The Grudge Informer
The Grudge Informer
David Dyzenhaus1
INTRODUCTION
One of the points of contention in HLA Harts and Lon L. Fullers 1958 debate2 was a
decision by a German postwar court, the Case of the Grudge Informer. Even those who
have forgotten the details of the debate usually remember the drama of this case. However,
the exchange between Hart and Fuller about it garnered little attention.3
This is a pity, as Harts treatment of the case is a rare example of engagement with an
actual problem of adjudicationit shows, contrary to received wisdom, that he and Fuller
were engaged in a real debate. As I will argue, it is a debate about judicial obligation in light
of a connection between what I call the doctrinal and the fundamental levels--the level of
judicial interpretation of legal doctrine and the level of judicial interpretation of what legality
itself requires. Indeed, I hope to show that the exchange supports Fullers claim, one which
has puzzled readers for fifty years, that Harts argument brought within the dispute between
natural law and positivism the issue of fidelity to law.4 I will also show that this connection
helps to advance the broader issues of legal philosophy involved in the later Hart-Dworkin
debate, one that has tended to eclipse the Hart-Fuller exchange, and which is often thought
also not to display real engagement. I will begin with Hart and then move through an
analysis of the case and Fullers response to Hart to the broader issues of legal philosophy.
1
According to Hart, a German court had in 1949 to decide a case where a woman was
prosecuted for the offence of illegally depriving her husband of his liberty a crime
punishable under the German Code of 1871, which had remained in force during the Nazi
era.5 In 1944, she had denounced her husband to the authorities for insulting remarks he had
made about Hitler while on leave from the army. She wanted to get rid of him because she
was having an affair. Under Nazi statutes, it was apparently,6 Hart says, illegal to make
such remarks, though the wife was under no legal duty to report him. The husband was
found guilty and sentenced to death, though it seems that he was not executed but sent to
the front.
The wifes defence was that she had acted in accordance with the law the statutes -
and so had not committed any crime. But the Court of Appeal, despite the fact that the
husband had been sentenced by a court for having violated a statute, found her guilty of
the offence of deprivation of liberty, because quoting from the judgment the statutes
were contrary to the sound conscience and sense of justice of all decent human beings.7
Hart reports that the reasoning was followed in many cases, and these were hailed as a
triumph of the doctrines of natural law and as signaling the overthrow of legal positivism.
But, he retorts, the unqualified satisfaction with this result seems to me to be hysteria.8
Harts point is that even if one applauds the objective of punishing the woman for
an outrageously immoral act, one should see that to achieve this a statute established
since 1934 had to be declared not to have the force of law and, Hart argues, the wisdom
of this course must be doubted. There were two other choices available to postwar
with a full consciousness of what was sacrificed in securing her punishment in this way. He
comments:
2
Odious as retrospective criminal legislation and punishment may be, to have pursued
it openly in this case would at least have had the merits of candour. It would have
made plain that in punishing the woman a choice had to be made between two evils,
that of leaving her unpunished and that of sacrificing a very precious principle of
morality endorsed by most legal systems. Surely if we have learned anything from
the history of morals it is that the thing to do with a moral quandary is not to hide it.
describe the treatment of the dilemma as if it were the disposition of the ordinary
case.9
Hart emphasizes that it is not a mere matter of form whether one leaves it to a court to
invalidate the statute in the way it did, that is, by pretending that it was merely interpreting
the law with no sacrifice of principle, or one requires that a statute be invalidated by a
retrospective statute. For if we adopt the Courts course, and assert that certain rules
cannot be law because of their moral iniquity, we confuse one of the most powerful,
because it is the simplest, forms of moral criticism. Rather, we should speak plainly and
Hart took this point to be a knock down one, the truly liberal answer, to Gustav
Radbruchs famous claim that legal positivism contributed to the failure of lawyers in prewar
Germany to respond adequately to the Nazis abuse of the legal order. . In his reaction to
this failure, Radbruch had concluded in short articles in 1945 and 1946,11 that one should
adopt the view that extreme injustice is no law. Thus, statutes lack the force of law when
they contravene fundamental principles of morality and should not be taken into account in
working out the legal position of any given individual in particular circumstances.12
3
Hart accuses Radbruch of extraordinary naivety for supposing that
insensitiveness to the demands of morality and subservience to state power in a people like
the Germans should have arisen from the belief that law might be law though it failed to
conform with the minimum requirements of morality. Hart recognises that the positivist
slogan law is law might have had a different history in Germany, acquiring a sinister
characteristic in contrast to its English history, where it went along with the most
enlightened liberal attitudes. But even if that is the case, Radbruch had latent in his
whole presentation of the issues to which the existence of morally iniquitous laws can give
rise something more disturbing than naivety. For he had only half digested the spiritual
Radbruch says, according to Hart, depends on an enormous overvaluation of the bare fact
that a rule may be said to be a valid rule of law, as if this, once declared, was conclusive of
the final question: Ought this rule of law to be obeyed?.13 Instead, one should adopt the
truly liberal answer and not let the fact that X is the law determine the issue whether X
should be obeyed.
Hart was well aware that the harshness of his own judgments could only be
accentuated by the fact that Radbruch and others, like Ulysses or Dante, testified from the
experience of a descent into Hell, from which they brought a message for human beings.
He was also aware that Radbruchs criticism of positivism involved an exercise of self-
criticism, for Radbruch, on his own account, had before the Nazi advent to power put
forward a basically positivist view of law.14 However, it was precisely the power of the
experience from which Radbruch spoke that bothered Hart, because that made Radbruchs
4
For Hart, the only way to avoid talking stark nonsense16 is to adopt the view of his
positivist predecessors, Jeremy Bentham and John Austin, and see that the question of the
validity of particular laws does not depend on their moral content. Rather, if laws reached a
certain degree of iniquity then there [is] a plain moral obligation to resist them and to
withhold obedience. He quotes with approval Austins example of the man who is
convicted of a crime punishable by death when the act he did was in fact trivial or even
beneficial. The man objects to the sentence that it is contrary to the law of God, but the
hanging [him] up, in pursuance of the law of which [he had] impugned the validity.17
Shortly after the 1958 debate between Hart and Fuller, HO Pappe pointed out that
the case on which they focused had been misrepresented. In fact, Pappe explained, the Court
and other courts which dealt with either the same or similar issues explicitly did not adopt a
higher law argument, preferring to focus on interpretation of the law. They concentrated
on matters such as the absence of a duty to inform and the privacy in which the remarks had
been made in order to sustain the conclusion that there had been an illegal deprivation of
Hart added a footnote to the reprinted version of Positivism and the Separation of
Law and Morals in which he summarised the respects in which, on Pappes account, he had
the facts of the case wrong, most pertinently because the Court,
after accepting the theoretical possibility that statutes might be invalid if in conflict
with natural law held that the Nazi statutes in question could not be held to violate it;
the accused was guilty of unlawfully depriving her husband of liberty, since she had
no duty to inform against him but did so for purely personal reasons and must have
5
realised that to do so was in the circumstances contrary to the sound conscience
Hart said that Pappes careful analysis should be studied. But he did not seem to
think it had implications for his own account of what we can think of as the dilemma of
legalitythe problem that judges, lawyers and those subject to the law face when the law is
used as an instrument of injustice. For he said that the case as he had understood it could be
This is an inadequate response. It shows that Hart failed to appreciate that the
problems Pappe exposed went well beyond the facts of a hypothetical case. Pappe did not
merely correct Harts account of the case but made profound jurisprudential points which
As we will see, Harts method of candour obscures not only the judges situation, but
also the citizens. In addition, because Hart does not attend to the problem of interpretation
faced by judges in a wicked legal system, he cannot appreciate how cases like that of the
Grudge Informer pose for judges the question of how to maintain the legal order legally in
good shape.21 They do so because they form a very important sub-category within the
category Dworkin terms hard cases. They are fundamental cases, where lawyers reasonably
disagree about the appropriate outcome, not only because they disagree about legal doctrine,
but also because their disagreement is influenced by their views about the very foundations
of legal order.
FUNDAMENTAL CASES
6
In the actual Case of the Grudge Informer, the Bamberg Court of Appeal overturned the
decision of the trial court. The latter found that the informer had not illegally deprived her
husband of his liberty because her report and the subsequent detention of her husband came
about because he had violated a valid law and through a properly carried out judicial
process.22 This decision, the Court of Appeal held, erred in that it inferred the legality of the
informers report from the legality of the court-martial which found the husband guilty. The
court of first instance thus failed to appreciate that the accused had used the court-martial as
a mere instrument to bring the criminal act. Her act was thus one of indirect perpetration.
But the act of the instrument which directly brought about the illegal resultthe court-
martial which found the husband guilty--was an act which simply interpreted and applied the
The salient difference, for the Court of Appeal, was that the accused was not under
any legal duty to bring about the result, whereas once her report was brought to the attention
of the court-martial, it was under a duty to decide as it did. The court-martial was simply
applying the positive law as it was intended to be applied. She knew, the Court of Appeal
reasoned, that her report would lead inexorably to a certain range of results, the least
harmful of which was a years imprisonment, the worst death. She thus deliberately made use
of laws that most Germans knew were designed to terrorize the population and which many,
The Court emphasized that its reasoning did not depend on a claim that the Nazi
statutes in issue violated the laws of nature. The laws, while grossly unfair, did not reach that
point because they did not command any positive conduct, only an omissionmaking no
public remarks of a certain sort. The Court did, however, imply that if the statutes had, for
example, commanded that people do something immoral and made it an offence not to do
7
so, it would have found that a court-martial that caused the detention of an individual who
had failed to carry out this positive duty would itself act illegally.
It is clear that the Court was determined to reach a conclusion that the woman was
guilty without incriminating the judges who were her instrument and its determination in this
regard gives its reasons a rather contrived quality. But while contrived, the reasoning is
distinctly legal, by which I mean that all the reasons the Court offers are legal reasons, and it
organizes those reasons into a chain of justification for the result it wished to reach.
Moreover, the legal reasons operate at two rather distinct levelsthe fundamental level of
reasoning about legal order and the role of judges in maintaining it and the doctrinal level at
At the fundamental level, the Court faces the question about what to make of judges
who are under a duty to interpret positive laws that are morally obnoxious but, in its view,
not so obnoxious that one can make a natural law argument that they are invalid. At the
doctrinal level, the Court has to consider how its sense of an answer to the first question
meshes with its sense of how best to interpret the substantive law. Assume that at both
levels judges regard themselves as under a duty to find an answer that coheres with the
animating principles of that level, respectively, the fundamental principles of legal order and
the principles of the substantive body of law. Because the Grudge Informer Case requires
the judge to give an answer at both levels, it is a special sub-category within the general
In this fundamental case, the Court assumes that judges are in a different position
from citizens when the law requires a morally obnoxious result but there is no legal duty on
the part of the citizen to bring that result to the attention of the state. That is because judges
8
are in virtue of their officetheir submission to authorityrequired to apply the law as it
The Court thus assumes that the judges on the court-martial had no option but to
decide as they did, and that is what marks the difference between the judges and the
informer. But it follows that if the statutes had imposed a duty on citizens to report such
remarks, the informer would also be able to rely on the statutes to absolve her of the crime
of illegal deprivation of liberty. Thus, the Courts distinction between the role of the judge
and the role of the citizen turns out, despite some suggestions to the contrary, to depend not
on any special judicial responsibilities but on contingent facts about the meaning of the
statutes.
The Court does enter a reservation, however, which is exactly Radbruchs. If these
statutes had reached a pitch of injustice sufficient to violate the laws of nature, the court-
martial judges would be guilty of a crime, as by implication the informer would be even if the
There are two problems with the Courts reasoning which, in conjunction, create a
problem for Hart. First, the Courts stance on the fundamental level means that in order to
get to the result it wanted it had to resort to a criminal doctrine in a way that not only looks
too contrived, but is also suspect as a piece of reasoning about the criminal law.23 In other
words, its stance at the fundamental level introduced distortions at the doctrinal level.
Hart would surely regard this problem as further evidence in favour of his proposal
that in such cases candour demands that one refrain from smuggling ones desire to have the
informer punished into claims about what the law requires, whether one does this through
the Radbruch Formula that extreme injustice is no law or through bending the doctrines
of the positive law. Indeed, it seems to show that the hypothetical case Hart considered was
9
in fact not very distant from the actual case. Thus he could say that the superior legal
solution if one resolved the moral dilemma in favour of punishing the informer was a statute
However, the second problem is one which affects Harts position and the Courts
with equal force and, as we will see, it makes the first problem one for Hart as well. It results
from the assumption that the court-martial judges were under a legal duty to decide as they
did, which left postwar courts with the option of, on the one hand, letting the informer go
unpunished or, on the other hand, if they were determined to find the informer guilty,
In fact, postwar courts faced with grudge informer cases were not confined to these
options. They could get to the same conclusion as the Bamberg Court of Appeal through
finding the informers guilty of complicity in illegal deprivation of liberty, though that
required also finding that the court-martial judges were culpablenot a mere instrument,
but the direct perpetrator. This option was adopted in 1952 by the Federal Supreme Court in
a very similar case, in which the Court quashed an acquittal of an informer by a lower court,
and referred the matter back for decision in light of its view of the law.24
As Pappe explains, the Federal Supreme Court found that the informer could not be
found guilty if the court-martial had acted legally, which it had not. There was no need to
rely, however, on a claim that the Nazi statute was incompatible with supra-positive law, as
the court-martial had not properly considered the statute or the established principles of
First, in regard to the statute, the Court emphasized that the remarks had to be made
in public. While superior courts during the Nazi era had developed a very wide interpretation
of publicity, to the point where the public included the smallest and most intimate circle,
10
there remained in place a requirement relating to the expectation that the remarks could be
repeated elsewhere. The Court thus held that the husband would have had to have had mens
rea in regard to a possible breach of confidence, which could not be assumed absent special
circumstances in the context of the relationship of extreme intimacy between husband and
wife. The court-martial thus derived its finding from an arbitrary and unlawful interpretation
of the law.
Second, the death sentence violated established principles of German criminal law.
The court-martial had a great deal of discretion when it came to sentence and a death
sentence was clearly disproportional given that the remarks were made between spouses and
that the interest supposedly protected by the statute was the military morale of the German
people, which was at most minimally impacted by the remarks. The court-martial decided
then, not on the basis of the positive law, but in response to administrative pressure to
Third, the accused was, as a result, an accessory before the fact to the crime of
unlawful deprivation of liberty and of attempted homicide. She was guilty because, fourth,
she had the intention to get rid of her husband by using the unlawful procedure of the court-
martial. The jury in the trial court had accepted her defence that she trusted in the legality of
the court-martial and, as a woman without higher education, she could not be expected to
recognise violations of the rule of law. However, the Court held that such a defence was
available only in normal times, when independent judges guarantee the liberty and dignity of
the individual, and the Nazi era was very different, since court practice often failed to deliver
This insight was well within the ken of an ordinary member of the public. There was,
despite the warped judgment of wide circles during the Nazi period, a lively
11
awareness in the population that administrative and legal authority could be abused
which served the purpose of political terror rather than the realisation of the law, had
actually led to a heightened popular sense of right and wrong rather than stifling it.
knowledge; it had also been found to be correct by the jury. A mistaken belief in the
legality of the court-martial procedure would be a defence only for a person who
could not be expected to share the insights of ordinary members of the public.
However, the facts of the case, as found by the jury, indicated that the accused, far
from acting to reveal the crime, just wished to make use of the best means to get rid
of her husband so that she could continue her adulterous way of life.25
Pappe went on to outline the ways in which common law judges could arrive at the
same result as the Federal Supreme Court, using the ordinary resources of their legal
orders.26 This exercise underlines his emphasis on the fact that the Court, though using a
different route from that of the Bamberg Court of Appeal, reached its conclusions without
having to rely on allegedly supra-positive legal principles. And that prepares the way for his
profound observation about a little noticed aspect of Harts 1958 article--that Hart relied on
the Case of the Grudge Informer in the context of his argument about judicial interpretation
of particular laws.
Because, says Hart, men were sentenced to death for criticism of the Nazi regime,
one should see that the choice of sentence might be guided exclusively by consideration of
what was needed to maintain the states tyranny effectively. A decision on these grounds
would be intelligent and purposive, and from one point of view the decision would be as it
ought to be. But, Hart asserts, that ought is immoral and thus we know that it cannot
12
sustain a claim that the distinction is false between law as it is and law as morally it ought to
be.27
Pappe observes that to regard such intimidation in the interests of the ruling party
as the declared purpose of the German criminal law is an arbitrary assumption. Not
only was the 1871 law on illegal deprivation of liberty in existence, but also the large part of
pre-Nazi law. Moreover, the judges were supposed to be independent and they had to
choose between applying the criminal law as established by statute, interpretation and
precedent, and, on the other hand, obeying the shifting and often contradictory
administrative directives of the government of the day. Finally, it has to be taken into
account that a dictatorial government can of course make laws to achieve its ends, but
cannot change the body of preexisting law wholesale. What it really desires is the ability to
decide arbitrarily. But it is unlikely to reveal its character by enacting a statute declaring that
courts should follow government orders regarding the definition and punishment of
crime, since such a statute would obviously not be law, even in Harts sense. It is thus
unsurprising, Pappe concludes, that the postwar courts should have concentrated on
questions of judicial interpretation and procedure rather than on that of the validity of
statutory law.28
these remarks of Hart were made in the course of his argument against the Legal Realist
claim that one can demonstrate a connection between law and morality from the fact that
when judges interpret particular laws they must rely on their own moral convictions. Hart
argued that this claim was far from impugning the positivist Separation Thesis that there is
no necessary connection between law and morality. First, reliance on values is not equivalent
to reliance on morality; second, in the hard cases on which the Realists rely there was often
13
even usually no answer determined by law. In other words, at most the Realists had shown a
necessary connection between law and value, not law and morality. But in fact they had
failed to show even this, since in the cases where judges had to rely on value, the reason for
such reliance was often that the law could not be said to supply an answer. The judge is in
the penumbra of personal choice or discretion, rather than in the core of legal certainty.29
Harts claim that from one point of view the decision would be as it ought to be
could thus mean two different things. He could mean that the court-martials decision was
value mandated by the legal order for interpreting this kind of statute. However, that the
value is morally obnoxious shows that there is no necessary connection between law and
morality. Alternatively, he could mean that the court-martial had to exercise a discretion,
infer a connection between law and morality from this because of the obnoxious nature of
the value. The second option is perhaps more likely,30 but it throws into doubt the
assumption Hart shares with the Bamberg Court of Appeal that the law required the court-
In addition, even if one were to accept that the best way to characterize the court-
martials situation is to say that its judges had a discretion, it seems, as Pappe points out,
odd to suppose that the value of intimidation in the interests of the regime is one a judge
may adopt and still be recognized as performing his office. The issue here is not simply that
the value is morally obnoxious. It is that the value seems radically inconsistent with the point
of view of legal order, which I think we can assume is the point of view to which Hart refers
when discussing the Legal Realists. It cannot, that is, be justified from the perspective of a
14
judge who regards it as part of his role to maintain legal order in good shape legally speaking,
Hart never says in so many words that judges have a duty to maintain their legal
orders in good shape legally speaking. But his later account of the internal point of view
which judges have to adopt in order for legal order to function at all surely amounts at the
least to such a claim. Since that account takes place at what I called earlier the fundamental
level, it has profound implications for Harts legal theory, for, in particular, his assertion that
his theory illuminates the predicament of individuals faced with unjust laws since it tells
them that the certification of law as valid is not conclusive of the question of obedience. For,
when more closely examined in light of the Case of the Grudge Informer, the assertion
As we know from The Concept of Law,32 Hart does not think that what judges do in
regard to the fundamental level can be captured in the language of obedience to commands.
Judicial duty has nothing to do with obedience and everything to do with judges accepting
that their rule of recognitionthe fundamental rule of legal order-- provides them with a
public, common standard of correct judicial decision.33 So if a rule is picked out as valid by
the rule of recognition, the judge has, from the point of view of legal order, to apply it. It is
hardly stretching Harts analysis, despite his claim that he was engaged in an exercise in
descriptive sociology,34 to say that the judge is under a duty to apply the rule, because that is
It follows for Hart that the court-martial judges were either under a duty from the
point of view of legal order to find the husband guilty or, from that same point of view,
could legitimately exercise their discretion by finding him guilty. Of course, Hart can
emphasize that the duty, like the possible exercise of discretion, has to be weighed against
15
the option of not finding the husband guilty, and it might seem obvious that there is no
contest here. But in either case Harts account leads to obscurity, not light.
In the first place, for a judge to decide not to apply a rule merely because it is
immoral is to cease being a judge. For Hart, the fundamental rule of legal order depends on
the judge continuing with the practice of applying valid rules. The judge is thus faced with a
situation far more complex than can be captured by what we can think of as the Hartian
formula: citizens should disobey extremely unjust laws. Moreover, it would be somewhat
contradictory for Hart to suppose that a Nazi-era judge should morally speaking have not
applied morally obnoxious law, but also that this same stance was morally inappropriate for
postwar judges.
Matters become even more puzzling if the situation is one which requires an exercise
of discretion. It unclear why Hart supposes that there was any sort of ought in play that
should inform such an exercise. For a judge to decide as he did either because he wanted to
help the Nazi state in its program of intimidation or because he was himself intimidated into
doing so is also to cease to be a judge. To find the husband guilty was, as the Federal
Supreme Court reasoned, to turn the office of judge into a cog in the machinery of a vicious
administrative regime and not to play a role in maintaining the Rechtsstaat. That is, the issue
was not merely that that finding was immoral, but that it undermined what we may think of
legal order. His idea of legalityof the normative structure of legal orderis that
normativity is what it takes to maintain a system of valid rules. But his analysis of the Grudge
Informer Case shows that, at least when it comes to the officials charged with the
maintenance of the most fundamental rule presupposed by his idea of legal order, his idea
16
obscures rather than illuminates the officials position when they are faced with an unjust
law.
Further, as I have already suggested, Harts idea is no less obscuring when applied to
the situation of the postwar courts. That the Bamberg Court of Appeal and the Federal
Supreme Court got to the same result by different doctrinal routes could, I suppose, be
thought to be evidence of the fact that the judges were determined to reach a result that was
not determined by the law, and so were exercising a discretion in Harts sense. But much
more plausible is that both Courts thought it imperative for legalitys sake to demonstrate to
the German people that there is a difference between legality and the Nazi administrative
regime. The Court of Appeal has much more difficulty in this regard. Its idea of legality,
which seems to have much in common with Harts, suggests that the court-martial judges
were faced with a stark choice between legal and moral duty, but not so stark as to overcome
their legal duty. And that in turn led to a less satisfactory piece of reasoning at the doctrinal
level.
In contrast, the Federal Supreme Court has far less difficulty, because its idea of
legality, implicit in its reasoning at the fundamental level, combines two elements: first, an
a wide view of the relevant positive law and to seek to reach conclusions that, to use
Dworkinian language, shows the law in its best light; second, an understanding of legal order
that is committed to an idea of legality that consists of principles, including the principle that
a court that decides matters that affect significant individual interests is radically different
from an arm of executive government that exists simply to carry out the orders of higher
officials.
17
This second component is itself justified by its service to the interests of the
individual and it may be that one could convincingly argue that the two components form
part of an overarching theory of legality. However, I wish to make a more limited point. It is
that the court-martial judges failed legality in respect of both components, while the
informer failed in that she turned her husband over to a body counting on the fact that it
would act, not as the court of law it was presented as being, but as a cog in a vicious
administrative regime.
FULLERS RESPONSE
In contrast to Hart, Fullers 1958 essay provided a theoretical framework sensitive to the
complexities of the situations faced by both the Nazi-era and the postwar courts. In Fullers
view, the debate between contending positions in legal theory was best understood as one
about different ideals of fidelity to law at the fundamental level. At that level, one had to take
into account what Fuller called the internal morality of law or inner morality of law.35
He argued that attention to the internal morality would have helped German lawyers before
the war maintain their fidelity to the ideal of law. After the war, German judges could have
dealt better with cases like that of the informer, had they focused on the deterioration of
legality. Thus, the debate at the fundamental level was one which had direct practical
implications, not least in that a judge faced with a certain kind of case would also be faced
with a choice between these ideals. Such a choice does not amount to an exercise of
discretion determined by factors external to law, since it is a choice about how best to
18
It is because Fuller sees that both levels are in play and that there is a direct
connection between them in judicial interpretation that he also notices both that the Nazi
statutes could not be said to compel the result that the husband was guilty and the
implications of this fact for the informers defense that she acted in accordance with the law.
In contrast, Hart, in analyzing the case, not only did not see that its characterization was a
matter of interpretation that had in part to take into account the interpretive complexity of
the situation of a judge, whether during the Nazi era or after the war. He also could not
appreciate such complexity, since for him issues of interpretation do not fall within the
scope of descriptive legal theory; rather, they are fit topics for normative political theory.36
In this regard, Fuller rejected Harts statement of the dilemma the postwar courts
faced because its solution say that the law is too evil to be obeyed is wrong-headed, at
least if it is offered to judges, since when a court refuses to apply something it admits to be
law moral confusion reaches its height. As Fuller recognized, Hart did not mean this
solution to be deployed by a court but by a legislature. However, Hart does not, Fuller says,
take into account the situation of drastic emergency in which the courts and Radbruch
allow the people to begin taking the law into their own hands, as might have occurred while
the courts were waiting for a statute.37 Germany had to restore both respect for law and
respect for justice and the attempt to restore both at once necessarily created painful
antimonies.38
Even more important is that the Grudge Informer Case could come into Harts
view only as one in which there is a stark confrontation between justice and valid law. In
this respect, there is little difference between his and Radbruchs positions. Indeed, Fuller
thought that the German indifference to the internal morality of law persisted in Radbruchs
19
new position after the war. For Radbruchs resort to notions of higher law a morality
that transcends positive law but which functions as a test for the validity of law may itself,
Fuller says, be a belated fruit of German legal positivism. Someone with a positivist
mindset might feel that the only way to escape one law is to set another off against it, and
there is a deep similarity between their positions. Both Hart and Radbruch resort to the idea
of a higher moral law in order to deal with the problems created by past legal injustice. For
them the higher law is a law which has the power or force to invalidate another law. Thus
they both in fact prefer the legal solution to come in the form of a frankly retroactive ruling.
The only difference is that Radbruch is prepared to allow judges to do what the legislature
has not done or will not do, which also means that he finds himself compelled to present
what they are doing as legal. Thus he argues that the prohibition on extreme injustice is a
What Hart and Radbruch therefore seem to share is an inability to appreciate the
moral and legal complexity of the situation of judges faced with an unjust law. As I
suggested earlier, this inability may not be confined to the situation of the judge. One can
infer from Pappes description of the Federal Supreme Courts reasoning that the Court
supposed that a law-abiding citizen of the Third Reich would not consign someone to a
system presided over by the courts-martial for making remarks about Hitler because she
would know that the accused would not be tried according to law--by impartial judges who
give an appropriate interpretation of the law. For such a citizen, the moral and legal issue was
thus too complicated for a theory that can only deal with a clean clash between what the law
20
In sum, the Case of the Grudge Informer supports Fullers position rather than
Harts. Moreover, as I will now suggest, one can draw some implications from this case for
the Hart-Dworkin debate. While that debate has largely been conducted in terms of the
appropriate stance for legal theory when it comes to judicial interpretation of particular laws,
my analysis of the Case of the Grudge Informer suggests that Fullers terrain of legality
that is, debate at the fundamental levelwill likely prove more fruitful.40
Dworkin has recently provided an analytic grid for understanding debates in philosophy of
law.41 His own work, he says, is mainly about law in the doctrinal senseit explores the
concept of the law of some place or entity being to a particular effect, that is, what is
involved in claims about what the law requires or prohibits or permits or creates.42
The doctrinal concept must not be confused, Dworkin asserts, with a sociological
concept of law, the concept used when we name a particular type of institutional social
structure. While such concepts have their uses in historical or social scientific inquiries, it is
not helpful to argue about whether some group really had a legal system.43 The sociological
concept of law is an imprecise criterial concept, where we share some criteria that make it
possible to use the same concept but not enough that the answer to such questions can be
determined.44 As a result, the argument, once much beloved of legal philosophers, whether
the Nazis had a legal system is bootless for that reason.45 The issue is not that the
sociological concept lacks boundaries, since the doctrinal concept figures among the
boundaries of the sociological concept in the following way: nothing is a legal system in the
sociological sense unless it makes sense to ask what rights and duties the system
21
recognizes.46 However, the relationship is not reciprocal since we can deny that the Nazis
had a legal system in the same way that we can deny that a game is a legal system, while
answering the question what rights and duties the putative Nazi law or the game
concept.47
There is also what Dworkin calls a taxonomic concept of law, one which supposes that
a community that has law in the sociological sense also has a collection of discrete rules and
standards.48 But he sees no point to a debate about whether moral standards really are legal
standards, since in such a debate the important issue is the doctrinal one of whether and
Finally, there an aspirational concept of law, which we often refer to as the ideal of
legality or the rule of law. A great deal, he says, turns on what we take to be the correct
is fully secured when officials are required to and do act only as established standards
permit or more substantivelegality holds only when the standards that officials accept
respect certain basic rights of individual citizens.50 In his view, any adequate account of the
aspirational conceptof the values of legality and the rule of lawmust give a prominent
place to the ideal of political integrity, that is, to the principle that a state should try to
govern so far as possible through a coherent set of political principles whose benefit it
extends to all citizens. Recognizing and striving for that dimension of equality is, I think,
essential to the legitimization of state coercive power.51 He does recognize that one might
think that the political and social value of legal order lies in another ideal: the ability of that
order to facilitate citizens planning and coordinate their activities in the interests of
22
individual and collective efficiency.52 This view describes one version of what Dworkin calls
political positivism.53
positivism,54 whereas Fullers is sociological because of his claim that for there to be law,
However, the idea that at the most fundamental level debates in philosophy of law
are about the political point of law was not only put forward by Fuller in the 1958 essay; in
The Morality of Law, he used, perhaps even coined, the term aspirational to express that
idea.56 Moreover, he made it clear in 1958 and in later work that the role of judges was to
Fuller and Dworkin agree, then, about the methodology of philosophy of lawit is a
political inquiry which has direct practical implications for legal officials in that it should
inform, among other things, the judiciarys decisions about how best to answer questions
about what the law requires. They differ mainly about the level on which the inquiry should
principles which Fuller identified as constituting the internal morality of law are, as Hart put
it, unfortunately compatible with very great iniquity.58 Indeed, Dworkin joined in the first
wave of positivist critique of Fullers The Morality of Law, arguing that Fullers failure in this
regard indicated that a critique of legal positivism has to start with a focus on its
However, to the extent that positivists have made any concessions to their critics,
these have not been at the doctrinal level.60 At this level, they have responded to Dworkins
argument about the way that moral principles figure in adjudication in two ways. Either they
23
say that he is correct but only contingently so, since everything depends on whether such
principles necessarily are included or incorporated into the legal materials of every legal
order. Or they say that even when a legal order requires judges to resort such moral
principles, their decisions are not fully determined by law. Both responses are united by the
thought that Dworkins theory is not a theory of law but a parochial account of adjudication.
They differ only in that the second response regards Dworkins account of adjudication as
Harts remark in The Concept of Law that the formal features of governance by rule are
closely related to the requirements of justice which lawyers term principles of legality and
that if this is what is meant by the necessary connection of law and morality, we may accept
it. And the concession holds despite Harts point about the compatibility of legality with
For Fuller does not deny that particular laws are enacted as instruments of the policy
of the powerful, nor that the powerful might use the law to enact evil policy. Indeed, he can
turn against legal positivists a version of what they consider to be a knockdown argument
against Dworkins positionthat Dworkins doctrinal focus conflates explaining law with
giving an account of how judges decide the law of a particular jurisdiction. That is, positivists
wrongly infer from the fact that the law, especially the statute law, of any jurisdiction is used
instrument.
That this is a mistake is reflected in European languages except for English in the
fact that, following the Romans, the generic term for law contains an element of right,
ius/Recht/droit, in contrast with the morally neutral words for particular laws,
24
lex/Gesetz/loi. Moreover, it is reflected in English in the fact that we speak of the rule of
law, not the rule of particular laws. The debate about the connection between law and
morality is thus not, as positivists wrongly suppose, about whether there is a connection
between the lawthe particular laws of a jurisdiction--and morality but about the
connection between law, that is, legality or the rule of law, and morality.
particular on the way in which Fullers challenge to legal positivism, a challenge about
conceptions of legality, was displaced by a debate between positivists and Dworkin about the
best way to understand judicial interpretation of the law. Of course, the fact that particular
laws can be used as instruments of evil throws into question any claim that legality
evidences a necessary connection between law and morality. But, as the Grudge Informer
Case shows, the kinds of question that are thrown up by such laws are not well answered by
legal positivism.
legality, to the need to maintain legal order so that it is in good shape legally speaking.
Fullers claim about a legal order not being so unless certain minimal requirements of
procedural justice are met is about this issue; more accurately, it is a claim about the formal
attributes of a legal order, attributes that require procedures for their realization.62
This clarification helps to show that Dworkin is mistaken that the relationship
between doctrinal conceptions of law and sociological concepts is one way. Recall that he
says that the doctrinal concept determines, though far from fully, the boundaries of the
sociological concept, but the relationship is not reciprocal since we can deny that the Nazis
had a legal system in the same way that we can deny that a game is a legal system, while
25
answering the question what rights and duties the putative Nazi law or the game
recognize.
However, as the Case of the Grudge Informer shows, in cases where the issues reach
to the foundations of legal order, the clash will be between different sociological conceptions
of legal order.63 Moreover, the choice between sociological conceptions will in fact help to
determine the conclusions judges reach at the doctrinal level, and determine in the fullest
possible sense available in moral argument, just the sense on which Dworkin relies in his one
right answer thesis. Where Dworkin goes wrong is in supposing that the aspirational concept
than name it. We also set out what we take to be its most fundamental formal
we have seen, at stake in fundamental cases is not just the doctrinal question of how to make
the law into the best it can be, but also the question of how best to sustain the project of
law, otherwise known as legality or the rule of law. At this level, Hart and Fuller side against
Dworkin in supposing that the sociological level is significant, even of prime interest, to
philosophy of law, though Fuller sides with Dworkin both in supposing that philosophy of
law inquires mainly into the normative, that is, political and moral point of law, and that, at
the doctrinal level, when judges differ most fundamentally, they do not at that point step
Thus when Fuller said that a legal order might so far depart from the morality of
order, from the inner morality of law itself, that it ceases to be a legal system,65 he was not
engaged in a bootless argument. It was, as we have seen, a serious question for postwar
26
German courts, whether there was legal order during the Nazi period. Neither of the two
postwar courts we have encountered could properly attend to the question they had to
answer without engaging that fundamental issue, as well as the doctrinal issue which in part
First, the relationship between the fundamental level, the level which Dworkin would
deem sociological, and the doctrinal level is not one way but reciprocal. Answers to
questions at the fundamental level partly determine answers at the doctrinal level.
Second, the fundamental level makes possible the doctrinal level. Recall Dworkins
description of the doctrinal levelit pertains to claims about what the law requires or
prohibits or permits or creates. That way of putting things neglects the question of how not
the law but law makes it possible for such claims to come into being and to be properly
adjudicated. Crucial here is the distinction mentioned earlier between the formal
of official actions with law, and so onand the procedures that pertain both to the making,
the fiat of law, and to laws interpretation, the reason or justification of the law.66
This distinction could explain why both domestic law and international law are both
law despite the fact that they do not have in common certain procedural mechanisms, for
example, an independent judiciary with authority to interpret the law. Indeed, how law or
legality makes it possible for doctrinal issues about rights and duties in international law to
be resolved might actually tell us more about the nature of law than the same issue in
domestic law, since in the latter domestic mechanisms exist that obscure the question.
Similarly, theories that seek to demote judges from their prominent role in constitutional or
rights adjudication might wish to engage more not only at the fundamental level, but within
that level more with formal criteria than procedural criteria, in order to make the case that
27
certain procedural mechanisms are not important, or not as important as, say Dworkin
troubling to legal positivism since it is the level at which it attempts primarily to operate. So
the challenge does not offer it the opportunity legal positivism has seized in regard to
debate because all the action is in the Hart-Fuller debate. Recall that my argument started
with a case, one to which Hart paid unusual attention, because he thought it supported
positivism. My argument that it does not invoked both Dworkinian and Fullerian themes.
Moreover, once the Fullerian fundamental level comes into view, one might be able to
understand better Dworkins theory, as the kind of overarching argument alluded to earlier,
which connects the fundamental and the doctrinal levels via an account of laws aspiration to
teat each individual with equal concern and respect. If that were right, then the terrain would
be properly opened up for a much more fruitful confrontation than the one between
Dworkin and the conceptual legal positivism espoused by many of Harts followers, and
perhaps by Hart himself after the 1958 debate. I mean here the confrontation between
Dworkins position and political positivism, as for example, represented by Jeremy Waldron.
So my final conclusion on Fullers behalf is that his understanding of legality marks out the
28
1
Professor of Law and Philosophy, University of Toronto. I thank those who attended a
philosophy seminar at McMaster University and a graduate seminar in the Law Faculty of
the University of Toronto for comments on drafts of this paper, Mark Bennett for extensive
in Jurisprudence and Philosophy (Oxford: Clarendon Press, 1983) 49, hereafter Hart; Lon L
Fuller, Positivism and Fidelity to Law- A Reply to Professor Hart, (1958) 71 Harvard Law
Reconsideration (2002) 15 Ratio Juris 186 and my The Dilemma of Legality and the Moral
Limits of Law in Austin Sarat, Lawrence Douglas, and Martha Merrill Umphrey, eds., The
Journal of Legal Studies 1; Five Minutes of Legal Philosophy (2006) 26 Oxford Journal of Legal
Studies 13.
29
12
Hart, 74.
13
Ibid., 74-5.
14
Ibid., 72-3.
15
Ibid., 72-4.
16
Ibid., 73. The phrase is John Austins, quoted from Austin, The Province of Jurisprudence
Determined (Library of Ideas edn., 1954) 185, but Hart clearly adopts it as his own.
17
Hart, Positivism and the Separation of Law and Morals, again quoting from Austin.
18
HO Pappe, On the Validity of Judicial Decisions in the Nazi Era, (1960) 23 Modern Law
she was not obliged to give evidence under oath, that the accused was under threat of
capital punishment, and that, without her sworn evidence, proof was likely to be
insufficient. However, she insisted on going ahead; ibid., 265. Pappe points out that the
Federal Supreme Courts finding does not inevitably lead to the conclusion that the judges at
the court-martial have to be punished, since they could claim a defence under the Criminal
30
Code of intimidation, although he also notes that this defence would not be that convincing,
as judges who fell out of favour usually saved their skins simply by resigning; ibid., 268.
26
Ibid., 268-9.
27
Hart, 69-70, at 70.
28
Pappe, 271-2.
29
Hart, 62-72.
30
If only because Harts remarks are in the middle of his discussion of the penumbra of
uncertainty; ibid, 612-14. However, his amplification of certain points in this discussion at
627-9 shows that Hart thought that his point about wicked oughts applied also to decisions
in the core.
31
See Mertens, Radbruch and Hart on the Grudge Informer: A Reconsideration and my
The Dilemma of Legality and the Moral Limits of Law for more detailed argument on this
that his theory is one of law, not adjudication, and that one cannot draw implications from
an exercise of judicial discretion for legal theory in general, and in particular for legal
positivism. However, the Grudge Informer Case shows how problematic this response is.
37
Fuller, 655.
31
38
Ibid., 657.
39
Ibid., 660.
40
Fuller also preferred the solution of a retroactive statute. But he was anxious to state that
his reason for this preference was not the positivist one that the statute would be the most
nearly lawful way of making unlawful what was once law. Rather, the statute would
symbolize a sharp break with the past, enabling the judiciary to return more rapidly to a
condition in which the demands of legal morality could be given proper respect; ibid., 661.
I disagree with Fuller. There was a point to the post-war courts trying to show that the
German legal order persisted into the Nazi era, providing both a set of standards against
which the activity of both officials and citizens could be evaluated and a basis for legal
reconstruction. See also Fullers doubts about Pappes argument in The Morality of Law (New
Haven: Yale University Press, 1969, revised edition) note 2 at 40, on the basis that it was odd
for postwar courts to interpret Nazi statutes in the light of their own standards and that it
was out of place to strain over questions of interpretation when the statutes were so full of
vague phrases and unrestricted delegations of power. (These doubts are clearly anticipated
in 1958, in Fuller at 655.) However, once one takes into account the first point, that the
standards were not exclusively postwar standards but standards that persisted into the Nazi
32
45
Ibid., 224
46
Ibid., 4.
47
Ibid.
48
Ibid.
49
Ibid., 4-5.
50
Ibid, 5.
51
Ibid., 13.
52
Ibid., 13.
53
Ibid., 171-83, 240.
54
Ibid., 30.
55
Ibid., 3.
56
Fuller, The Morality of Law, chapter 1.
57
Fuller, 670; Lon L. Fuller, Anatomy of the Law (Westport: Greenwood Press, 1976) 19, first
published in 1968.
58
Hart, The Concept of Law, 206-7.
59
Ronald Dworkin, Philosophy, Morality, and Law: Observations Prompted by Professor
Fullers Novel Claim, (1965) 113 : University of Pennsylvania Law Review 672, at 687.
60
Unless one considers inclusive legal positivism as a kind of wholesale capitulation to
Dworkin.
61
Hart, The Concept of Law, 206-7. And we can recall here that in Harts critique of Radbruch
he did say that the principle against retroactive punishment is a very precious principle of
morality endorsed by most legal systems; see text to note XX. For insightful discussion of
33
Harts remarks, see Nigel Simmonds, Law as a Moral Idea (Oxford: Oxford University Press,
2007), 70-6.
62
I owe this important observation to Jeremy Waldron, though I put it somewhat differently
than he.
63
Hart says of such cases that in them nothing succeeds like success, thus giving up the
Kenneth I Winston has pointed out that Harts legal theory addresses only one mark of legal
authority the final determination of the content of the law. Hart thus misses the other
mark of authority, its congruence with legal principles. Attention to this other mark
ordinarily, the province of the judge brings one almost inevitably into an internal
perspective on law which is out of kilter with positivist claims about discretion and the
Separation Thesis. See Winston, Introduction, Winston, ed., The Principles of Legal
Order: Selected Essays of Lon L. Fuller (Durham, NC: Duke University Press, 1981) 16-23.
Compare Frederick Schauer, Fullers Internal Point of View, (1994) 13 Law and Philosophy
285, for an apparent concession that there is this second mark and that attention to it has
these effects, all the time arguing that it is still important to maintain a positivistic, external
point of view.
34