisting on the long term persistence of conflict and trauma. (72)
Given how far Hobson has moved on other issues, his resistance on the
two remaining points about censorship-disguise and the function of dreams
should not be considered permanent. When one factors out the more eccen-
tric of his opinions, there appears little about his current view of dreams that
contradicts Freud’s and much that seems merely to re-describe it. It is there-
fore curious to observe Hobson continuing to speak disdainfully about Freud,
while edging ever closer to him.
NOTES
1. Nielsen (2000) estimates that people report dreams after being woken from REM sleep
around 80 percent of the time, but only around 12 to 25 percent of the time when woken from
NREM sleep.
2. With the rare exception of those suffering from nighttime sleep-related eating disorder.
3. Others have offered a few more similarly speculative examples (Schönhammer 2005),
but stronger evidence is required to see these as solid grounds for the general conclusion that
typical dreams are best explained (largely) physiologically.
4. The evidence from both laboratory and naturalistic sleep research is that typical dreams
are the exception rather than the rule (Domhoff 1996, 198; Snyder 1970, 148).
5. Relating to input source and modulation, as per Hobson et al.’s (2003) AIM model.
6. According to Solms, “The possibility that the reported loss of dreaming in these patients
is attributable to amnesia for dreams rather than true loss of dreams has been excluded not only
by REM awakening but also by neuropsychological examination of memory functions in
dreaming versus nondreaming patients” (2000, 850).
7. Solms observes that this area of the brain has to do with the “output” end of visual
processing in waking life, pertaining to “abstract aspects,” yet the fact that damage to it causes
complete cessation of dreaming but damage to other areas associated with visual processing do
not suggests that in dreaming it becomes the “input” end. This is in line with Freud’s view that
dreaming involves “regression”—a reversal of the normal processes of thought from high- to
low-level organization (Solms and Turnbaull 2002, 208–211; 1900, S.E., V: 543).
8. See Perogamvros and Schwartz (2012, 1941–42) for a review.
9. This accords with Perogamvros and Schwartz’s (2012, 1943) observation that dream
maintenance needs to be explained alongside dream production. They hold that their own
neurobiologically motivated Reward Activation Model “supports some of Freud’s claims about
dreaming because it proposes that activation of the SEEKING reward-related system (urge for
Freud) could generate dreaming (production) and could preserve dream continuation (mainte-
nance).”
10. The narrative coherence of dreams has been corroborated by content analysis (Domhoff
2005).
11. See also the discussion on psychical energy in Chapter 2.
12. Hobson’s current proposal (2014, 70) offers no such explanation: “The bizarreness of
dreams . . . [is] the understandable by-product of two non-conflictual factors: (1) the unavail-
ability of the real world space-time continuum and (2) the chaotic nature of the REM sleep
brain activation process.”
13. Hobson actually regards dream consciousness as epiphenomenal to REM processes, and
hence without a true function (see Metzinger 2009, 160). This is not far away from Freud’s
view: “It was indeed to be expected that dreaming, even though it may originally have been a
process without a useful purpose, would have procured itself some function in the interplay of
mental forces” (1900, S.E., V: 579).
Chapter Eight
Mind-Blocks
185
186 Chapter 8
nothing within his theory that precludes the censor playing some role in
dream production, or that precludes some dreams that are not obviously
wish-fulfilling nevertheless turning out to be so. He states, “It also seems
likely that [Freud’s] idea of wish fulfillment holds true for some unknown
number of dreams” (2003, 143). He also observes that Freud “is responsible
for the idea that dreams may be the product of figurative thought” (ibid.), a
reference to aspects of Freud’s dream-work that vindicates to a degree the
idea that dreams can be interpreted. Also, though he argues against Freud’s
claim that the function of dreams is to guard sleep, there is nothing in his
argument that implies that dreams cannot serve this function. Thus Dom-
hoff’s theory is compatible with at least weakened versions of all of Freud’s
key claims.
Despite this compatibility between their theories, Domhoff takes a force-
ful approach to arguing against Freud’s theory. He lists a number of areas in
which he thinks the empirical evidence speaks decisively against Freud. This
list is immensely valuable, as it constitutes perhaps the most exhaustive
empirically-based case against Freud’s theory. In this chapter I will address
the most important of his objections. How I determine an objection’s impor-
tance is by its degree of relation to Freud’s five key claims about dreams, as
analyzed in Chapter 1. Of course, Freud had a great many things to say about
dreams and one could not reasonably expect all of these to be true. But the
falsity of many of these claims would have no bearing on the core of his
theory, since they are not key claims or essential to his arguments for the key
claims. Examples of these peripheral claims include that a dream is like “a
firework that has been hours in the preparation, and then blazes up in a
moment” and that all important speeches in dreams are unoriginal reproduc-
tions from memories. I will pass over objections to such claims as inessential
to our current concerns, and instead focus on the more important objections.
DOMHOFF’S OBJECTIONS
this part of the book, my concern in assessing this evidence is with whether it
does indeed warrant such rejection.
Children’s Dreams
Freud believed that, from a very young age, children dream. He also thought
that their dreams were mostly straightforward wish-fulfillments, and used
this supposed fact to motivate his general wish-fulfillment claim. Domhoff,
however, argues that the scientific evidence shows that very young children
(up to age three) do not dream, that young children’s (ages three to five)
dreams are almost always “static and bland . . . not at all like Freud’s anecdo-
tal examples,” and that “there are no signs of wishes in children’s cognitively
impoverished dream reports” (2003, 136). The evidence he cites is based on
the work of David Foulkes (1982, 1999).
Assessment
The objection that young children’s dreams are “static and bland” is impor-
tant, since if there is to be any hope for Freud’s claim that dreams are wish-
fulfillments, then one would expect most young children’s dreams to con-
form to it. In contrast, the objection that very young children do not even
dream is not important, since, if correct, it would not undermine the wish-
fulfillment claim, as this is based on considerably more evidence than just the
few examples of children’s dreams that Freud cites in support. Indeed, even
if Freud’s evidence had been limited only to children’s dreams, most of these
are from children over three years old—of the fifteen dreams of children he
describes in his writings, only four are from children under three (Colace,
2010, 14–16). It is therefore only the finding that children’s dreams are
“static and bland” that would be damaging to Freud.
Both the alleged findings cited by Domhoff are in any case dubious. This
is particularly so given that there are general reasons why one might doubt
such results. The study’s conclusions are drawn from dream reports, but
children’s capacity to report dreams is plausibly limited by their ability and
willingness to articulate their experience. While observing that some authors
report dreams in children younger than two, Colace points out that “it is very
difficult to discover” at what stage children start dreaming until they have
enough language ability to express themselves (2010, 59). Also, against
Foulkes’s claims that young children’s dreams are short and bland, Resnick
et al. (1994) draw the following conclusion: “In contrast to previous studies,
our data indicate that young children are able to give long, detailed reports of
their dreams that share many formal characteristics with adult dream reports”
(30). This includes their finding that the rate of dream reporting in four- and
five-year-olds is almost the same as that of eight-, nine-, and ten-year-olds.
They also offer a potential explanation of the discrepancy with Foulkes’s
188 Chapter 8
For children’s dreams are . . . significant and not puzzling. Here, incidentally,
we have a further argument against tracing the origin of dreams to dissociated
Mind-Blocks 189
cerebral activity during sleep. For why should a reduction in psychical func-
tioning of this kind be a characteristic of the state of sleep in the case of adults
but not in that of children? (1901, S.E., V: 643)
As we can see . . . children’s dreams are not senseless. They are intelligible,
completely valid mental acts. You will recall what I told you of the medical
view of dreams and of the analogy with unmusical fingers wandering over the
keys of a piano. You cannot fail to observe how sharply these children’s
dreams contradict this view. It would really be too strange if children could
perform complete mental functions in their sleep while adults were content
under the same conditions with reactions which were no more than “twitch-
ings.” (1916, S.E., XV: 127–28)
This argument has stood the test of time, as the findings of Colace and other
researchers show. These findings put considerable pressure on all theories
that regard bizarreness as intrinsic to the dream process.
Additional details match the evidence. Due to his developing views about
infantile sexuality and dream symbolism, Freud came to believe that dream-
distortion was more prevalent in children’s dreams than he had originally
thought. Thus it is not to be expected that all children’s dreams lack bizarre-
ness or are straightforwardly wish-fulfilling. This is consistent with Colace’s
findings (2010, 129–51). Also, Freud’s theory predicts that the bizarreness
would be greater in psychologically disturbed children. This prediction has
been confirmed by Colace (2010, 127) and other researchers (Foulkes et al.
1969).
Further, Colace conducted studies that show that dream-bizarreness cor-
relates to the development of “superego functions” in children. Superego
functions include “the ability to experience a sense of guilt” and the “acquisi-
tion of moral rules.” Freud’s later development of his theory of dreams is that
the role of the “censor” is performed by the superego, so measuring the
correlation between bizarreness and superego functional development is a
test of Freud’s censorship-disguise explanation of bizarreness. The results of
three studies of this by Colace were positive. This provides qualified evi-
dence in support of Freud’s theory.
I say it is qualified because it was found that there is also positive correla-
tion with certain cognitive abilities, such as linguistic skills, attention and
discrimination ability, and perceptive organizational ability—though inter-
estingly there is no correlation with other cognitive abilities, such as descrip-
tive abilities, memory, or imaginative ability (Colace 2010, 144, 146). Co-
lace asserts that the correlations are compatible:
The findings from my studies suggest that dream bizarreness production may
require certain cognitive prerequisites, but also that dream bizarreness is clear-
ly affected by the level of development of the superego functions of personal-
ity, as Freud thought. . . . We may view cognitive abilities as a necessary
190 Chapter 8
Verdict
Dream-Work
Freud distinguished between the manifest and the latent content of dreams,
and argued that the latter was transformed into the former by the dream-
work. Domhoff objects that “No convincing nonclinical studies . . . have
demonstrated the operation of the dream-work” (2003, 137). He also states
that research suggests that “more information is available in the manifest
content of the dream than would be expected if the dream-work had made the
manifest dream-thoughts relatively meaningless” (ibid.).
Assessment
wishes that drive dreams belong to the id, while the preconscious thoughts
that these wishes utilize in order to find their way into consciousness belong
mainly to the ego. There was however a longer-term effect: Freud gradually
began to recognize that the formation of dreams owes more to the ego than
he had previously thought (Weiss 1992).
The original version of Freud’s theory already incorporated the ego’s
involvement through the operation of secondary revision. As we saw in
Chapter 1, according to Freud this operation is responsible for giving the
dream a façade of intelligibility. We also saw that this operation is continu-
ous with waking thought—indeed, the direct evidence for it comes from the
presence in the dream of such wake-like thoughts as “This is only a dream.”
Both these points suggest that secondary revision is carried out by the ego.
Freud did not express this directly in the original version of his theory, and
correspondingly underplayed the role of secondary revision in dreams. But
he corrected this in his latest version, described posthumously in An Outline
of Psycho-Analysis (1940), in which he makes clear that this operation is an
ego function. That this is so makes likely that a dream that has been so
worked on will carry the imprint of the ego’s concerns—that is, not only will
the dream contain ego thoughts as part of the preconscious material that
makes its way into the dream, but also, in being subsequently worked on, it is
likely to reflect the ego’s preoccupations.
Moreover, given that the ego has an obvious and general adaptive func-
tion—in that it serves to solve problems to the benefit of the individual—and
given also that the dream material presented to it typically carries the imprint
of conflict—of an uneasy compromise between a motivating wish and a
censorious counter-wish—one might expect the ego to serve adaptive ends in
its operation of secondary revision. As the ego works over the preconscious
material presented to it, it does so in a way that reduces residual tension, for
example by attempting to solve problems the dreamer faces in waking life. 5
Freud’s own early example of secondary revision, the thought that “This is
only a dream,” appears to serve precisely this purpose of reducing anxiety.
Such anxiety is, on Freud’s account, the result of the underlying dream
conflict at its rawest: a forbidden wish recognized in the dream by the censor.
In more disguised dreams the conflict is buried under layers of preconscious
thoughts that may nevertheless bear its stamp in their own inner tensions, and
it is on these that secondary revision may attempt a palliative solution. Hence
there is considerable justification from within Freud’s mature theory for
dreams serving adaptive purposes.
This relates to another study that Domhoff mentions. Greenberg et al.
(1992) studied the dreams of two subjects and obtained results favorable to
the hypothesis that dreams address problems from waking life. They found,
moreover, that the apparently successful resolution of the problem in the
dream tended to have a positive effect in waking life. Domhoff takes this
Mind-Blocks 193
study as having negative implications for Freud’s claim about the dream-
work. But even the authors of the study recognize this is not so. They state,
“Our data are neutral to the question of whether these problems could also be
seen as a disguised expression of hidden drives or wishes” (546). They later
summarize their conclusion thus: “Perhaps alongside or perhaps in place of
the view that the dream is trying to hide significant content, is the idea that a
dream is a direct expression of what is troublesome and of efforts to cope
with those issues” (547–48). The recognition that their results have no signif-
icant bearing on Freud’s claim about the dream-work reflects my point that
Freud’s theory is compatible with dreams serving adaptive purposes.
I think by now the point is well-illustrated. None of the studies Domhoff
relies on has a significant bearing on Freud’s claim about the dream-work,
and it is hard to imagine a nonclinical study that would have such a bearing.
That does not, of course, mean that the claim should be accepted, but it does
mean that nonclinical research has not ruled it out.
Verdict
Dream Emotion
Assessment
In some dreams the affect does at least remain in contact with the ideational
material which has replaced that to which the affect was originally attached. In
others, the dissolution of the complex has gone further. The affect makes its
appearance completely detached from the idea which belongs to it and is
introduced at some other point in the dream, where it fits in with the new
arrangement of the dream-elements. (1900, S.E., V: 463; my italics)
He illustrates the point with an example from one of his own dreams, in
which his latent thought of his premature death is represented in the dream as
the death of another man, accompanied by his indifference, while the fear
attached to this thought is displaced to an entirely different scene in the
dream, that of the appearance of impending warships, where it would on a
straightforward rendering have seemed appropriate. Thus the fact that emo-
tion matches dream-content does not mean that displacement has not taken
place.
Though it would have little bearing on any claim central to Freud’s theo-
ry, the assertion that Domhoff quotes, that dream emotion is “overwhelming-
ly appropriate to content,” if accurate, would indicate that Freud exaggerated
the degree to which emotion may be inappropriate to dream-content. The
assertion, however, does not seem accurate. Domhoff reports a study in
which “the type of emotion, or lack thereof, was appropriate to the dream
situation in 60% of the dreams” (2005; Foulkes et al. 1988). This still leaves
forty percent where emotion is not appropriate to dream-content. Thus if
these results are correct, then Freud was not completely off the mark in
pointing out that dream emotion often does not match dream-content. More
importantly, one would want an explanation of that forty percent of cases
where emotion does not match content, which Freud’s theory provides.
It is also worth mentioning a result that is often reported as unfavorable to
Freud’s theory, but may instead be favorable. In numerous studies it has been
found that negative emotions, such as anxiety and anger, occur much more
frequently in dream reports than positive emotions (Domhoff 2005). This is
often taken as evidence against Freud’s wish-fulfillment claim, even though,
as we saw in Chapter 2, Freud’s theory does not imply that the emotions in
the dream-content should be positive. 7 What is interesting is that this result,
which is based on the judgments of blind coders (that is, independent asses-
sors), differs markedly from the view that the dreamers themselves have
about their dreams. In one study it was found that “the specific emotions
were negative by 2:1 according to coders, but positive in general mood
according to the dreamers themselves by 2.5:1” (Domhoff 2005; Strauch and
Meier 1996). Domhoff laments that there is “no ready explanation for these
Mind-Blocks 195
contrasting results” (ibid.). But Freud’s theory provides just such an explana-
tion—that there exists a latent content of dreams that is different from the
manifest content.
Verdict
Wish-Fulfillment
Domhoff criticizes Freud’s claim that all dreams represent the fulfillment of
wishes by reiterating well-known objections. According to Domhoff, anxiety
and punishment dreams are evidence against the claim, and Freud’s defense
against these objections is merely “theory-saving” (2003, 139). Further, trau-
matic dreams decisively refute the claim, as Freud himself admitted. As we
saw in Chapter 2, Freud treated traumatic dreams as exceptions to the rule,
and he revised his theory in such a way as to maintain the spirit, if not the
word, of his original claim—specifically, his new claim was that all dreams
are the attempted fulfillment of a wish. Domhoff argues that traumatic
dreams are too widespread and persistent to be treated as an exception (ibid.,
140).
Domhoff goes on to counter Solms’s neuroscientific defense of the wish-
fulfillment claim. He points out that Solms’s argument that the dopaminergic
SEEKING system is the driver of dreams “is greeted with skepticism by
other neuropsychologists and neurophysiologists because dopamine produc-
tion is about the same in waking and REM and is probably only one aspect of
a complex neurochemical mixture” (141). He also argues that the finding that
“dopaminergic blockers do not eliminate dreams” counts against Freud’s
theory.
Assessment
I have already discussed the objection from anxiety and punishment dreams
in Chapter 2. I do not think it fair to say that Freud’s defense against this
objection is “theory-saving,” since the account he offers of such dreams
comes from the core of his theory. The particular explanation of such dreams
that he gives is drawn from his model of a conflict between a forbidden wish
and a repressive counter-wish. As such, to call his explanation “theory-sav-
ing”—with the obvious connotation of ad hocness—is unjust, since it is
illustrative of central tenets of his theory.
Posttraumatic stress disorder (PTSD) dreams are, however, a decisive
refutation of Freud’s original wish-fulfillment claim. But Domhoff’s objec-
196 Chapter 8
With the traumatic neuroses things are different. . . . We should not, I think, be
afraid to admit that here the function of the dream has failed. I will not invoke
the saying that the exception proves the rule: its wisdom seems to me most
questionable. But no doubt the exception does not overturn the rule. (1933,
S.E., XXII: 29)
What he means is that the fact that dreams fail to fulfill wishes in the case of
trauma does not show that the function of dreams is not to fulfill wishes. This
is a correct and well-known fact about functions. The failure of a purported
function to be manifested does not prove that it is not the true function. For
example, that many people are blind does not prove that it is not the function
of the eyes to see.
Freud was prepared to accept that traumatic dreams were more wide-
spread than just the “war neuroses.” He writes, “We must admit that child-
hood experiences, too, are of a traumatic nature” (1933, S.E., XXII: 30).
Given the emphasis he put on childhood trauma, this suggests that, at the
time he made these remarks, he believed that, quantitatively but not function-
ally speaking, non-wish-fulfilling dreams are a significant subset of all
dreams. This, as I have argued, does not negate his claim that the function of
dreams is to fulfill wishes. At best it weakens the inductive argument for the
claim, but, as we saw in Chapter 2, this was not his main argument.
With regard to Solms’s defense of the wish-fulfillment theory, Domhoff
remarks that his “strong emphasis on the dopaminergic system does not
speak to posttraumatic dreams” (2003, 141). It is true that a fully developed
theory should address posttraumatic dreams and it is not clear how Solms’s
theory would do this. Nevertheless, posttraumatic dreams should not be taken
as falsifying Solms’s ideas. Solms’s hypothesis is that the dopaminergic
system is involved in the generation of dreams, consistent with Freud’s claim
that wishes drive dreams. This claim is not refuted by PTSD dreams. These
dreams prima facie refute the claim that all dreams represent the fulfillment
of a wish, but not the claim that all dreams are motivated by a wish.
Domhoff’s subsequent criticisms of Solms miss the mark. First, Solms
does not deny that dopamine is one aspect of a complex neurochemical
mixture—his hypothesis is that it is necessary, not sufficient, for dreams
(Solms 2003, 249). This relates also to Domhoff’s remark that dopamine
production is about the same in waking and REM. It is based on the follow-
ing criticism by Gottesmann (2003, 153):
Mind-Blocks 197
Verdict
Function
Freud thought that the function of dreams was to preserve sleep by serving as
a safety-valve for psychical excitation seeking discharge. Domhoff claims
that this is contradicted by several different lines of evidence: (1) the fre-
quency and regularity of dreaming suggests that dreaming cannot be mainly
concerned with dealing with episodic urges; (2) content analysis reveals that
dreams seldom involve desires like hunger and thirst; (3) Foulkes’s study of
children’s dreams shows that very young children rarely dream, yet they
sleep well enough; (4) leucotomized schizophrenics rarely report dreams, yet
sleep normally; and (5) Solms’s study of non-dreamers, which he claims is
supportive of Freud’s theory, nevertheless shows that about half of these
people did not have significant sleep disturbance.
Assessment
The first criticism is a strange one, as Freud’s theory is not that dreams only
exist to cope with episodic urges, but with the recurrent pressure of uncon-
scious excitation. This is likely to be strongest during REM sleep, as in this
phase, the limbic system and associated areas are highly active. The predic-
tion of Freud’s functional claim would be, therefore, that dreaming is very
frequent during this period, and that is indeed the case. So the evidence here
is actually in line with Freud’s claim.
The second criticism is also unconvincing. Hunger and thirst dreams may
be rare because the subjects of dream research rarely go to bed hungry or
thirsty enough for these desires to threaten sleep, or because of some similar
reason. Whatever the case, Freud used hunger and thirst dreams merely as
examples to motivate his claim, where what is at issue is whether dreams
predominantly help to guard against unconscious impulses.
The third criticism is unconvincing for reasons I have previously raised.
The rarity of dream reports in preschool children may have more to do with
inability to report than with absence of dreams. We simply have no good way
of knowing the extent to which very young children dream. Moreover, on
Mind-Blocks 199
Freud’s theory, the main disturbers of sleep are repressed impulses, which
very young children do not have, so one would not expect frequent waking
even in the absence of dreams.
The fourth criticism is more serious. Domhoff reports that a team of
Canadian researchers in the 1970s found that lobotomized schizophrenics
reported fewer dreams than a control group of unlobotomized schizophren-
ics, yet there were no obvious differences in the sleep of the two groups (Jus,
Jus, Gautier, et al. 1973; Jus, Jus, Villeneuve, et al. 1973). Domhoff consid-
ers this a refutation of Freud’s functional claim.
The above criticism should be considered in conjunction with the next,
which relates to Solms’s studies. Solms (1995) found that brain-damaged
patients who reported no dreams had significantly more disturbed sleep than
brain-damaged patients whose dreaming was unaffected. He interpreted this
as favorable to Freud’s functional claim. Domhoff, however, considers it
evidence against the claim, since about half of the non-dreaming patients
indicated no significant sleep disturbance. According to Domhoff, this, along
with the previously mentioned study on schizophrenics, constitutes decisive
evidence against the claim that “dreams are necessary to preserve sleep”
(2004, 9).
But herein lies the chief fault of Domhoff’s criticism. Freud’s claim is
that the function of dreams is to preserve sleep, not that dreams are necessary
to preserve sleep. The two claims are not equivalent. A simple example
illustrates. The function of guard dogs is to protect property, but that does not
mean that guard dogs are necessary to protect property. The fact that many
properties remain safe without guard dogs does not negate the claim that that
is their function. Similarly, dreams can serve the function of preserving sleep
without it being the case that they are necessary for sleep.
As it happens, many of those in Solms’s study could not dream because
of impairment to their appetitive motivational SEEKING system, for as we
have seen, one of the main reasons for loss of dreaming is damage to this
system. So their loss of dreaming would have gone hand-in-hand with a loss
of sleep-disturbing stimuli. In which case, their lack of sleep disturbance
does not count against Freud’s functional claim. This is also true of the
schizophrenics in the studies that Domhoff cites, who had damage to circuits
of the SEEKING system. Solms correctly points out that a better way to test
Freud’s claim is to focus on patients who have lesions that prevent dreams
but whose motivational system is still intact. As yet, though, such a study has
not been carried out.
Even if such a study were to yield some negative results, though these
would constitute disconfirming evidence, they would not be a decisive refu-
tation. It could be, for example, that in these cases sleep is preserved because
people develop compensatory mechanisms—like stronger than normal re-
pression during sleep (cf. 1900, S.E., V: 577–579). 8 As such, a reasonable
200 Chapter 8
way to assess the functional claim is to look for a general trend. That is what
Solms did in the more limited study described above, and he found that
absence of dreaming is indeed associated with increase in sleep disturbance.
A recent review of the scientific evidence by Guénolé et al. (2013)
presents a more sober assessment than Domhoff’s. It shows that there is
some evidence against Freud’s claim, such as the fact that primary anoneira
(the “constitutional inability to dream”) is not known to be linked with in-
somnia—though the authors point out that studies of this rare condition are
few and subject to methodological limitations. It also shows that there is
some evidence in favor of the claim, such as the finding that dream reporting
in neurotic patients—who presumably have more unconscious arousal, hence
more need of a safety-valve—is significantly higher than controls. The au-
thors’ overall conclusion is as follows:
As the authors point out, given the limitations of the studies reviewed, con-
siderably more research is needed to properly test Freud’s claim.
Verdict
Free Association
Domhoff (2003, 143) asserts that Freud’s theory rests exclusively on the
method of free association. He objects to the method, firstly, on the grounds
that there is no evidence it has probative value. He supports this by reporting
a study by Foulkes that attempted to use free associations to interpret dreams,
but which led to the author concluding that the method was “inherently
arbitrary” (Foulkes 1996, 617).
Domhoff then argues that social psychology indicates the “power of sug-
gestion in a therapeutic context” (2003, 143). This provides an alternative
explanation of Freud’s results—that they are the result of “persuasion and
conversion.” He concludes that this “does not mean that all psychoanalytic
sessions have been shown to be exercises in suggestion,” but “it does mean
that the burden of proof is now on Freudians to demonstrate that any thera-
Mind-Blocks 201
peutic data they use to make claims about dreams are not confounded by this
extremely important variable” (ibid.).
Assessment
There are several possible reasons why this may be so. The most obvious
is that the method does not work—that is, does not enable the recovery of
dream meaning—and the appearance that it does can be otherwise explained.
This is the alternative that Domhoff favors. But there are other possible
reasons for the negative results. One is that the method works, but only in the
hands of an especially skilled practitioner. This is in line with Bleuler’s
comment (see Chapter 6) that dream interpretation is more an art than a
science. Another is that studies to date have been inadequate. This would be
even more likely if, as many have suggested, Freud has falsely described his
method. Recall from Chapter 3, Glymour’s criticism that Freud’s description
does not match the practice of the method. My subsequent analysis lends
some support to this criticism, since it allows that not all associations are
useful for interpretation. If this is so, then it may be that researchers have
been testing the wrong method and need to pay closer attention to how Freud
practiced his dream interpretations rather than to how he described the tech-
nique. All in all, the lack of scientific evidence supporting Freud’s method
should be worrying to Freudians, though it does not in itself indicate that the
method is “inherently arbitrary.” Alternative explanations exist and need to
be explored.
Domhoff adds weight to his preferred explanation of the negative results
by providing an account of why it might be that the method often appears to
work. He points to experimental evidence of how a therapist’s suggestions in
the course of dream interpretation can influence people into believing in
invented childhood events (Mazzoni et al. 1999). He also suggests that a
combination of the authority of the therapist and cognitive dissonance is
what leads many clients to believe their therapist’s interpretations (Domhoff
2003, 54–55).
The studies Domhoff refers to pose a serious challenge to Freud’s meth-
od. They show how easy it is to influence people’s beliefs through dream
interpretation in a therapeutic context. It should be noted that the method of
interpretation used in the studies was not that of Freud’s, though it bore
superficial resemblances to it. Nevertheless, this evidence adds considerable
plausibility to the suggestion that apparent interpretative success is illusory.
The conclusion to be drawn from the studies is that the dreamer’s accep-
tance of an interpretation does not prove its success. A word of caution,
though. This does not imply that the dreamer’s responses during analysis
cannot be evidence at all. In Chapters 3 and 4, I argued that subtle pheno-
menological characteristics of the dreamer’s responses can constitute evi-
dence in favor of an interpretation. In particular, if the connection between a
dream element and an accompanying association strikes the dreamer as sig-
nificant, that can be evidence for the hypothesis that that association was a
cause of that element. These points are not undermined by the above studies
for three reasons. First, the above studies have nothing to say about the subtle
Mind-Blocks 203
First, I do not think that one can begin straightaway with a test of the
reliability of the method. As things stand, the method is too under-described
and unsystematic for this. My thinking here is guided by the argument in
Chapter 6 about the lessons that can be drawn from telescopy and microsco-
py: one needs to first establish that the method can work in certain cases, and
then see to what extent this can be generalized. Once the method has been
rendered more systematic and rigorous, in line with one’s initial findings,
one will be in a better position to establish its general reliability. The initial
focus should therefore be on validating the method in circumstances condu-
cive to a correct interpretation. Let me sketch such a study. 9
to ignore it, and simply hope that there is enough evidence from the
recording of the session and from the shared cultural understanding of
the judges with the client for them to make good judgments. A second,
is to acquaint the judges with other therapy sessions involving that
client, to the extent that they gain familiarity with the client’s inner
world. A third is to have the judges conduct in-depth interviews with
the client prior to their assessment. An interview could also be used as
a means of assessing the dreamer’s subjective appraisal of the connec-
tions between the dream elements and the free associated thoughts.
5. The second team of judges, those evaluating the overall interpretation
given the evidence, have a more straightforward task. They are asked
to grant that the relevant free associations are dream-thoughts, and on
this basis form a judgment about the analyst’s overall interpretation,
assigning a probability value that corresponds to their degree of belief
in the correctness of this interpretation. This judgment also draws on
other evidence present in the recording—including the client’s infor-
mation about what preceded the dream and her verbal and non-verbal
behavior, in particular, her overall response to the interpretation.
6. Finally, appropriate statistical analyses are carried out on the results.
These include analyses of inter-judge consistency and a Bayesian
analysis to provide an overall probability value for the interpretation
(Sturrock 1994).
Verdict
CONCLUSION
In this chapter and the last I have examined the most prominent scientific
criticisms brought against Freud’s theory. We have seen that neither Hob-
206 Chapter 8
observation that the high arousal of REM sleep correlates with increased
dreaming.
There is also other evidence that I have not discussed above, coming from
studies, old and new, that show that material suppressed during the day tends
to show up in dreams. For example, in a study published in 1931, Malamud
and Lindner showed pictures to patients receiving psychotherapy that con-
tained elements of “emotional significance” to the patient. Van de Castle
reports, “the investigators found that the conflict-charged elements were
omitted from the initial description of the pictures but were subsequently
incorporated into dreams” (1994, 218). Several other studies at the time had
similar findings (ibid., 217–218). More recently, Wegner et al. (2004) found
that suppression of thoughts of a target person before sleep increased dream-
ing about that person. These findings are consistent with the Freudian idea
that repressed thoughts find their way into dreams.
In another study (Köhler and Prinzleve 2007), researchers found that
people free associating with forgotten elements of dreams as compared with
remembered elements report more unpleasant feelings and show greater
physiological signs of discomfort (as measured by skin conductance re-
sponses). This is in line with Freud’s posit that dream forgetting is due to
repression. They also found that people free associating with elements of
their own dreams—whether remembered or not—as compared to somebody
else’s also report more unpleasant feelings and show greater physiological
signs of discomfort. This evidence is favorable to Freud’s censorship-dis-
guise claim (D2), and also offers some support for the relevance of free
associations (D1).
NOTES
1. As Colace (2010, 26–27) correctly points out, Freud did not consider all children’s
dreams to be free from distortion (1900, S.E., IV: 131n1).
2. See Kächele et al. (2009, 295–96) for a discussion of the limitations of CCRT in
psychoanalytic contexts.
3. To put it in quite these terms may be misleading. On Freud’s definition, most interpreta-
tions, even very unadventurous ones, belong to the latent content. For example, in discussing
children’s dreams, Freud explains that, though their interpretation is straightforward, this does
not mean that they belong to the manifest content: “But when we examine these dreams more
closely, we shall recognize a small piece of dream-distortion even in them, a certain distinction
between the manifest content of the dream and the latent dream thoughts” (1916, S.E., XV:
128). He means that, though the content of the thought finds its way into the dream, the attitude
of the thought—that which defines it as a wish—does not. Thus for Freud, if there is any
transformation of a thought, even if it does not affect its content, then that thought belongs to
the latent content.
4. A reason why many think that Freud’s theory is incompatible with an informative
manifest content is that he himself seemed to say so. For example: “It is natural that we should
lose some of our interest in the manifest dream. . . . [I]t is as a rule like a piece of breccia,
composed of various fragments of rock held together by a binding medium, so that the designs
that appear on it do not belong to the original rocks imbedded in it” (1916, S.E., XV: 181–82).
There are a few points that must be borne in mind, however. First, this is not a denial that the
manifest content is informative or related to waking concerns. Second, Freud had, at the time,
rhetorical and strategic reasons for warning against too much preoccupation with the manifest
content, in particular due to his wariness of dissenting lines of thought (Spanjaard 1993, 154).
Third, these remarks do not represent his final thoughts on this issue—as I go on to explain,
Freud increasingly came to see the ego as playing a more active role in shaping the dream-
content. Fourth, what matters most is not Freud’s occasional remarks, but whether or not his
theory is in fact compatible with an informative manifest content.
5. This is different from, but compatible with, Freud’s original explanation of problem-
solving in dreams—that it is due to restless preconscious thoughts recruiting unconscious
energy in order to complete their daytime activities (1900, S.E., V: 564).
6. Fisher and Greenberg (1996) mention three studies—Smith (1986), Robbins and Tanck
(1978), and Cavellero (1987)—whose results are in line with Freudian predictions but which
the authors dismiss as having negligible confirmation value.
7. Even at the time he wrote The Interpretation of Dreams, Freud was aware of results
similar to those that Domhoff cites. He mentions a study in which it was found that “57.2 per
cent of dreams are ‘disagreeable’ and only 28.6 per cent positively ‘pleasant’” (1900, S.E., IV:
134).
8. This is consonant with my point in Chapter 5 that “falsifiability” is rarely, if ever,
absolute. Thus though Solms (2014, 209) is right to seek an experiment that would offer clearer
confirmation or disconfirmation of the hypothesis, he is wrong to regard this as “a critical
experiment to decide the issue once and for all.” Freud perhaps had a better understanding of
this—as he remarked (in a different context), “If for the sake of studying it, we isolate one
particular psychic function, such as dreaming, from the psychical machinery as a whole, we
make it possible to discover the laws that are peculiar to it; but when we insert it once more into
Mind-Blocks 209
the general context we must be prepared to discover that these findings are obscured or im-
paired by . . . other forces” (1933, S.E., XXII: 29).
9. Many of the principles outlined here are based on Sturrock (1994).
10. A cause for optimism that it could be successful is Popper’s belief that “Freud’s analyses
of dreams given in his book are fundamentally correct” (1983, 164). If such an otherwise
trenchant critic could come to this view, then it is conceivable that sufficiently objective judges
could reach a similar judgment.
11. It ignores the fact that there are numerous reasons in favor of Freud’s account. The
argument from parsimony is strong when the only point in favor of a theory is that it solves a
problem that can be otherwise solved more simply.
12. Other researchers, noting that REM sleep involves genital arousal and a surge in sex
hormones, have conducted studies that support the assertion that most dreams reflect sexual
wish-fulfillment (McNamara 2014).
Conclusion
Finding the Seed of Gold
In Plato’s dialog Euthyphro, Socrates asks the eponymous young man who is
about to prosecute his own father in the name of piety, “What is piety?” To
each successively more sophisticated answer that Euthyphro (or Socrates on
his behalf) gives, Socrates proffers a counterargument. The outcome of this
process is an impasse—the two men are not able to arrive at an acceptable
definition. Such an outcome is common to Plato’s dialogs. The Greek term
for it is aporia.
So it seems with our dialog. We have arrived at an impasse, an aporia,
not knowing whether we should accept or reject Freud’s theory of dreams.
This stands in contrast with the certainty of others. Freud was certain his
theory was true. Glymour, Grünbaum, Sand, Hobson, Domhoff, and count-
less other critics seem certain it is not. The evidence we have looked at
warrants neither attitude, hence our verdict is apposite. But as with Plato’s
dialogs, the value lies not in attaining a resolution, but in the process of
attempting it—not in the destination, but in the journey. In this, I hope, much
has been learned. Here is a brief review.
In Chapters 1 and 2, we saw that Freud’s rationale for his theory com-
bines evidence from dream interpretations with theorizing about the mind.
Critics have claimed that the latter derives from nineteenth-century specula-
tive neurobiology. This is an exaggeration. Freud’s theorizing of the mind
does draw on speculative theory, but it does not derive from this. It is rather
the product of a complex cocktail of imagination, observation, and inference.
His models are not a priori constructions, but the products of a to-and-fro
between theoretical insight and empirical judgment. Nevertheless, the cock-
211
212 Conclusion
The Future
Aleman, André, and Frank Larøi. 2008. Hallucinations: The Science of Idiosyncratic Percep-
tion. Washington: American Psychological Association.
Alvarez, Anne, and Anthony Lee. 2004. “Early Forms of Relatedness in Autism: A Longitudi-
nal Clinical and Quantitative Single-Case Study.” Clinical Child Psychology and Psychiatry
9(4): 499–518.
Anderson, Michael C., and Collin Green. 2001. “Suppressing Unwanted Memories by Execu-
tive Control.” Nature 410: 366–369.
Anderson, Michael C., and Benjamin J. Levy. 2006. “Encouraging the Nascent Cognitive
Neuroscience of Repression.” Behavioral and Brain Sciences 29(5): 511–513.
Anderson, Michael C., and Benjamin J. Levy. 2009. “Suppressing Unwanted Memories.” Cur-
rent Directions in Psychological Science 18(4): 189–194.
Anzieu, Didier. 1986. Freud’s Self-Analysis. Translated by Peter Graham. London: Hogarth
Press.
Aserinsky, Eugene, and Nathaniel Kleitman. 1953. “Regularly Occurring Periods of Eye Motil-
ity, and Concomitant Phenomena, During Sleep.” Science 118: 273–274.
Aumann, Carolin, Olaf Lahl, and Reinhard Pietrowsky. 2012. “Relationship between Dream
Structure, Boundary Structure and the Big Five Personality Dimensions.” Dreaming 22(2):
124–135.
Berlin, Heather A. 2011. “The Neural Basis of the Dynamic Unconscious.” Neuropsychoanaly-
sis 13(1): 5–31.
Blackmore, Susan. 2006. “Sleeping Giant.” The Guardian, April 6. http://www.theguardian.
com/commentisfree/2006/apr/06/votingonfreud1.
Blass, Rachel B. 2001. “The Limitations of Critical Studies of the Epistemology of Freud's
Dream Theory and their Clinical Implications: A Response to Spence and Grünbaum.”
Psychoanalysis and Contemporary Thought 24(2): 115–151.
Blass, Rachel B. 2002. The Meaning of the Dream in Psychoanalysis. New York: SUNY Press.
Bonaparte, Marie, Anna Freud, Ernst Kris, Eric Mosbacher, and James Strachey, eds. 1954.
The Origins of Psycho-Analysis, Letters to Wilhelm Fliess, Drafts and Notes: 1887–1902.
London: Imago Publishing.
Borch-Jacobsen, Mikkel. 1996. Remembering Anna O.: A Century of Mystification. Translated
by Kirby Olson, in collaboration with Xavière Callahan and the author. New York and
London: Routledge.
Brenner, Charles. 1969. “Dreams in Clinical Psychoanalytic Practice.” The Journal of Nervous
and Mental Disease 149(2): 122–132.
Brown, Harold I. 1985. “Galileo on the Telescope and the Eye.” Journal of the History of Ideas
46(4): 487–501.
217
218 Bibliography
Carhart-Harris, Robin L., and Karl J. Friston. 2010. “The Default-Mode, Ego-Functions and
Free-Energy: A Neurobiological Account of Freudian Ideas.” Brain: A Journal of Neurolo-
gy 133(4): 1265–83.
Cavallero, Corrado. 1987. “Dream Sources, Associative Mechanisms, and Temporal Dimen-
sion.” Sleep: Journal of Sleep Research & Sleep Medicine 10(1): 78–83.
Chalmers, Alan F. 1985. “Galileo’s Telescopic Observations of Venus and Mars.” British
Journal for the Philosophy of Science 36: 175–184.
Chalmers, Alan F. 1999. What is This Thing Called Science? Third edition. Maidenhead: Open
University Press. Citations refer to 2005 reprint.
Cipolli, Carlo, Roberto Bolzani, Roberto Massetani, Luigi Murri, and Alberto Muratorio. 1992.
“Dream Structure in Parkinson's Patients.” The Journal of Nervous and Mental Disease
180(8): 516–523.
Clavelin, Maurice. 1974. The Natural Philosophy of Galileo. Cambridge, MA: MIT Press.
Colace, Claudio. 2003. “Dream Bizarreness Reconsidered.” Sleep and Hypnosis 5: 105–128.
Colace, Claudio. 2010. Children’s Dreams: From Freud’s Observations to Modern Dream
Research. London: Karnac Books.
Colace, Claudio. 2013. “Are the Wish-Fulfillment Dreams of Children the Royal Road for
Looking at the Functions of Dreams?” Neuropsychoanalysis 15(2): 161–175.
Colace, Claudio, and Bruna Tuci. 1996. “Early Children’s Dreams are Not Bizarre.” Sleep
Research 25: 147.
Davidson, Donald. 1984. “On the Very Idea of a Conceptual Scheme.” In Donald Davidson,
Inquiries into Truth and Interpretation, 183–98. Oxford: Oxford University Press.
Dennett, Daniel C. 2014. “What Is Dreaming For, If Anything?” In Nicholas Tranquillo (ed.),
Dream Consciousness: Allan Hobson’s New Approach to the Brain and its Mind, 113–117.
Vienna Circle Institute Library, volume 3. Cham: Springer International Publishing.
Domhoff, G. William. 1996. Finding Meaning in Dreams: A Quantitative Approach. New
York and London: Plenum Press.
Domhoff, G. William. 2000. “The Problems with Activation-Synthesis Theory.” http://
www.dreamresearch.net/Library/domhoff_2000e.html.
Domhoff, G. William. 2003. The Scientific Study of Dreams: Neural Networks, Cognitive
Development, and Content Analysis. Washington, DC: American Psychological Associa-
tion. Citations refer to 2007 printing.
Domhoff, G. William. 2004. “Why did Empirical Dream Researchers Reject Freud? A Critique
of Historical Claims by Mark Solms.” Dreaming 14(1): 3–17.
Domhoff, G. William. 2005. “The Content of Dreams: Methodologic and Theoretical Implica-
tions.” In M. H. Kryger, T. Roth, and W. C. Dement (eds.), Principles and Practices of
Sleep Medicine. Fourth edition, 522–534. Philadelphia: W. B. Saunders.
Edelson, Marshall. 1985. “The Hermeneutic Turn and the Single Case Study in Psychoanaly-
sis.” Psychoanalysis & Contemporary Thought 8(4): 567–614.
Edelson, Marshall. 1986. “Causal Explanation in Science and in Psychoanalysis: Implications
for Writing a Case Study.” The Psychoanalytic Study of the Child 41: 89–127.
Edelson, Marshall. 1988. A Theory in Crisis. Chicago: University of Chicago Press.
Ellenberger, Henri F. 1970. The Discovery of the Unconscious: The History and Evolution of
Dynamic Psychiatry. New York: Basic Books.
Erdelyi, Matthew H. 2006. “The Unified Theory of Repression.” Behavioral and Brain Sci-
ences 29(5): 499–511.
Erikson, Erik H. 1954. “The Dream Specimen of Psychoanalysis.” Journal of the American
Psychoanalytic Association 2:5-56.
Fisher, Seymour, and Roger P. Greenberg. 1977. The Scientific Credibility of Freud’s Theories
and Therapy. New York: Basic Books.
Fisher, Seymour, and Roger P. Greenberg. 1996. Freud Scientifically Reappraised: Testing the
Theories and Therapy. New York: John Wiley & Sons.
Forrester, John. 1997. Dispatches from the Freud Wars: Psychoanalysis and its Passions.
Cambridge, MA, and London: Harvard University Press.
Fosshage, James L., and Clemens A. Loew, eds. 1987. Dream Interpretation: A Comparative
Study. Revised edition. New York: PMA Publishing.
Bibliography 219
Foulkes, David. 1982. Children’s Dreams: Longitudinal Studies. New York: Wiley.
Foulkes, David. 1996. “Dream Research: 1953–1993.” Sleep 19: 609–624.
Foulkes, David. 1999. Children’s Dreams and the Development of Consciousness. Cambridge,
MA: Harvard University Press.
Foulkes, David, James D. Larson, Ethel M. Swanson, and Max Rardin. 1969. “Two Studies of
Childhood Dreaming.” American Journal of Orthopsychiatry 39(4): 627–643.
Foulkes, David, Brenda Sullivan, Nancy H. Kerr, and Lisa Brown. 1988. “Appropriateness of
Dream Feelings to Dreamed Situations.” Cognition & Emotion 2(1): 29–39.
Freud, Sigmund. 1966. The Standard Edition of the Complete Psychological Works of Sigmund
Freud. Translated and edited by James Strachey. London: Vintage, The Hogarth Press, and
The Institute of Psycho-Analysis. Citations refer to the 2001 Vintage edition.
Freud, Sigmund. 2001. On Dreams. Translated by M. D. Eder. Mineola, New York: Dover
Publications.
Fridhandler, Bram, Tracy D. Eells, and Mardi J. Horowitz. 1999. “Psychoanalytic Explanation
of Pathological Grief: Scientific Observation of a Single Case.” Psychoanalytic Psychology
16(1): 34–57.
Friston, Karl J. 2009. “The Free-Energy Principle: A Rough Guide to the Brain?” Trends in
Cognitive Sciences 13(7): 293–301.
Friston, Karl J. 2010. “The Free-Energy Principle: A Unified Brain Theory?” Nature Reviews
Neuroscience 11(2): 127–138.
Gaillard, Jean-Michel, and Ali Moneme. 1977. “Modification of Dream Content after Preferen-
tial Blockade of Mesolimbic and Mesocortical Dopaminergic Systems.” Journal of Psychi-
atric Research 13(4): 247–256.
Galileo Galilei. 1957. The Starry Messenger. In Discoveries and Opinions of Galileo, translat-
ed by Stillman Drake, 1–86. Garden City, NY: Doubleday.
Galileo Galilei. 1967. Dialogue Concerning Two Chief World Systems. Translated by Stillman
Drake. Berkeley, California: University of California Press.
Gardner, Sebastian. 2006. Irrationality and the Philosophy of Psychoanalysis. New York:
Cambridge University Press.
Gest, Howard. 2004. “The Discovery of Microorganisms by Robert Hooke and Antoni Van
Leeuwenhoek, Fellows of the Royal Society.” Notes and Records of the Royal Society of
London 58(2): 187–201.
Geymonat, Ludovico. 1965. Galileo Galilei. Translated by Stillman Drake. New York:
McGraw Hill.
Gilhooley, Dan. 2002. “Misrepresentation and Misreading in the Case of Anna O.” Modern
Psychoanalysis 27(1): 75–100.
Glymour, Clark. 1983. “The Theory of Your Dreams.” In Robert S. Cohen and Larry Lauden
(eds.), Physics, Philosophy and Psychoanalysis: Essays in Honour of A. Grünbaum, 57–71.
Dordrecht: D. Reidel Publishing Company.
Goldman, Alvin I. “Argumentation and Interpersonal Justification.” Argumentation 11(2):
155–164.
Gottesmann, Claude. 2003. “Each Distinct Type of Mental State Is Supported by Specific Brain
Functions.” In Edward F. Pace-Schott, Mark Solms, Mark Blagrove, and Stevan Harnad
(eds.), Sleep and Dreaming: Scientific Advances and Reconsiderations, 152–154. New
York: Cambridge University Press.
Greenberg, Ramon, Howard Katz, Wynn Schwartz, and Chester Pearlman. 1992. “A Research-
Based Reconsideration of the Psychoanalytic Theory of Dreaming.” Journal of the
American Psychoanalytic Association 40(2): 531–550.
Griffin, Joe, and Ivan Tyrrell. 2007. “The Dream of Irma’s Injection: Freud Reinterpreted.”
http://www.why-we-dream.com/freuddream2.htm.
Grubrich-Simitis, Ilse. 2000. “Metamorphoses of The Interpretation of Dreams: Freud’s Con-
flicted Relations with his Book of the Century.” The International Journal of Psychoanaly-
sis 81(6): 1155–1183.
Grünbaum, Adolf. 1977. “Is Psychoanalysis a Pseudo-Science? Karl Popper versus Sigmund
Freud.” Zeitschrift für Philosophische Forschung 31(3): 333–353.
220 Bibliography
Grünbaum, Adolf. 1979. “Is Freudian Psychoanalytic Theory Pseudo-Scientific by Karl Pop-
per's Criterion of Demarcation?” American Philosophical Quarterly 16(2): 131–141.
Grünbaum, Adolf. 1984. The Foundations of Psychoanalysis: A Philosophical Critique. Berke-
ley, Los Angeles, and London: University of California Press. Citations refer to paperback
edition, first printed 1985.
Grünbaum, Adolf. 1993. Validation in the Clinical Theory of Psychoanalysis: A Study in the
Philosophy of Psychoanalysis. Madison, CT: International Universities Press.
Grünbaum, Adolf. 2005. “Critique of Psychoanalysis.” In Lavinia Gomez (ed.), The Freud
Wars: An Introduction to the Philosophy of Psychoanalysis, 109–137. London and New
York: Routledge. First published in The Freud Encyclopedia, ed. E. Erwin (London and
New York: Routledge, 2002).
Guénolé, Fabian, Geoffrey Marcaggi, and Jean-Marc Baleyte. 2013. “Do Dreams Really Guard
Sleep? Evidence For and Against Freud's Theory of the Basic Function of Dreaming.”
Frontiers in Psychology 4: 17. doi: 10.3389/fpsyg.2013.00017
Hansson, Sven O. 2006. “Falsificationism Falsified.” Foundations of Science 11(3): 275–286.
Harman, Gilbert H. 1965. “The Inference to the Best Explanation.” The Philosophical Review
74(1): 88–95.
Hartmann, E., D. Russ, M. Oldfield, R. Falke, and B. Skoff. 1980. “Dream Content: Effects of
L-DOPA.” Sleep Research 9: 153.
Heath, Sir Thomas L., ed. 1912. The Method of Archimedes, Recently Discovered by Heiberg:
A Supplement to the Works of Archimedes, 1897. New York: Cambridge: University Press.
Hobson, J. Allan. 1988. The Dreaming Brain. New York: Basic Books.
Hobson, J. Allan. 2002. Dreaming: An Introduction to the Science of Sleep. Oxford: Oxford
University Press.
Hobson, J. Allan. 2005. Dreaming: A Very Short Introduction. Oxford: Oxford University
Press.
Hobson, J. Allan. 2009. “REM Sleep and Dreaming: Towards a Theory of Protoconscious-
ness.” Nature Reviews Neuroscience 10(11): 803–813.
Hobson, J. Allan. 2013. “Ego Ergo Sum: Toward a Psychodynamic Neurology.” Contemporary
Psychoanalysis 49(2): 142–164.
Hobson, J. Allan. 2014. “The William James Lectures on Dream Consciousness.” In Nicholas
Tranquillo (ed.), Dream Consciousness: Allan Hobson’s New Approach to the Brain and its
Mind, 3–79. Vienna Circle Institute Library, volume 3. Cham: Springer International Pub-
lishing.
Hobson, J. Allan. 2014. “Response to Commentaries on the William James Lectures on Dream
Consciousness.” In Nicholas Tranquillo (ed.), Dream Consciousness: Allan Hobson’s New
Approach to the Brain and its Mind, 237–256. Vienna Circle Institute Library, volume 3.
Cham: Springer International Publishing.
Hobson, J. Allan, and Karl J. Friston. 2012. “Waking and Dreaming Consciousness: Neurobio-
logical and Functional Considerations.” Progress in Neurobiology 98(1): 82–98.
Hobson, J. Allan, and Robert W. McCarley. 1977. “The Brain as a Dream State Generator: An
Activation-Synthesis Hypothesis of the Dream Process.” American Journal of Psychiatry
134(12): 1335–1348.
Hobson, J. Allan, Edward F. Pace-Schott, and Robert Stickgold. 2003. “Dreaming and the
Brain: Toward a Cognitive Neuroscience of Conscious States.” In Edward F. Pace-Schott,
Mark Solms, Mark Blagrove, and Stevan Harnad (eds.), Sleep and Dreaming: Scientific
Advances and Reconsiderations, 1–50. New York: Cambridge University Press. First pub-
lished in 2000, Behavioral and Brain Sciences 23(6): 793–842.
Hobson, J. Allan, Edward F. Pace-Schott, and Robert Stickgold. 2003. “Dream Science 2000:
A response to commentaries on Dreaming and the brain.” In Edward F. Pace-Schott, Mark
Solms, Mark Blagrove, and Stevan Harnad (eds.), Sleep and Dreaming: Scientific Advances
and Reconsiderations, 231–247. New York: Cambridge University Press.
Hon, Giora, and Sam S. Rakover. 2001. Explanation: Theoretical Approaches and Applica-
tions. Dordrecht: Springer.
Hopkins, Jim. 1991. “The Interpretation of Dreams.” In Jerome Neu (ed.), The Cambridge
Companion to Freud, 86–135. New York: Cambridge University Press.
Bibliography 221
Hopkins, Jim. 1999. “Freud and the Science of Mind.” In Simon Glendinning (ed.), The
Edinburgh Encyclopedia of Continental Philosophy, 377–389. Edinburgh: University Press.
Hopkins, Jim. 2012. “Psychoanalysis, Representation and Neuroscience: The Freudian Uncon-
scious and the Bayesian Brain.” In Aikaterini Fotopoulou, Donald Pfaff, and Martin A.
Conway (eds.), From the Couch to the Lab: Trends in Psychodynamic Neuroscience,
230–265. Oxford: Oxford University Press.
Hull, David L., Peter D. Tessner, and Arthur M. Diamond. 1978. “Planck’s Principle.” Science
202: 717–723.
Jones, Ernest. 1953. Sigmund Freud: Life and Work, vol. 1, The Young Freud 1856–1900.
London: Hogarth Press.
Jones, Ernest. 1955. Sigmund Freud: Life and Work, vol. 2, Years of Maturity 1901–1919.
London: Hogarth Press.
Jones, Ernest. 1957. Sigmund Freud: Life and Work, vol. 3, The Last Phase 1919–1939.
London: Hogarth Press.
Jus, A., K. Jus, J. Gautier, A. Villeneuve, P. Pires, R. Lachance, and R. Villeneuve. 1973.
“Dream Reports after Reserpine in Chronic Lobotomized Schizophrenic Patients.” Vie Méd-
icale au Canada Français 2(9): 843–848.
Jus, A., K. Jus, A. Villeneuve, A. Pires, R. Lachance, J. Fortier, and R. Villeneuve. 1973.
“Studies on Dream Recall in Chronic Schizophrenic Patients after Prefrontal Lobotomy.”
Biological Psychiatry 6: 275–293.
Kächele, Horst, Joseph Schachter, Helmut Thomä, and The Ulm Psychoanalytic Process Re-
search Study Group. 2009. From Psychoanalytic Narrative to Empirical Single Case Re-
search: Implications for Psychoanalytic Practice. Psychoanalytic Inquiry Book Series, vol-
ume 30. New York and London: Routledge.
Kaplan-Solms, Karen, and Mark Solms. 2002. Clinical Studies in Neuro-Psychoanalysis. New
York: Karnac Books.
Karla, R., M. V. Natu, R. S. Deswal, and A. K. Agarwal. 2000. “Dream Features in Psychiatric
Patients on Multiple Psychoactive Drugs.” Human Psychopharmacology: Clinical and Ex-
perimental 15: 525–528.
Köhler, Thomas, and Michael Prinzleve. “Is Forgetting of Dreams Due to Repression? Experi-
mental Investigations Using Free Associations.” Swiss Journal of Psychology 66(1): 33–40.
Kramer, Milton, Roy M. Whitman, Bill J. Baldridge, and Paul H. Ornstein. 1970. “Dream
Content in Male Schizophrenic Patients.” Diseases of the Nervous System 31: 51–58.
Kris, Anton O. 1983. “The Analyst's Conceptual Freedom in the Method of Free Association.”
The International Journal of Psychoanalysis 64: 407–411.
Kumar, Santosh, Subhash Soren, and Suprakash Chaudhury. 2009. “Hallucinations: Etiology
and Clinical Implications.” Industrial Psychiatry Journal 18(2): 119–126.
Lakatos, Imre. 1970. “Falsification and the Methodology of Scientific Research Programmes.”
In Imre Lakatos and Alan Musgrave (eds.), Criticism and the Growth of Knowledge. New
York: Cambridge University Press.
Laplanche, Jean, and Jean-Bertrand Pontalis. The Language of Psycho-Analysis. Translated by
Donald Nicholson-Smith. London: Karnac.
Lena, I., S. Parrot, O. Deschaux, S. Muffat-Joly, V. Sauvinet, B. Renaud, M.-F. Suaud-Chagny,
and C. Gottesmann. 2005. “Variations in Extracellular Levels of Dopamine, Noradrenaline,
Glutamate, and Aspartate Across the Sleep-wake Cycle in the Medial Prefrontal Cortex and
Nucleus Accumbens of Freely Moving Rats.” Journal of Neuroscience Research 81(6):
891–899. doi: 10.1002/jnr.20602
Lipton, Peter. 2001. “Is Explanation a Guide to Inference? A Reply to Wesley C. Salmon.” In
Giora Hon and Sam S. Rakover (eds.), Explanation: Theoretical Approaches and Applica-
tions, 93–120. Dordrecht: Springer.
Lipton, Peter. 2004. Inference to the Best Explanation. London and New York: Routledge. First
published 1991.
Luborsky, Lester, and Paul Crits-Christoph. 1990. Understanding Transference: The Core
Conflictual Relationship Theme Method. New York: Basic Books.
Lynch, Kevin. 2014. “The Vagaries of Psychoanalytic Interpretation: An Investigation into the
Causes of the Consensus Problem in Psychoanalysis.” Philosophia 42(3): 779–799.
222 Bibliography
Macmillan, Malcolm. 1997. The Completed Arc: Freud Evaluated. Cambridge, MA, and Lon-
don: MIT Press. First published 1991.
Magee, Bryan. 1974. Popper. London: Woburn Press.
Magnus, P. D. 2003. “Success, Truth and the Galilean Strategy.” The British Journal for the
Philosophy of Science 54(3): 465–474.
Malamud, William and Forrest E. Linder. 1931. “Dreams and their Relationship to Recent
Impressions.” Archives of Neurology & Psychiatry 25(5): 1081–1099.
Marinelli, Lydia, and Andreas Mayer. 2003. Dreaming by the Book: Freud’s Interpretation of
Dreams and the History of the Psychoanalytic Movement. Translated by Susan Fairfield.
New York: Other Press.
Masson, Jeffrey M., trans. and ed. 1985. The Complete Letters of Sigmund Freud to Wilhelm
Fliess 1887–1904. Cambridge, Mass. and London: The Belknap Press.
Matalon, Nadav. 2011. “The Riddle of Dreams.” Philosophical Psychology 24(4): 517–536.
Mazzoni, Giuliana A. L., Elizabeth F. Loftus, Aaron Seitz, and Steven J. Lynn. 1999. “Chang-
ing Beliefs and Memories through Dream Interpretation.” Applied Cognitive Psychology
13(2): 125–144.
McCarley, Robert W., and J. Allan Hobson. 1977. “The Neurobiological Origins of Psychoana-
lytic Dream Theory.” The American Journal of Psychiatry 134(11): 1211–121.
McNamara, Patrick. 2014. “How Sex Rules our Dreams.” Aeon, April 25. http://aeon.co/maga-
zine/altered-states/was-freud-right-about-dreams-all-along/.
Metzinger, Thomas. 2009. The Ego Tunnel. The Science of the Soul and the Myth of the Self.
New York: Basic Books. Citations refer to paperback edition, first printed 2010.
Michael, Michael. 2008. “On the Validity of Freud’s Dream Interpretations.” Studies in History
and Philosophy of Biological and Biomedical Sciences 39: 52–64.
Midgley, Nick. 2006. “The ‘Inseparable Bond between Cure and Research’: Clinical Case
Study as a Method of Psychoanalytic Inquiry.” Journal of Child Psychotherapy 32(2):
122–147.
Moskovitz, Charlene, Hamilton Moses, and Harold L. Klawans. 1978. “Levodopa-Induced
Psychosis: A Kindling Phenomenon.” American Journal Psychiatry 135(6): 669–675.
Nagera, Humberto, ed. 1969. Basic Psychoanalytic Concepts on the Theory of Dreams, The
Hampstead Clinic Psychoanalytic Library, volume 1. New York: Basic Books.
Nielsen, Tore A. 2000. “A Review of Mentation in REM and NREM Sleep: ‘Covert’ REM
Sleep as a Possible Reconciliation of Two Opposing Models.” Behavioral and Brain Sci-
ences 23(6): 851–866.
Panksepp, Jaak. 1998. Affective Neuroscience: The Foundations of Human and Animal Emo-
tions. Oxford: Oxford University Press. Citations refer to paperback edition, first printed
2005.
Perogamvros, Lampros, and Sophie Schwartz. 2012. “The Roles of the Reward System in
Sleep and Dreaming.” Neuroscience & Biobehavioral Reviews 36(8): 1934–1951.
Popper, Karl R. 1959. The Logic of Scientific Discovery. London: Hutchinson.
Popper, Karl R. 1963. Conjectures and Refutations: The Growth of Scientific Knowledge.
London and New York: Routledge & Kegan Paul. Citations refer to the 2000 reprinting of
the fifth edition (1989).
Popper, Karl R. 1974. “Replies to My Critics.” In Paul A. Schilpp (ed.), The Philosophy of Karl
Popper, vol. 2. La Salle, IL: Open Court.
Popper, Karl R. 1983. Realism and the Aim of Science: From the Postscript to the Logic of
Scientific Discovery. Edited by W. W. Bartley III. London and New York: Routledge.
Citations refer to the 2005 printing of the paperback edition.
Putnam, Hilary. 1974. “The Corroboration of Theories.” In Paul A. Schilpp (ed.), The Philoso-
phy of Karl Popper, vol. 1, 221–240. La Salle, IL: Open Court.
Resnick, Jody, Robert Stickgold, Cynthia D. Rittenhouse, and J. Allan Hobson. 1994. “Self-
Representation and Bizarreness in Children’s Dream Reports Collected in the Home Set-
ting.” Consciousness and Cognition 3(1): 30–45.
Robbins, Paul R., and Roland H. Tanck. 1978. “The Dream Incident Technique as a Measure of
Unresolved Problems.” Journal of Personality Assessment 42(6): 583–591.
Bibliography 223
Rooke, Sally E., Donald W. Hine, and Einar B. Thorsteinsson. 2008. “Implicit Cognition and
Substance Use: A Meta-Analysis.” Addictive Behaviors 33(10): 1314–1328.
Ruse, Michael. 2010. “Evolution and the Idea of Social Progress.” In Denis R. Alexander and
Ronald L. Numbers (eds.), Biology and Ideology from Descartes to Dawkins, 247–275.
Chicago and London: University of Chicago Press.
Ryckman, Nathan A., and Anthony J. Lambert. 2015. “Unsuccessful Suppression is Associated
with Increased Neuroticism, Intrusive Thoughts, and Rumination.” Personality and Individ-
ual Differences 73: 88–91.
Salmon, Wesley C. 2001. “Reflections of a Bashful Bayesian: A Reply to Peter Lipton.” In
Giora Hon and Sam S. Rakover (eds.), Explanation: Theoretical Approaches and Applica-
tions, 121–136. Dordrecht: Springer.
Sand, Rosemarie. 1993. “On a Contribution to a Future Scientific Study of Dream Interpreta-
tion.” In John Earman, Allen I. Janis, Gerald J. Massey, Nicholas Rescher, and Massimo
Pauri (eds.), Philosophical Problems of the Internal and External Worlds: Essays on the
Philosophy of Adolf Grünbaum, 527–546. Pittsburgh: University of Pittsburgh Press.
Sand, Rosemarie. 2012a. “The Unhappy Fate of Freud’s Dream Theory.” In Mark A. Holow-
chak (ed.), Radical Claims in Freudian Psychoanalysis: Point/Counterpoint, 83–95. Lan-
ham, MD: Jason Aronson.
Sand, Rosemarie. 2012b. “Reply to Dr. Michael.” In Mark A. Holowchak (ed.), Radical Claims
in Freudian Psychoanalysis: Point/Counterpoint, 97–101. Lanham, MD: Jason Aronson.
Scarone, S., G. Spoto, G. Penati, R. Canger, and E. A. Moja. 1976. “A Study of the EEG Sleep
Patterns and the Sleep and Dream Experience of a Group of Schizophrenic Patients Treated
with Sulpiride.” Arzneimittel-Forschung 26(8): 1626–1628.
Schnabel, Konrad, and Jens B. Asendorpf. 2013. “Free Association as a Measure of Stable
Implicit Attitudes.” European Journal of Personality 27: 39–50.
Schönhammer, Rainer. 2005. “‘Typical Dreams’: Reflections of Arousal.” Journal of Con-
sciousness Studies 12(4–5): 18–27.
Schur, Max. 1972. Freud: Living and Dying. London: Hogarth Press and The Institute of
Psych-Analysis.
Searle, John R. 1998. Mind, Language and Society: Philosophy in the Real World. New York:
Basic books.
Seitz, Philip F. D. 1966. “The Consensus Problem in Psychoanalytic Research.” In L. A.
Gottschalk and A. H. Auerbach (eds.), Methods of Research in Psychotherapy, 209–225.
New York: Appleton-Century-Crofts.
Smith, Robert C. 1986. “Studying the Meaning of Dreams: Accurate Definition of the Indepen-
dent Variable.” Biological Psychiatry 21(11): 989–996.
Snyder, F. 1970. “The Phenomenology of Dreaming.” In L. Madow and L. Snow (eds.), The
Psychodynamic Implications of the Physiological Studies on Dreams, 124–151. Springfield,
IL: C.C. Thomas.
Solms, Mark. 1995. “New Findings on the Neurological Organization of Dreaming: Implica-
tions for Psychoanalysis.” The Psychoanalytic Quarterly 64: 43–67.
Solms, Mark. 2000. “Dreaming and REM Sleep are Controlled by Different Brain Mecha-
nisms.” Behavioral and Brain Sciences 23(6): 793–1121.
Solms, Mark. 2003. “Forebrain mechanisms of dreaming are activated from a variety of
sources.” In Edward F. Pace-Schott, Mark Solms, Mark Blagrove, and Stevan Harnad (eds.),
Sleep and Dreaming: Scientific Advances and Reconsiderations, 247–252. New York: Cam-
bridge University Press.
Solms, Mark. 2013. “Freud’s ‘Primary Process’ versus Hobson’s ‘Protoconsciousness.’” Con-
temporary Psychoanalysis 49(2): 201–208.
Solms, Mark. 2014. “Do You Agree that Freud’s Theory Was Erroneous with Respect to (1)
the Genesis of Dreaming (Release of Repressed Infantile Wishes) and (2) the Bizarreness of
Dreams (Disguise and Censorship of Unacceptable Wishes)?” In Nicholas Tranquillo (ed.),
Dream Consciousness: Allan Hobson’s New Approach to the Brain and its Mind, 205–209.
Vienna Circle Institute Library, volume 3. Cham: Springer International Publishing.
Solms, Mark, and Oliver Turnbull. 2002. The Brain and the Inner World. New York: Other
Press.
224 Bibliography
Spanjaard, Jacob. 1993. “The Manifest Dream Content and its Significance for the Interpreta-
tion of Dreams.” In Sara Flanders (ed.), The Dream Discourse Today, 153–180. The New
Library of Psychoanalysis, volume 17. London: Routledge.
Strauch, Inge, and Barbara Meier. 1996. In Search of Dreams: Results of Experimental Dream
Research. New York: SUNY Press.
Sturrock, Peter A. 1994. “Applied Scientific Inference.” Journal of Scientific Exploration 8(4):
491–508.
Sulloway, Frank J. 1992. Freud, Biologist of the Mind: Beyond the Psychoanalytic Legend.
Cambridge, MA, and London: Harvard University Press.
Szentagotai, Aurora, and Diana Onea. 2007. “Is Repressive Coping Associated With Suppres-
sion?” Journal of Cognitive and Behavioral Psychotherapies 7(2): 127–138.
Timpanaro, Sebastiano. 1976. The Freudian Slip: Psychoanalysis and Textual Criticism.
Translated by Kate Soper. London: NLB. Original work published 1974.
Tuckett, David. 1993. “Some Thoughts on the Presentation and Discussion of the Clinical
Material of Psychoanalysis.” The International Journal of Psychoanalysis 74: 1175–1189.
Van de Castle, Robert L. 1994. Our Dreaming Mind. New York: Ballantine Books.
Van Helden, Albert. 1974. “The Telescope in the Seventeenth Century.” Isis 65(1): 38–58.
Vogel, Gerald W. 1978. “An Alternative View of the Neurobiology of Dreaming.” American
Journal of Psychiatry 135(12): 1531–1535.
Wallerstein, Robert S., and Harold Sampson. 1971. “Issues in Research in the Psychoanalytic
Process.” The International Journal of Psychoanalysis 52(1): 11–50.
Walton, Douglas N. 1988. “Burden of Proof.” Argumentation 2 (2): 233–254.
Wegner, D. M., R. M. Wenzlaff, and M. Kozak. 2004. “Dream Rebound: The Return of
Suppressed Thoughts in Dreams.” Psychological Science 15(4): 232–236.
Weiss, Joseph. 1992. “Dreams and their Various Purposes.” In Melvin R. Lansky (ed.), Essen-
tial Papers on Dreams, 213–235. New York and London: New York University Press.
Wells, Frederic L. 1912. “Critique of Impure Reason.” The Journal of Abnormal Psychology 7:
89–93.
Wells, Frederic L. 1913. “On Formulation in Psychoanalysis.” The Journal of Abnormal
Psychology 8: 217–227.
Westen, Drew. 1998. “The Scientific Legacy of Sigmund Freud: Toward a Psychodynamically
Informed Psychological Science.” Psychological Bulletin 124(3): 333–371.
Westman, Robert S. 1986. “The Copernicans and the Churches.” In David Lindberg and Ro-
nald L. Numbers (eds.), God and Nature: Historical Essays on the Encounter between
Christianity and Science, 76–113. Berkeley and London: University of California Press.
Wittgenstein, Ludwig. 1966. Lectures and Conversations on Aesthetics, Psychology, and Relig-
ious Belief. Edited by Cyril Barrett. Malden: Blackwell. Citations refer to 2003 reprint.
Wohlgemuth, Adolf. 1923. A Critical Examination of Psycho-Analysis. London: G. Allen &
Unwin.
Wollheim, Richard. 1993. The Mind and Its Depths. Cambridge, MA, and London: Harvard
University Press. Citations refer to paperback edition, first printed 1994.
Woodworth, Robert S. 1917. “Some Criticisms of the Freudian Psychology.” The Journal of
Abnormal Psychology 12: 174–194.
Yu, Calvin K.-C. 2007. Brain Mechanisms of Dreaming. PhD dissertation, University of Cape
Town.
Ziman, John M. 1978. Reliable Knowledge: An Explanation of the Grounds for Belief in
Science. New York: Cambridge University Press. Citations refer to 1996 reprint of the
Canto edition.
Zubin, Joseph. 1964. “Criteria for the Evaluation of the Results of Psychotherapy.” American
Journal of Psychotherapy 18: 138–144.
Index
225
226 Index
Occam’s razor, 48–49 repression, 17, 21, 32, 57, 60–61, 131–136,
Oedipus complex, 47–48, 140n8 160, 180, 201, 207, 214
On Dreams, xx research program, 117, 140n1
Otto, xxi, xxiii. See also Otto’s illness, resistance, 17, 19–21, 22, 101, 135, 159,
dream of 176, 214
Otto’s illness, dream of, 76, 90, 103 revision of dream theory, Freud’s, 44–46,
An Outline of Psycho-Analysis, 192 127
over-determination, 46–49, 146, 191 Ricoeur, Paul. See hermeneutic
rigor. See scientific standards
Panksepp, Jaak, 166, 177 royal road, xiii, xxv
personal justification, 74–75, 96, 97, Royal Society, 151
159–160 Ruse, Michael, 117
philosopher’s stone, xi–xii, 215
philosophy of science. See history and Salmon, Wesley, 136–137
philosophy of science schizophrenics, 198–199
placebo effect, 133–134, 135 scientific revolution, xi
pontine reticular formation, 169 scientific standards, xiii, 62, 118, 130, 153,
Popper, Karl, xii, xv, 118, 119–130, 139, 159, 213; reliable epistemic enterprise,
144, 147, 167, 209n10, 212 144; rigorous practice, 144–150, 161n9
Popper-Lynkeus, Josef, xviii, 27–28 scientistic, xvi
posttraumatic stress disorder dreams. See secondary process, 26n9, 36, 63n20, 181,
traumatic dreams 215
preconscious, 31, 32–34, 40, 41–42, 47, secondary revision, 10, 16, 53, 192
51–53, 54, 191–192, 208n5 SEEKING system, 177, 179, 195, 197,
primary process, 26n9 199, 206, 214
principle of charity, 4–5, 76 selection bias, 148
principle of constancy, 35, 38, 215 self-analysis, xix, 97, 203
private data, problem of, 147–150, 151, sleep preservation. See function of dreams
161n6 slips of the tongue, xvii, 41
problem-solving function of dreams, Solms, Mark, xv, 119, 165, 166, 177–178,
192–193, 206 179, 181–182, 195, 196–198, 199–200
professor, xxi, xxvin2, 11–12, 19, 29–30 somatic sources of dreams, 51, 172
Project for a Scientific Psychology, xiv, 37 specimen dream. See Irma’s injection,
pseudoscience, xv, 117–118, 120, 139 dream of
psychosis, xviii, 36 straightforward dreams. See dreams of
psychotherapy. See therapy convenience
PTSD dreams. See traumatic dreams strikingness. See felt significance
punishment dreams, 43, 63n14, 195 structural model, 62n1, 191
Putnam, Hilary, 122 Studies on Hysteria, xviii, 131
subjectivity, 96
reality-testing, 35, 39, 181, 197 suggestion, 97, 97–98, 148, 149, 159–160,
recurring dreams, 178 206
reflex apparatus, 35, 37, 38, 44, 59, 215 Sulloway, Frank, xix, 146
regression, xix, 15, 52, 63n8, 184n7 superego, 62n1, 63n5, 189, 191, 206, 214
REM, xv, 33, 166, 167, 169–170, 171–172, supper-party dream, 43, 70–71, 77–79, 81,
176, 176–178, 183, 184n1, 198, 200, 88, 93, 103, 148, 149, 156
206, 209n12 symbolization, xix, 15, 36, 170, 175
representability, considerations of. See
dramatization talking cure, 132
Index 229
unconscious, 31–36, 44, 47, 51–52, 119, yellow beard, dream of, 11–13, 19, 29, 103
135, 181
unfalsifiability. See falsifiability
About the Author
231