Professional Documents
Culture Documents
W03 Sigaut. 2003. Technology
W03 Sigaut. 2003. Technology
TECHNOLOGY
Franois Sigaut
In order properly to speak about techniques, one has first of all to know something
about them. Now there is a science that deals with techniques, called technology,
which, in France, has not received the recognition it deserves.
(Marcel Mauss 1948:71)
Broadly speaking, technology is the way people do things.
(Lynn White, Jr 1940:141)
TECHNOLOGY
other hand, the role of technics in the history of humankind and in the
differentiation of societies is so obvious that no mythology, ideology or
philosophy has managed to ignore it altogether. Some arts separate human
beings from other animals: fire, cooking, ornamentation. Othersfarming,
ceramics, weaving, metallurgy, etc.separate some human groups from others
and, according to many mythologies, civilized peoples from savages. Sometimes
the arts have been regarded as the gift of a benevolent god or goddess (e.g.
Triptolemus, inventor of the plough and patron of agriculture). But more often
they are said to have been stolen, bringing down all manner of misfortune onto
the heads of either the thief (Prometheus) or mankind in general (Genesis).
Invention is double-edged, benign or malignant, as ambiguous as the civilizing
hero of mythology, who is often depicted as ambitious, shifty, cunning: in short
a trickster. Perhaps no other mythic figure is more widespread. He is found
among peoples whose level of material culture is often considered to be the
lowest, like the Australian Aborigines. But he is also alive and well today, for
example in our science-fiction figure of the mad scientist. The mad scientist
has the same character traits as the trickster, the same adventures befall him,
and he often suffers the same fate, as the victim of his own inventions. Our
century has given birth to a host of mad scientists: Wellss Doctor Moreau or
Capeks Professor Rossum, the inventor of the robotThe line is long, going
back through Frankenstein and Faust to our classical Antiquity, where
Daedalus was probably the prototype.
There is reason to believe therefore that, ever since their emergence as a
species, human beings have puzzled over that strange thing called technics,
something they acquire without knowing how, that they possess but which
possesses them even more, that is not a part of them but without which they
would not be what they are. I have just mentioned science fiction as a modern
form of this questioning. Philosophy is another, and here again our century has
been ambitious and long-winded in its pursuit of a tradition that goes back to
pre-Socratic thinkers. The authors are innumerable (see for example Mitcham
and Mackey 1973); I need mention only a few well-known names of the period
between 1900 and 1960, such as Berdiaeff, Ellul, Heidegger, Mumford, Ortega
y Gasset, Sombart and Spengler. The philosophy of technics, and the
ideologies for which that philosophy furnishes one of the favourite media of
expression, have a long history on which we have either only very general views
(Auzias 1971, Moser 1973, Schuhl 1938) or, on the contrary, highly detailed but
specialized studies (Adas 1989, Espinas 1897, Herf 1985). Although we may
still be ignorant of most of this philosophical tradition, we must be aware that it
exists, because it gave rise to the anthropology and history of technics proper.
Two works seem to me to be perfectly paradigmatic in this respect. They are
Lewis Mumfords Technics and Civilization (1934) and Andr Leroi-Gourhans
Le Geste et la parole (19645). Both are particularly clear illustrations of how
history, in the first case, and anthropology, in the second, may be used to serve
explicitly philosophical ends.
421
CULTURE
TECHNOLOGY
CULTURE
At the end of his essay, Lafitte came to the conclusion that the science of
machines, which he called mechanology, is a social science, a branch of
sociology (1972 [1932]:109). Here engineer and anthropologist meet, for
Lafittes mechanology is manifestly the same as the ethnologists technology.
Both have the same purpose: to understand technical facts for what they are,
namely social facts.
DESCRIBING TECHNICAL FACTS: OPERATIONS,
PATHS AND NETWORKS
Our starting point, then, must be the observation of facts. But which facts? We
cannot simply go out and observe techniques, since we do not yet know what
they look like. What we see, to paraphrase White (1940), is people doing
things: a neighour painting his shutters, the plumber fixing a bathroom leak,
the barber cutting a customers hair, a mechanical shovel digging a trench in
the street, and so on. However dissimilar these activities may be, they all have
points in common. First of all, they are actions; this is obvious but must be
borne in mind. Next, they are material actions, in the sense that they all make a
material change in something. Finally, they are intentional, and are so on
several levels. For instance, the shovel driver deliberately digs a trench of a
certain gauge: he is following instructions from his foreman. But he also
knowsor someone else knowsthat this trench will hold pipes, that the pipes
will carry natural gas, that the natural gas will be used for cooking and for
heating the houses in the area, and so on. In some ways all human activities are
responses to such successive goals, which, step by step, affect every aspect of
life in society; so this cannot be the feature that distinguishes technical
activities from all others. What does distinguish them, perhaps, is that, as in the
above examples, the social goals have taken the form of material needs, and
these have become the agents true goals. The activities we are concerned with
are not simply material, they are intentionally material. That is perhaps their
most characteristic trait.
In a sense, this remark is implicit in Mausss definitionindeed it was already
present in the system of Aristotles causes (Mason 1895:15). Moreover, technical
action looks like a special case of what, in cybernetics, is called a goal-directed
system or mechanism (Couffignal 1978:26). Interesting though they may be,
however, I will not go into these cybernetic theories here, as it would further
distract us from our purpose, which is to indicate how directly to observe facts. If
it is important to note that the material changes are intentional, this is because it
enables us to identify these changes as the main factor that gives technical actions
their meaning. In no society are ultimate social and material goals separable. But
there are some activities in which the social goals have become material
imperatives for their agents; and these are what we customarily call technical.
In practice, this means that the goal of every action is to bring a physical
system from some state to another. This can be represented as follows:
424
TECHNOLOGY
Let us further assume that the change is an elementary one, that it is either
impossible or meaningless to identify any intermediate steps between (N) and
(N+1): we will call the corresponding technical act an operation. The
operation is someone doing something when that something is the smallest
material change that can be usefully observed. The operation so defined is the
first kind of technical fact that can be observed directly.
Our task is taking shape. It consists of identifying and describing the
operations performed within the human group with which we are concerned.
But before proceeding, it should be pointed out that no hypothesis has yet been
made as to the nature of the physical systems or the changes made in them. The
physical system can be the human body, as in many of Mausss (1936) techniques
du corps (techniques of the body). The changes may be as simple as moving
something from one place to another; they may even be non-changes, to the
extent that the goal is to retard or arrest certain undesirable natural processes,
for instance in canning or freezing food. The concept of operation applies to
all these cases; it does not depend on the nature of either the system or the
changes introduced by human actions. All that is required is that the changes
must be observable by the methods of the natural sciences.
The last point is crucial, for our ability to analyse action into operations
depends ultimately on the accuracy of the methods of scientific observation
that are used. In the eighteenth century, for example, grinding wheat la
grosse, as they called it in rural areas of France, involved a grand total of two
operations: crushing the grain between the millstones and then sifting it. Today
the working of a modern mill involves dozens of operations. Both the number
and the precision of the operations have clearly increased with the increasing
accuracy of the material means of analysing the product.
This raises another, more general, problem. What should be used as the
basis for identifying operations: our sciences or native knowledge? The
answer to this question depends on how well the one translates into the other.
In the Far East, the making of beer from rice or other grains has been
described by many ethnographers (e.g. Toffin 1987), and the practices involved
are very different from those used in making European beers. In most cases the
grains are ground and cooked, and a starter that has been prepared beforehand
is then added. I was personally never able to make sense of these practices,
despite good, detailed descriptions, until I learned from a Japanese colleague
that the starter was in fact a culture of micro-organisms that saccharified the
cereal starch (Ankei 1986). In Europe, saccharification, induced by the
germination of the grain (malting), is entirely separate from fermentation, and
425
CULTURE
so the two operations are readily identifiable in the descriptions. In Asia, they
are much harder to distinguish because the starter plays a part in both.
Of course this is only our own personal interpretation of indigenous
practices, and it would be a mistake to attribute it to the people we observe, if
only as empirical foreknowledge or unconscious intuition. Nevertheless,
experience shows that, without such interpretation, it is indeed difficult to
carry out ethnographic observation, and the results are often unusable. The
problem of access to indigenous knowledge stands, then, but unless we use the
methods of observation afforded us by the natural sciences, we must renounce
all hope of solving it.
As a rule, operations do not occur in isolation but as parts of a sequence that
can be called a path [Translators note: the author uses the French technical
term filire, which designates a die for the threading of screws or through
which metal is extruded to make wire; the idea is that of an obligatory passageway] (Gille 1978:16). Path is an almost intuitive notion, and most technical
works contain a wide variety of examples (Figure 1ac). In the example of beer
making, the path is composed of two specific operations, saccharification and
fermentation, along with a number of other, secondary, ones. Each path is part
of a larger process. In Europe, the brewing of beer is preceded by growing
barley and hops and culturing yeast; it demands a variety of devices that have
had to be made by the corresponding craftsmen; it burns fuel, and so on. Step
by step we realize that all the paths present in any one society are interwoven, in
some way or other, into a sort of network, which is in fact the economic
organization of that society.
It would clearly make no sense to try to describe in detail the overall
economic network of a society, however simple. It would be a never-ending
task, and the result would be unusable, which was exactly the difficulty raised
by Heider in the passage to which I referred above. The notion of network is a
limit-notion, whose sole purpose is to remind us that paths cannot be regarded
as isolated units any more than can operations. What we must do is to locate the
technical facts within the social space: such concepts as operation, path and
network are only instruments with which to do so.
It is easy to understand that the identification of technical facts requires a
consideration not only of their physical and chemical entailments but also of
their positioning on a path. It was biochemistry and biology that enabled us to
identify a particular stage in beer-making as starch saccharification. But it is
the goal of producing beer that gives meaning to saccharifying starch, for this
operation is indispensable to the fermentation that follows. Saccharification
does not appear in other food-preparation paths; or, if it does, we must expect it
to be in some completely different way that makes a different operation of it.
The preparation of cereals for cooking is full of such cases. In India, the parboiling of rice is a type of pre-cooking that is separate from the cooking proper,
basically because it takes place before, rather than after, the grain is milled
(Gariboldi 1974). The parboiled rice is then dried, which makes it possible to
426
TECHNOLOGY
Figure 1 Some examples of how technical paths (sequences of operations) are currently
represented.
(a) The making of a wooden shoe, France, 1940s. The sequence represented shows the
final operations in hollowing out the shoe. It includes realistic depictions of the tools, the
movements and the product. (From Enquiry no. 1810 with Auguste Becoulet, clog-maker
at Cuze (Doubs), 1943, investigators Barr and Perreau, Archives of the National
Museum of Popular Arts and Traditions, Paris)
(b) Production and distillation of crude glycerine, United States, 1900s. The diagram
portrays realistically the machinery employed. (After Lamborn 1918:605)
(c) Traditional processing of cereals, South India, 1980s. This is an abstract
representation (a flow-chart). Materials, products and operations are identified by name
only. Implements are not represented. (After Kimata 1987:44)
427
CULTURE
store it and to treat it more or less like ordinary rice. In Central America maize
is treated to an operation much like parboiling, nixtamalization (Muchnik
1981:28, Katz et al. 1974). The difference is that the kernels are not left to dry
but are crushed while still moist between two stones (mano/metate) to produce
a kind of dough, which is made into tortillas. That is why nixtamalization and
parboiling are not the same, but different operations. And it is also why
Mesoamerican grindstones, however much they may resemble Old World
prehistoric stones, must be regarded as different tools with a different
functional significance.
All of this points to one conclusion: defining an operation by its physicochemical properties is not enough to identify it properly. Its location on the
path must also be defined. Crushing, cutting, washing, drying, sifting
and kneading are not operations but empirical or indigenous categories of
action. They have some meaning on a physico-chemical level, and that is why
they are handy tagsall too handy, perhaps. For the purposes of technological
analysis, however, unless they are redefined in terms of rigorously observed
operations, such tags are insufficient, or worse, a source of error. We must know
exactly not only what is crushed, cut up or washed, but also whythat is,
where along the path the action of crushing, cutting, etc. occurs. In eighteenthcentury Europe, fallow land was ploughed several times, at set times of the year;
each ploughing had its particular characteristics and name (fallowing,
stirring, laying up, and so on): each of these ploughings was an observable
reality, while ploughing in general is only an abstraction (Sigaut 1977).
Because operations can be observed, they must be our starting point if we are
not to find ourselves prisoners of purely common-sense categories.
This rule is especially important for archaeology and museology, that is for
all areas of study that start from the physical object, the tool. It is obviously
necessary to study artefacts because they are often all we have. But given our
ignorance of the true nature of the operations in which they were used, there is
a serious danger of grouping them arbitrarily. Our own notion of knife, for
example, is so familiar to us that we have a hard time seeing beyond it. Thus
Leroi-Gourhan (1965, 2:125) felt confident enough to write that the
palaeontology of the knife can be traced back without a break to the first tools
(Figure 2). This statement assumes that there is only one type of knife (our
own), of which others are simply more or less incidental variations.
Surprisingly, the author seems unaware of the fact, reported as early as 1867 by
Pitt-Rivers (1906:74), that our everyday hafted knife appears only with metal,
and is altogether different from cutting implements made from stone. LeroiGourhan seems especially unaware of the extraordinary diversity among
present-day knives, of which Figure 3 gives only some idea, as it does not
show an Inuit womans knife (Mason 1892), an Indonesian or African reaping
knife (Fischer 1937, 1939), a North American hooked knife, or the odd-looking
(to our eyes) Indian stand-knifewhich is held upright with the foot, the
object to be cut being brought down on the knife (Figure 4)not to mention
430
TECHNOLOGY
431
CULTURE
the whole range of knives that can be bought in our modern hardware stores.
Clearly nothing useful can be gained from lumping all these together as
knives, because neither the notion of knife nor the corresponding idea of
cutting really makes sense. Both belong to those prenotions that Durkheim
warned us about and which are certainly largely to blame for the near-paralysis
that has long gripped the anthropology of technics.
Figure 3 A textbook selection of some recent knives. From left: knives from Persia and
West Africa; a woodcarving knife, Eskimo; a knife for cutting the umbilical cord, Gilyak;
a razor from Turkey; hunting knives, Orotche and Eskimo; a tool for scaling fish from
Japan. This selection shows quite clearly that tools put together on the basis of their form
(knives) and of their mode of working (cutting by pression), but with their functions
either unknown or ignored, cannot constitute a unit of any analytical value. (After
Hirschberg and Janata 1980:163)
432
TECHNOLOGY
Figure 4 Sickle-knife or stand-knife, India. The blade is held in a vertical plane, cutting
edge upward and backward, with the help of the right foot. The vegetables to be cut are
pressed against the edge with both hands. In many Indian languages, this tool is given
the same name as the sickle. Moreover, true sickles are sometimes used in the same way,
the handle being held down by the foot instead of using the small plate shown here. Any
concept of the knife that does not include examples like this one is incomplete, and
therefore invalid as an analytical tool. (After Mahias 1985:180)
433
CULTURE
Product family
chicha
sake
beer (sensu stricto)
Obviously this list, which is not original (see Ankei 1986 for sources), does not
exhaust the topic. Making beer is always a long process involving many
operations made up of numerous variable elements. Each of these elements
must in turn be analysed in this manner. But the method remains the same: it
consists of plotting the analysis along two axes: a vertical y-axis, which charts
the successive transformations by means of which a human group appropriates
certain natural objects, and a horizontal x-axis, along which are placed the
various alternatives for carrying out each transformation. The operation is the
point at which the two axes intersect.
The scope of this article makes it difficult to go much beyond a general
presentation of a few basic concepts. And perhaps it is not even desirable to go
further, in as much as anthropologists are still in the first stages of their
434
TECHNOLOGY
thinking on this subject, and have not yet achieved consensus on their
conceptual vocabulary. In such circumstances, any moderately serious
discussion runs the risk of turning into a semantic battle. Nevertheless, I would
like to clarify two important points: first, the notion of technical lineage that
has just been proposed, and second, with respect to artefacts, the ill-recognized
but indispensable notion of their workings, which falls between structure and
function.
The actual expression technical lineage goes back some thirty years
(Simondon 1969 [1958]: 409), but the idea is older still. It is implicit in two
virtually inescapable common-sense observations: that techniques progress;
and that this comes about by successive, orderly accretions, which creates an
impression of linear progression (the internal-combustion engine was invented
after the steam engine, the alternator after the direct-current dynamo, and so
on). Perhaps no one has expressed these two concepts as vigorously as PittRivers, who took them as guidelines for his museum displays, and in particular
his tool collections (Pitt-Rivers 1906, Sigaut 1990).
Technological evolutionism of this kind seems so simplistic today that it is
almost too easy to criticize. But if as a theory it is indefensible, as a problem it is
not, for it is not possible simply to brush away the appearances it was supposed
to explain. It may seem preferable to compare the process of innovation to a
game of Scrabble (Callon and Latour 1985) rather than to one of dominoes
(Pitt-Rivers 1906:19), but the two analogies point to the same process of
recombining pre-existing elements with a view to achieving goals that remain
basically the same: to do something better, cheaper and faster. From a
diachronic perspective, then, the evolution of techniques inevitably gives the
impression of a veritable orthogenesis. It has even been argued that this
apparent predictability is one of the distinctive features of technical as opposed
to scientific evolution, the latter being fundamentally unpredictable (Cazenobe
1987).
Clearly this traditional idea of the technical lineage lies at the heart of an
extensive debate. But the concept of lineage, as presented here, falls outside this
debate because it is purely descriptive. It is designed for analysing techniques
as they are defined above: as alternative ways of carrying out a given operation.
Techniques are first of all actions, movements of the body and limbs. But these
movements are mechanical in nature and produce all sorts of results,
mechanical and otherwise. It is because human movements bring into play
something other than themselves that they are effective. This other thing can be
called an effect, a term borrowed from physics, as in Joule effect, Doppler
effect, etc. To come back to our example of beer-making, we have seen that
there was always one indispensable operation, starch saccharification. If we
compare the technographic descriptions of this operation, we obtain a host of
different techniques. But, at a first level of analysis, all these techniques use one
of three effectsinsalivation, starter, or maltingwhich can be distinguished
by biochemical means. Every starch saccharification technique belongs to one
435
CULTURE
of the three corresponding technical lineages. Or, to put it another way, it is the
use of a given effect in a given operation that identifies a technical lineage.
Contrary to the old idea of technical lineage discussed above, the concept as
it has just been defined is an analytical and descriptive tool entirely free of
diachronic connotations. It groups techniques in terms of their nature, not
their history. This is approximately equivalent to Deforges meaning (1985:95
125), although he uses principle where I would have used effect. Other terms
are possible as well (method, process, and so forth). But whichever finally
prevails, it is important to understand that an effect is not a technique: it is only
one of the components of a technique. Centrifuging, for example, is an effect
involved in operations as dissimilar as separating uranium isotopes, testing
blood, separating milk, spinning the wash, certain circus acts and fair rides, and
so on. It is clear that the only thing they have in common is centrifugal motion,
and that the centrifuge effect can be used to identify a technique only once the
operation concerned has itself been unambiguously identified.
And yet effect and technique are frequently confused. This can be explained
by the important role of effects in technicians discourseengineering is
essentially a science of effectsand by the spectacular, not to say miraculous,
nature of the innovations that make use of hitherto unknown effects, like the
electromagnet in the nineteenth century or todays microwave oven. The
popular history of technics has always been fascinated with innovations of this
kind. But it is precisely the confusion between effect and technique that has
spelled the failure of all attempts to classify techniques, however necessary it
was to try. It was the same confusion that led someone like Lucien Febvre
(1935) to think that a distinction could be made between a technical history and
a social history of technics. There is no such thing as a non-social or an asocial
history of technics. What Febvre meant by a technical history of technics was
no doubt a history of effects; if such a history is possible, it is probably a branch
of the history of science.
The current inability to distinguish, in artefacts, between workings and
function is a consequence of the common confusion of effect and technique.
Workings and function are classic cybernetic concepts (Couffignal 1978
[1963]:28); we also run across them in neurobiology (Paillard 1976:37,
1986:1385)indeed they are necessary for describing any goal-directed
system. In technology, they correspond to two questions that spring to
everyones mind when confronted with an unfamiliar object: What is it for?
(function) and How does it work? (workings). To these must be added the
concept of structure, answering the question What is it? or What is it made
of ? It is therefore all the more startling to find technological literature
commonly confusing the two (Cazenobe 1985, 1987, Sigaut 1987b). The
shortcomings in the notion of knife were just a case in point. We think we are
talking about function when we say that a knife is used for cutting. But cutting
alone (Leroi-Gourhan 1965, 2:125) is not a function, it is at most a category of
working modes, and a heterogeneous and arbitrary one at that. The butcher
436
TECHNOLOGY
who cuts the steak I have just ordered does not go about it in the same way or
with the same kind of knife as will I, an hour later, when I cut up the steak on
my plate. For cutting to be a function, we must know exactly what is cut, in
what context and to what precise purpose; in other words, we must know what
operation we are talking about.
It is the location within a specific operation, with all the finalities this
implies, that defines the function of an artefact. Its workingshow it actually
workslie in the way it intervenes in the effect that is to be produced. And its
structure is the set of geometric and physical properties that result from the
artefacts manufacture and use: shape, size, materials, assembly, solidity,
elasticity, resistance to heat, and so on. Any study of an artefact must therefore
include three levels of analysis: the structure of the artefact, the way it works
and its function. On the structural level the analytic procedure is obvious: it
means using all available means of investigationgeometric, physical, etc.to
describe the object itself. This is the natural starting point for the study of any
object, the results of which are limited only by the capacity of the physical
apparatus used. The difficulty begins with the next level. Some aspects of an
objects workings are dictated by its shape (as with baskets, files, flutes): these
can be discovered by the same methods of investigation as before. The
archaeological study of artefacts is an example. But stone tools, the
prehistorians staple, are a very special case. If, for example, it were possible to
reproduce an ancient musical scale on a lithophone (Condominas 1965:333
53), the same experiment would make no sense with a violin. On the whole, the
unknown object does not speak; you have only to think of a collection of
knitting needles. Museum store rooms and old hardware catalogues are full of
such objects, perfectly ordinary for those familiar with them, but perfectly
puzzling for the layman; and in the latter case, the interpretations can be
hilarious. A humorous book presenting just such interpretations was written by
Macaulay (1979), but he was making a serious point. Given an unidentified
object, the means of investigation available allow us to describe its structure and
to figure out something about its workings, but they do not allow us to discover
much more, least of all its function. The only way to accomplish this is to
proceed by analogy, that is to compare the object under study with similar
objects whose function is already known. The danger comes when the analogy
is implicit and therefore unreasoned. For instance, for a long time any object
that resembled a sickle was considered to be a sicklethat is, a tool used to
reap cereals (function). In the case of stone-bladed sickles, this interpretation
seemed to be corroborated by an examination of wear marks (silica gloss). But
the available ethno-historical information on reaping techniques tells us that
the sickle is anything but a primitive tool. There are at least seven lineages of
reaping techniques that use tools more simple than the sickle, which is never
found among peoples unacquainted with metal, nor even among most peoples
who, although possessing iron tools, do not use animal powersuch as those in
sub-Saharan Africa and Indonesia (Sigaut 1978, 1991). Moreover, new
437
CULTURE
methods of examining wear marks have shown that those observed on stone
sickles matched those made by the cutting of grasses, reeds, rushes, etc., as
well as by the cutting of cereals proper (Anderson-Gerfaud 1983). And so today
flint blades with silica gloss are no longer regarded, as they once were, as
providing unequivocal evidence of agricultural activity. Their interpretation
has become more complex and, at the same time, more interesting. But in order
to arrive at this stage, researchers had to stop confusing workings with
function.
The foregoing presentation of technographic analytical concepts may have
tried the patience of some readers, and perhaps all the more sorely because it is
still a very sketchy outline. Are these concepts really valid and useful, or might
better ones not be found? The question must remain open until a real
discussion of the subject is undertaken by contemporary anthropologists. The
only conviction that I would like to share is that, without adequate concepts
those proposed above, or otherswe may continue to see a more or less
interesting philosophy of technics; but there can be no scientific technography,
and therefore no real anthropology of technics.
TECHNICAL INTELLIGENCE: KNOWLEDGE AND
SKILLS
So far we have been looking at technics from the outside, and have not
concerned ourselves with the knowledge and skills they require. And yet the
importance of these is unquestionable, as is obvious in the case of the machines
and sophisticated devices that surround us. It becomes even more obvious
when we consider that, aside from a bare dozen innate reflexes, the child must
learn step by step, one movement at a time, every way of doing things that will
make him or her a normal person. We rediscover some of these along with the
child, for we adults had internalized and, in a sense, forgotten them long ago.
The fading of knowledge in the process of assimilating it, or, better, literally
embodying or incorporating it, is an essential feature of effective action, and
thus of technics. The same holds both for machines and for the nervous system.
Learning is not complete until the transmitted patterns of action have become
automatic; a machine cannot be considered to be perfected until it no longer
needs constant repairs or adjustments. The perfect machine has often been
compared to a black box, which means not only that you can no longer see how
it works, but also that you no longer need to know. An experts nervous system
is an even more impenetrable black box. After a century of study, it is still
something of a mystery how typists can type as fast as they do (Salthouse
1984). The knowledge built into a machine can always be retrieved, at least in
theory. But we still seem ill-equipped to identify the skills embodied in our own
nervous systems.
Skills should not be confused with knowledge, however; the condition for
effective action is not knowledge but skillsit is not sufficient to know how a
438
TECHNOLOGY
car is driven or a piano played if one is actually to be able to drive a car or play
the piano. Turning knowledge into skills takes a learning period of anything
from a few minutes to many years, but the learning is always necessary.
Similarly, turning the idea for a machine into the real thing necessitates a whole
process, which Simondon has called concretization (1969 [1958]: 1949), at
the end of which the initial abstract combination of elements has disappeared,
leaving in its place a new material object. It is for this reason that technics
cannot simply be reduced to applied science. As early as the last century,
Reuleaux (1865:viii) vigorously denounced this misconception. More recently,
the example of Marconis invention of the radio demonstrated that scientific
knowledge takes on a different function, meaning and content even when it
passes into the technical realm (Cazenobe 1983, 1986). In short, it stops being
scientific, and that is why it is artificial and futile to search for a distinction
between so-called scientific techniques and what we call crafts or traditional
techniques. The savage mind does not exist; the Neolithic paradox even less
(Lvi-Strauss 1962:224). There are just two kinds of practice: in one,
knowledge is subservient to material action; in the other, action is subservient
to knowledge. In daily life, the two are usually so intimately bound up that they
are often claimed to be indistinguishable. If we were to follow this argument,
however, we would have to abandon any attempt at making distinctions with
regard to human behaviour, and with it the task of analysis. In reality, the rare
cases for which we have sufficiently detailed studies would seem to indicate that
technical innovation does not result directly and necessarily from the progress
of knowledge alone. It even happens that stating a problem in overly scientific
terms leads away from the solution: a good example might be the graphiola
epidemic, the epiphytic disease that, since 1970, has destroyed nearly all French
and English elms. In over ten years of assiduous research, biologists have not
found a cure; indeed they have not even been able to rediscover the remedy
already developed in the 1850s, at the time of an earlier outbreak of the same
disease (Fleury 1988)! The accumulation of knowledge follows its own logic,
and this logic alone does not necessarily produce effective solutions.
In other words, knowledge is necessary for action, but the fact that it is
embodied in the very process of action means that this is no longer the kind of
knowledge we are used to talking aboutthat is, knowledge as the goal of
action. And this has led to a certain reticence on the subject among researchers,
even when they have not preferred to ignore it altogether. Technical
intelligence, if we may use this expression to designate the knowledge and skills
used in the course of technical action, is nearly virgin research territory. There
are few overviews of the subject (Gille 1978:141677, Leplat and Pailhous
1981), especially by anthropologists (Chamoux 1978). Field studies are hardly
less scarce (Chamoux 1986, Chick and Roberts 1987, Sagant 1987), and, on a
point as essential as the analysis of basic body movements and postures,
research has only recently begun to make a comeback (Bril 1984, Devine 1985,
Kawada 1988) after a blackout that has lasted since the late 1930s (Boas
439
CULTURE
1938:119). This blackout was part of a total eclipse which affected not only
technology in general, but also neighbouring disciplines such as psychology
and ethology (Ardrey 1969, Griffin 1984). It was due to a veritable censorship
exercised by doctrines such as functionalism and structuralism in
anthropology, and behaviourism in psychology. In the last twenty years,
ethological field studies, mainly of primates, have exploded the traditional
frameworks. This work has gone hand in hand with the development of what
we may call experimental psycho-archaeology, with the slightly paradoxical
result that the repertoire of technical gestures of chimpanzees or Homo habilis
is, if not actually better known, at least better studied than that of
contemporary men and women. We know everything, or nearly everything,
about chimpanzees techniques for cracking nuts or fishing out termites. About
human body movements used in pounding and grinding grains, shredding
roots or peeling fruit, we know just enough to be able to say that they vary from
one cultural area to another, and that the difference is often striking. But since
Masons articles (1892, 1899), no work has been carried out on the ways of
using a knife. And on such a fundamental subject as manual grasping, it is
significant that we owe the only information we have (Marzke and Shackley
1986, Napier 1980) to a triple detour through primatology, psycho-archaeology
and robotics.
In fact the study of technical intelligence and its component elements
(knowledge, skills, gestural habits, etc.) is no more advanced than it is because,
in ordinary research situations, the difficulties and misunderstandings
overwhelm the student. This is why specific research situations that reduce the
difficulties are of particular interest. Marzke and Shackleys work provides us
with three such situations: in primatology and in experimental archaeology,
because body movements that appear too ordinary to be worth observing in
todays human beings suddenly become fascinating in apes or in relation to
prehistoric artefacts; and in robotics, because in order to transfer human skills
to a machine, they must first be made explicit, at least partially or by analogy. A
fourth propitious situation is in sports, because this is an area in which detailed
analysis of skills is of obvious concern (Simonet 1986), especially in top-level
competition, where the financial stakes are huge and the race to stay at the top
unrelenting. Moreover, sports have something in common with experimental
archaeology, in that, be it the boomerang or the crossbow, equitation or sailing,
these are areas in which experimenters have a ready audience and willing
sponsors (e.g. Spruytte 1977).
Sports can also be compared to various organizations that, for purposes of
pedagogy or social criticism, encourage their followers to turn to techniques
other than those prevailing at a given time. These movements go back to the
beginning of the century, and can perhaps be traced directly to the writings of
H.D.Thoreau (Walden, 1854). As alternative technologies and the like, they
were fashionable in the 1970s. Subsequently having come under attack
(Schnaiberg 1982), these movements have today turned to criticizing
440
TECHNOLOGY
CULTURE
its own difficulties. Technical intelligence is hidden behind the results of the
technique by a whole variety of mechanisms: the deceptive ease of the acrobat
or the magician presents us with the classic stage image. In a sense, the ideology
of applied science succumbed to this mystification. Anthropology must not fall
into the same trap.
When all is said and done, we know next to nothing of what goes on behind
the scenes (Reuleaux 1877:3). We only know that we must go and see, and that
in order to do this we must find situations and methods that permit
observation. That is what I have been saying here.
ADAPTATION AND INNOVATION
In the last two centuries, technical development has been so spectacular that
the European mind has, in certain respects, been almost captivated (Adas
1989). The notion of the industrial revolution was perhaps the first indication
of this fascination. The expression had appeared in the writings of Marx and
Engels by the middle of the nineteenth century (Rioux 1971:8), and with
Toynbee (1884) and Mantoux (1973 [1906]) it became a veritable historical
paradigm, which was to provide the inspiration for the notion of the
agricultural revolution (Gras 1925:20832) and for Childes (1936) notions of
the Neolithic and urban revolutions. Concurrently, evolutionists relied heavily
on techniques and artefacts to distinguish successive prehistoric periods
(Thomsen 1836, Lubbock 1865), or stages of civilization (Morgan 1877,
Mumford 1934), or modes of production (Marx) and economic forms (Hahn
1896, Forde 1934).
Today no one still takes evolutionists seriously. And yet we have inherited
some of their deeply rooted mental habits. Some are simply handy, like the
Thomsen-Lubbock typology (Palaeolithic, Neolithic, etc.) which, with a few
adjustments, remains the frame of reference for prehistorians today. Others are
more debatable, like the categories hunter-gatherer, agriculturalist, pastoralist,
etc. that anthropologists have commonly used following the precedent of Forde
(1934). This vocabulary implies that, on a certain level, societies are defined by
their techniques, a hypothesis that has only to be stated to reveal how risky it is.
Moreover, in the case of a category like hunter-gatherer, the hypothesis is all the
more indefensible in that the real criterion of classification is not positive but
negative, not the presence of hunting and gathering techniques but the absence
of farming. On this basis, it has been claimed that man has spent 99 per cent of
his existence as a hunter-gatherer (Lee and DeVore 1968:3), but this claim is
not based on any tangible evidence, any more than are the countless theories
implying the uniqueness of the industrial, post-industrial or technological
society compared with all other forms of society that have gone before. Of
course techniques play a role in the way societies function. But if this role is
assumed to be self-evident from the outset (as in the ideology of progress),
there is no chance that it will ever be elucidated. This fundamental error may
442
TECHNOLOGY
well be part of the reason for the demise of evolutionism, despite the real and
little-known richness of other aspects of the evolutionists thought (Sigaut
1990). The same error has proved all the more fatal, since it seems to have
elicited, as a counter-reaction, the equally erroneous idea that techniques are
entirely determined by social relationships. While the steam engine certainly
does not explain industrial capitalism, capitalism does not explain the steam
engine either. This type of alternative merely proves that the problem was badly
defined in the first place.
And perhaps it was badly defined simply because it was defined too hastily.
Using techniques as a criterion assumes that they are already known, which
locks the researcher into false common-sense assumptions from the start. For
example, we continue to characterize the Neolithic by the presence of
agriculture. But since we know so little about Neolithic agriculture, that is
about the techniques it used, this is a purely verbal exercise. Moreover, some of
the things we do know are strangely ignored. For example, agriculture and
animal husbandry are usually considered to be essentially food-producing
activities; most recent research in cultural ecology is conducted on this
premiss. But this ignores the importance of clothing. As far as we know, the
value of skins and fleeces, in the Near East, was immeasurably greater than that
of meat, which would explain the domestication of sheep and goats much more
cogently than do the strictly food-producing hypotheses (Sigaut 1980). But the
simple fact of taking clothing into consideration breaks up the classic categories
of hunter-gatherers and farmers. For where do you put peoples like the
eighteenth-century Iroquois, Pawnees, and others, who farmed for their food
but hunted for their clothing?
Pre-Columbian America turns out to be a particularly rich field for analysis.
Because of the absence of metal tools and work animals, and because of the few
basic plants cultivated (maize, manioc, potato), its agrarian systems were much
less varied than those of the Old World. And if the zones in which there was no
agriculture, for mainly ecological reasons, are set aside, it would seem that
clothing habits, resources and techniques were what made the difference.
Clothing is important throughout the Pacific region, from Peru to New
Mexico; garments are usually woven from fibres obtained from either
cultivated plants or domestic animals, and most agricultural tasks are
performed by men. In the eastern regions, on the other hand, from Argentina
to the Saint Lawrence, clothing is either absent (in the tropical zones) or made
of the skins and furs of hunted animals, and most agricultural tasks are
performed by women. It is no doubt not a matter of chance that societies with
the strongest hierarchies or a state organization developed in the Pacific zone.
And it is evident that only by studying and comparing the functions of clothing
and weaving techniques (e.g. Murra 1962, Driver 1969 [1961]: maps 8, 21, 26)
will we perhaps come to understand this development.
This example raises a set of problems that are important for many, though no
doubt not all, techniques: those of their geographical distribution. The problems
443
CULTURE
are old ones (Mason 1894) but, despite the efforts of a few diehards (e.g. the
articles of D.S.Davidson cited in Brace 1980; Driver and Massey 1957), they
attract little attention today. It is obvious that a certain type of ethnological
mapping is a sterile exercise at best (Bromberger 1984). But cartography as a tool
is less to blame than the way in which it has been used. In the case of techniques,
at any rate, their geographical distribution is a sort of synthesis of several types of
very basic social processes: processes of adaptation to surroundings or, better,
management of resources and physical constraints, on the one hand; and
processes of producing and transmitting innovations, on the other.
Adaptive processes were the first to come to mind. Human geography and,
long before that, climate theory attempted to explain societies in terms of
adaptation; today cultural ecology has taken up the same challenge. The trouble
with the notion of adaptation is that it assumes, a priori, an identifiable
environment to which societies then proceed to adapt. But environment in this
sense has no observable existence. The only truly observable environment is the
one a group perceives and uses, and, in particular, the set of resources and
constraints that it recognizes as such in speech and practice. Each social group,
then, has its own environment, which can be and often is different from that
of another group sharing the same territory. The only way to identify this
environment is by technological analysis, to which all ecological analyses should
be subordinated.
In the case of non-Mediterranean Europe, for instance, technological (and
dialectological) analysis immediately shows sod or turf to be one of the decisive
factors in the development of farming techniques specific to this region (Sigaut
1975). Now, with a few rare exceptions (Evans 1970:56), the abundant
literature on European agricultural history ignores sod as an environmental
factor, which means that an essential piece is missing from all theories which
have, up to the present, tried to explain this history. Another example, which we
have already used here, is that of fermented beverages: chichas, beers and saks.
Technological analysis teaches us to differentiate them. It also teaches us that
the level of amylase in sprouted grains is a decisive environmental factor. Barley
has an exceptionally high level, and Europe makes beer. Rice, on the other
hand, has a very low level, and all the rice-growing regions of Asia make sak.
Now the industrial destiny of beers and saks has clearly been quite different,
as has the evolution of the modes and rituals of consumption associated with
each. The point is obviously not to explain all differences between European
and Far Eastern societies by the greater or lesser suitability of soils for growing
sod, or by the higher or lower level of amylase in barley and rice. The points are
simply (1) that these factors belong fully to the environments specific to these
societies, and (2) that only by technological analysis is it possible to identify
them. It must be understood that the notions of environment and adaptation
have no scientific content outside this analytic context. And that is probably
why, since the beginning of human geography, most research on the theme of
adaptation has yielded such disappointing results.
444
TECHNOLOGY
Compared with adaptation, the themes of the production and the circulation
of innovations have stimulated much less research, even though they are just as
important for human geography. Edmonsons (1961) model of the circulation of
innovations, according to which innovations springing up at random in a
uniformly inhabited space and spreading at a constant rate tend to collect at the
centre of the space, has oddly enough received no attention at all. Rather than
starting in Iraq and diffusing outward, the Neolithic appears to a considerable
extent to have converged on it, Edmonson concluded (1961:72), and this view
has, on the whole, been confirmed by subsequent research. The model needs to
be adjusted to take into account the physical and social differences within real
inhabited spaces. But, with this one reservation, it appears to be of considerable
heuristic value. For Europe, in any case, the concentration of the most highly
developed forms of agriculture and industry in the region extending from the Po
River plain to the Low Countries, which has been observed since the late Middle
Ages and which has once more become apparent after having been masked for
two centuries by the coal and steam episode, must no doubt be explained in this
way. However uniform the geographical environment is assumed to be,
innovation inevitably introduces differences that increase with time: this is
perhaps the essential lesson to be learned from Edmonsons model.
Geography is no less important to the production of innovations than to
their circulation, though obviously for different reasons. The tendency for all
workshops or crafts to gather in one area or village is probably quite universal,
since it has been observed in so-called primitive societies (Malinowski 1935,
1:1223) as well as in our modern industrial societies, and most notably in
Californias Silicon Valley. The location of resources or infrastructures is only
part of the explanation of this phenomenon. An interpretation in terms of
exchanges (Lowie 1934:1089, 146, Lemonnier 1981) is broader in scope, but
leads to a chicken-and-egg alternative: was it the need to exchange that led to
specialization, or specialization that made it necessary to exchange? The only
solution, it seems to me, is to bring into consideration a too often neglected
element: the production and reproduction of technical skills.
THE SOCIAL PRODUCTION OF TECHNICAL SKILLS
As I have already argued, skills are notor not simplyforms of knowledge.
They cannot exist apart from permanent practice. They cannot be written
downat least not all of themnor transmitted by speech, as in classroom
teaching. They owe their existence to constant renewal in the course of practical
action, a renewal of which innovation and apprenticeship are only particular
aspects (on the necessity for apprenticeship, see Wynn, in this volume, Article
6). The French say cest en forgeant quon devient forgeron, the English,
practice makes perfect; both proverbs are more apposite than ever, whether
they are applied to high-level sporting competition, heavily automated
industry, or computer sciencewhich is all the more outstanding an
445
CULTURE
illustration as it was unexpected. We were expecting Big Brother, that is, the
victory of completely algorithmized, centralized expertise. What we have, at
least for the time being, are practical jokers, pirates, bugs and virusesin
short, computerized anarchybecause neither the machines nor their users
are behaving as expected (Gonin and Joffrin 1983, Denning 1988, Marshall
1988). This is nothing new; it could be called the law of the irreducibility of
skills. It has been observed throughout the history of communicating
machines (Perriault 1989). Indeed it runs through the entire history of
technics, which from this point of view might be interpreted as a constantly
renewed attempt to build skills into machines by means of algorithms, an
attempt constantly foiled because other skills always tend to develop around the
new machines. Mason already saw this nearly a century ago: It is sometimes
said that the substitution of unerring machinery has taken away the cunning of
the human hand. The case is not nearly as bad as that, however. No change of
apparatus can deprive the human race of geniuses, for the man of knack will be
found excelling in the handling of the new machines (Mason 1895:26).
What concerns us here, however, is that the production (including the
reproduction) of skills is inseparable from the production of material goods,
and, consequently, the way societies organize themselves to producethe
Marxists mode of productiondepends as much on the skills as on the
goods to be produced. It seems that this point has never really been made. After
having recognized the failure of purely economic explanations, for example,
industrial historians turned to cultural explanations (Why isnt the whole
world developed?, Easterlin 1981). This was real, but only partial, progress, in
that the notion of culture invoked seemed too broad to be an effective guide to
research (Dellheim 1987). To precisely what cultural differences can variations
in productivity be attributed (Clark 1987)? My hypothesis is that the mode of
production and reproduction of skills, as only ethnography can reveal it, plays
an essential role.
It is probably no coincidence that more and more students, ethnologists or
not, are turning to ethnography in their attempt to understand what really goes
on in factories, stores, offices, etc. In the last ten years, vocabulary borrowed
from ethnology has come to be routinely used in the sociology of the sciences,
as for example in a recent book entitled Academic Tribes and Territories (Becher
1989). And journalists have not been far behind; after the October 1987 stock
market crash, the media virtually swamped us with articles on that exotic
people, the Golden Boys, their various tribes, dialects, rituals, identity tags,
customsincluding clothing and sexual practicesetc. (Cam and Filloux
1987a). Clearly these elements are not pieces of folklore, in the derogatory sense
of the term, but external manifestations of the culture within which are
produced the skills needed for stock markets to work. Of course, to work they
also need machines (telephones, computers, and so on) and theoretical
knowledge (the Golden Boys are university trained). But these would be useless
without the skills developed on the spot, of which one of the most striking to
446
TECHNOLOGY
the outsider is the sign language used on the floor of the Paris Futures
Exchange (Cam and Filloux 1987b; a similar language was used in the Chicago
Grain Exchange at the beginning of the century: Perrignon de Troyes 1926).
We may note in parentheses that the example of the Golden Boys points up a
dimension needed by economics if it is to become a true science: economists
must take into account the way markets work in reality and not only in theory.
But the main interest of this example is its visibility. The Golden Boys are only
a special case of a universal phenomenon, the constitution of identifiable social
units around certain skills, which in this case happened to make the front pages.
The problem, then, is to elucidate the relationship between the social units
and the skills produced in them, which is only a more general way of putting
the problem raised earlier with regard to geographical examples like Silicon
Valley. It is still rare to find research conducted from this perspective. The work
I have come across (e.g. Darr 1985, Kuhn 1985) suggests the following series
of hypotheses:
1
2
3
4
Normal social life requires that each person acquire a minimum number of
materially and socially effective practices.
The skills underpinning these practices can be produced and transmitted
only in groups of a certain size, not too small and not too large.
In order to last any length of time, these groups must have a recognizable
and recognized identity, a minimum of internal cohesion, procedures for
recruiting and initiating new members, for settling conflicts, and so on.
Furthermore, each group must balance its exchanges with the outside,
which depends on the efficacy of the skills it produces. An effective group
will see its social status rise. One that is not effective will decline until it
falls under the domination of other groups or disappears altogether.
These hypotheses call for further development, which would exceed the scope
of the present article; but just to give some idea, I will comment briefly on two
points: the skill-producing group itself and the notion of efficacy.
Why does an individual need to belong to a group, and not just any group, to
acquire the skills he needs to live? On one level the answers are so obvious that I
need not go into themfor example, there is no innovation without tradition,
and no tradition without a social unit to carry it on. But there is also something
more than these self-evident truths. There is the fact, long ago recognized by
Espinas (1897:5378), that reality is real only if there is some social consensus
to recognize it as suchthat is, if others allow the person to interpret his or her
feelings and perceptions as evidence of reality. Of course, these others are not
just anyone. They must be individuals whom the person knows and with whom
he or she shares the same general experience. Which is to say that others are a
group whose identity is defined by common abilities, and whose numbers are
restricted by the condition that everyone must know everyone else to several
persons or at the most several dozens.
447
CULTURE
TECHNOLOGY
But is this result really so odd? Is it not rather our habit of separating the
material and social spheres that should be questioned? Material efficacy is
effective only if it is also socialthat is what the example of todays sports tells
usbut it can be effective in a number of ways. Efficacy is always a generator of
identity; and in sports it produces little else, which is their main peculiarity. In
this sense sports reveal, as it were, some of the most deeply ingrained
tendencies of our society in their pure state (Bromberger et al. 1987). Efficacy is
an indispensable notion in technology, and it enters into the very definition
given by Mauss; the problem lies in knowing what it represents in each case.
What conception of efficacy, for example, underpins Indias caste system?
Within our own contemporary European conception, the caste system remains
ill understood, despite several generations of research. But if, contrary to our
custom, we were to include in our conception of efficacy such notions as purity
and pollution, merit and blame, etc., the caste system might begin to seem less
strange. In particular, the obsession of its members with hierarchical rank is
perhaps not without analogies to the obsession with rank characteristic of our
own sportsmen and sportswomen.
EXCLUDING TECHNICS FROM THE IDEA OF
SOCIETY: THE CASE OF SLAVERY
The main problem in technology consists in the inadequacy of the concepts at
our disposal to account for the facts. Extant concepts are often inadequate for
the purposes of straight technographic description; and when we try to
broaden the analysis, the gap only widens, as we have just seen with the notion
of efficacy. The present situation is the product of a long history of building in
or embodying technical facts and relationships, which effectively removes them
from the realm of conscious awareness even as they are produced. Or, to quote
Akrich (1989:53), the construction of a technical systemcan be considered
complete only when its effects are denied. Traditional anthropology and
history have perpetuated the denial by developing their impossible, atechnical
conception of society. Perhaps an example in which this conception is shown
clearly to be inadequate will demonstrate the need for a change of approach.
Let us take ancient slavery as that example. This is an old problem, older
even than that of Indias castes, since, according to Finley (1980:35), the first
modern view on the subject was published by Reitemeier in 1789. Two
centuries and several thousand titles later (Miller 1985), the solution seems as
far away as ever. The accumulated corpus of knowledge is impressive, but no
meaningful pattern seems to be emerging: that is the implicit conclusion to
which Finley himself has come at the end of a long life of research devoted in
large part to the problem (1980:66). What, then, is still needed for this corpus,
and our vision of it, to make sense?
The answer, probably, is a good technologists assessment of the situation.
Indeed, until very recently, all historians of Greek economics were divided
449
CULTURE
among three beliefs. The first, to which Finley himself (1965) contributed
liberally, was that technical development in Greece was weak, or at any rate had
prematurely ground to a halt. The second was that slavery was one of the main
causes of this weakness. The third, made famous by Engels, was a sort of mirror
image of the second: slavery tempered technical inadequacy, it was the
consequence rather than the cause of the insufficiency.
But insufficient compared with what? There is no need to be a deep thinker
to realize the paucity of supporting evidence for these beliefs. They are even in
blatant contradiction with the facts. Greece was one of the first regions to
develop the use of iron, not only for ornamentation and weaponry, but also for
tools, including everyday tools such as those used in agriculture. It was in the
Greek and Greek-speaking world that the first rotary-motion machines
appeared: olive crushers, mills, water-lifting devices such as the saqiya, noria,
Archimedes screw, etc., to which must be added new lever and screw presses. If
we go on to count architecture, animal-drawn vehicles, shipbuilding, war
machines, glass-making, etc., the evidence becomes overwhelming. Classical
Antiquity was the theatre of a veritable technical revolution the likes of which
can be found only in the Near East during the fourth millennium BC and in
our present-day technical revolution. That a scholar of Finleys calibre failed to
see this can be explained only by what could be called a cultural blindness that
besets otherwise literate Western intellectuals when it comes to things
technical.
What about slavery, then? It cannot, of course, have been either the cause or
the consequence of a technical stagnation that did not exist. On the contrary,
we are led to ask whether it did not play a positive role in Antiquitys
remarkable technical development, since the two were associated. But what
role? When one looks into a question like the milling of grain, for example, one
notes that, over a period of four centuries, ancient society went from the simple
saddle-quern to the water mill, and at the same time it went from a situation in
which grinding grain was womens work to one in which it was mens work.
Could there then be some relation between the three simultaneous changes: the
technical innovation, the new sexual distribution of labour, and slavery?
Such, at least, is the hypothesis suggested by the facts. For in most societies
the sexual division of labour is one of the components of sexual identity, which
makes any change especially difficult. In particular, a freeman cannot normally
perform womens tasks without ceasing to be a manas was the case of the
famous North American transvestites, berdaches (Callender and Kochems
1983)or without ceasing to be freewhich would have been the outcome in
Ancient Greece. Following this line of thinking, slavery in Antiquity would
have been a systematic means of making men do womens work, which would
explain its vital importance for the economy. Slavery made it possible to take a
number of tasks out of the family or home, to which they had until then been
confined, and to turn them into commercial trades. Moreover, slaves were
bought. Therefore the financial capital they represented had to be made to bear
450
TECHNOLOGY
fruit: technical innovation was the means. In this respect, the same holds true
whether the capital comprises humans or animals. The effort to increase the
returns of animal labour was, for thousands of years, one of the principal
sources of technical innovation; why then should human labour be any
different? And could not grinders, mills, presses and other machines of
Antiquity have been to the domestication of men more or less what the
tribulum, the ard and wheeled vehicles had been to the domestication of animals
three thousand years earlier?
Stated in such succinct terms, the hypothesis can only be incomplete; and
perhaps false as well. But this is not the place to elaborate on it. For our present
purposes the hypothesis is more interesting for its exemplary value than for its
content. It is most remarkable for its total dismissal of the vast historiographic
corpus on slavery in Antiquity mentioned above. In two centuries of assiduous
research on slavery, it would seem that the authors scrupulously examined
every possibility except the one to which all the facts point: that there might be
some relation between slavery, technical innovation and the sexual division of
labour (for a partial exception, see Pryor 1977:4, 234). How can such an
extraordinary blind spot be explained if not by a prevailing vision of society
from which technics are systematically excluded?
THE FUTURE OF TECHNOLOGY
We will make this last remark our conclusion. The anthropology of technics is
just now coming into existence. Of all the branches of anthropology, it is
probably the one that has the most work to do and the fewest workers.
Following a promising start in the nineteenth century, it was nearly abandoned
for three generations. In such circumstances, it is not surprising that
unresolved difficulties and unanswered questions outweigh positive results.
But we must not let this discourage us. For the very gains made by
anthropology in other areas tend to highlight the continuing obscurity of this
branch. In fact, if anthropologists do not want to see their whole discipline
doomed to stagnation, they will have to make an effort to reintegrate technics
into their vision of culture and society. New voices have been heard recently in
support of this view (Reynolds 1983, Beckerman 1985, Pfaffenberger 1988).
What will this new, balanced vision of the future look like? It is too early to say;
we can only suppose that it will be very different from the present one, mainly
perhaps in regard to todays so-called advanced societies, in which an excess of
information and commentary on technological innovation is just as great an
obstacle to understanding as are ignorance and lack of information for ancient
and exotic societies.
Never has there been so much talk about technics, but perhaps never has the
abundance of words had a greater effect of masking reality. One of the perverse
effects of informatics is that users are obliged to expend considerable effort
simply in protecting themselves from irrelevant information (Dennett 1986,
451
CULTURE
Bennett 1988), and this would explain the disappointing growth rate in
productivity (Baily 1986). Those who attempt to think seriously about technics
today find themselves the victims of this effect, which is why we have left it
largely to one side in the foregoing discussion. If we refuse to follow popular
wisdom, which contrasts present-day high technology with, shall we say, the
low technology of a mythical former age, we suddenly realize that there is little
left to say. We do indeed have computers, satellites and some other items that
our ancestors did not, and it is unquestionable that our society has changed
since their time. But so what? Our great-g randparents boasted in
approximately the same terms about their own Wonders of Science (Figuier
186770), and two centuries before them Charles Perrault (1688, I:72) was
already entranced with the prodigious progress of Science and the Arts over
the last fifty or sixty years. In the thirteenth century, Roger Bacon voiced
similar sentiments, which there is no reason to deny to the Romans, Greeks,
Sumerians, and so on. What makes our modernity different from these earlier
versions? We have no answer.
It is no accident that, in this situation, we have almost no alternative but to
turn to history. For perhaps the most direct means of demystifying the image of
modernity that stands between us and the reality of our societies is to look back
at history. That being said, and despite a development out of all proportion to
that of anthropology, the history of technics still falls far short of providing us
with the means of interpreting our present and future with any greater clarity.
Until now, it has managed more to corrode traditional ideas than to develop
new ways of thinking. This is not surprising, since the whole edifice of the
social sciences was built as if technics did not exist or were of only minor
importance. How could technics be made to fit in smoothly? Technology cannot
hope to develop without engendering conflict. The future of technology, as a
social science, is as unpredictable as the future of technics and society today.
REFERENCES
Adas, M. (1989) Machines as the Measure of Men, Ithaca and London: Cornell
University Press.
Akrich, M. (1989) La construction dun systme socio-technique. Esquisse pour une
anthropologie des techniques, Anthropologie et Socits 13(2):3154.
Anderson-Gerfaud, P. (1983) A consideration of the uses of certain backed and
lustred stone tools from late Mesolithic and Natufian levels of Abu Hureyra and
Mureybet (Syria), in M.-C.Cauvin (ed.) Traces dutilisation sur les outils nolithiques
du Proche Orient, Lyon: Maison de lOrient.
Ankei, T. (1986) Discovery of Sak in Central Africa: mold-fermented liquor of the
Songola, Journal dAgriculture traditionnelle et de Botanique applique 33:2947.
Ardrey, R. (1969) Preface to E.Marais, The Soul of the Ape, Cape Town and London:
Rousseau and Blond.
Auzias, J.-M. (1971) La Philosophie et les techniques, Paris: P.U.F.
Baily, M.N. (1986) What has happened to productivity growth?, Science 234: 44351.
452
TECHNOLOGY
Becher, T. (1989) Academic Tribes and Territories: Intellectual Enquiry and the Cultures of
Disciplines, Milton Keynes: Open University Press.
Beckerman, S. (1985) Pour une anthropologie exprimentale: ltude des techniques,
Techniques et Culture 5:13948.
Bennett, C. (1988) Dmons, machines et thermodynamique, Pour la Science 123: 91
7.
Boas, F. (1938) The Mind of Primitive Man, New York: Macmillan.
Brace, C.L. (1980) Australian tooth-size clines and the death of a stereotype, Current
Anthropology 21(2):14164.
Bril, B. (1984) Description du geste technique: quelles mthodes?, Techniques et
Culture 3:8196.
Bromberger, C. (ed.) (1984) Lethnocartographie en Europe, Technologies, Idologies,
Pratiques 4, 14, Actes de la Table ronde internationale organise par le Centre
dEthnologie Mditerranenne, Aix-en-Provence, 2527 November 1982.
Bromberger, C., Hayot, A. and Mariottini, J.-M. (1987) Allez lO.M.! Forza Juve!,
Terrain 8:841.
Callender, C. and Kochems, L.M. (1983) The North American Berdache, Current
Anthropology 24(4):44370.
Callon, M. and Latour, B. (1985) Les paradoxes de la modernit, Prospective et Sant
36:1325.
Cam, F. and Filloux, F. (1987a) Do you speak golden boys?, Libration 23 October:5.
(1987b) La plante des signes, Libration 31 October-1 November:11.
Cazenobe, J. (1983) La Vise et lobstacle, tude et documents sur la prhistoire de londe
hertzienne, Cahiers dhistoire et de philosophic des sciences, 5, Paris: Belin.
(1985) Essence et naissance de llectro-aimant, History and Technology 2(2):
177201.
(1986) Limailles, lampes et cristaux, essai sur les origines de llectronique, Cahiers
dhistoire et de philosophic des sciences, 15, Paris: Belin.
(1987) Esquisse dune conception opratoire de lobjet technique, Techniques et
Culture 10:6180.
Chambers, R., Pacey, A. and Thrupp, L.A. (1989) Farmer First, Southampton:
Intermediate Technology.
Chamoux, M.-N. (1978) La transmission des savoir-faire: un objet pour lethnologie
des techniques?, Techniques et Culture 3 (1st ser.):4683.
(1986) Apprendre autrement, in P.Rossel (ed.) Demain lartisanat, Paris and
Geneva: P.U.F. and Cahiers de lIUED.
Chick, G.E. and Roberts, J.M. (1987) Lathe craft: a study in part appreciation,
Human Organization 46(4):30517.
Childe, V.G. (1936) Man Makes Himself, London: Watts.
Clark, G. (1987) Why isnt the whole world developed? Lessons from the cotton mills,
Journal of Economic History 47(1):14173.
Condominas, G. (1965) LExotique est quotidien, Paris: Plon.
Couffignal, L. (1978 [1963]) La Cyberntique, Paris: P.U.F.
Darr, J.-P. (1985) La Parole et la technique, lunivers de pense des leveurs du Ternois,
Paris: LHarmattan.
Daumas, M. (1965) Introduction, in M.Daumas (ed.) Histoire gnrale des techniques,
II: Les premires tapes du machinisme, Paris: P.U.F.
Deforge, Y. (1985) Technologie et gntique de lobjet industriel, Paris: Maloine.
453
CULTURE
454
TECHNOLOGY
Hahn, E. (1896) Demeter und Baubo, Versuch einer Theorie der Entstehung unsres
Ackerbaues, Lbeck: Max Schmidt.
Haudricourt, A.-G. (1987) La Technologie science humaine. Recherches dhistoire et
dethnologie des techniques, Paris: Editions de la Maison des Sciences de lHomme.
Heider, K.G. (1970) The Dugum Dani, A Papuan Culture in the Highlands of West New
Guinea, New York: Wenner-Gren Foundation.
Herf, J. (1985) Reactionary Modernism: Technology, Culture and Politics in Weimar and
the Third Reich, Cambridge: Cambridge University Press.
Hirschberg, W. and Janata, A. (1980) Technologie und Ergologie in der Vlkerkunde, vol.
1, Berlin: Dietrich Reimer.
Ingold, T. (1988) Tools, minds and machines: an excursion in the philosophy of
technology, Techniques et Culture 12:15176.
Katz, S.H., Hediger, M.L. and Valleroy, L.A. (1974) Traditional maize processing
techniques in the New World, Science 184:76573.
Kawada, J. (1988) Les techniques du corps et la technologic traditionnelle, in J.
Kawada (ed.) Boucle du Nigerapproches multidisciplinaires, Tokyo: Institut de
Recherches sur les langues et les cultures dAsie et dAfrique.
Kimata, K. (1987) Grain crop cookery in South India, in S.Sakamoto (ed.) A
Preliminary Report of the Studies on Millet Cultivation and its Agro-Pastoral Culture
Complex in the Indian Subcontinent, Kyoto: Kyoto University.
Kuhn, T.S. (1985) Scientific development and lexical change, unpublished Ms,
Thalheimer Lectures, The Johns Hopkins University, 1219 November 1984.
Lafitte, J. (1972 [1932]) Rflexions sur la science des machines, Paris: Vrin.
Lamborn, L.L. (1918) Modern Soaps, Candles and Glycerin, New York: Van Nostrand.
Landes, D.S. (1969) The Unbound Prometheus, Technological Change and Industrial
Development in Western Europe from 1750 to the Present, Cambridge: Cambridge
University Press.
Latour, B. and Woolgar, S. (1979) Laboratory Life, The Construction of Scientific Facts,
Beverly Hills and London: Sage.
Le Chtelier, H.-L. (1925) Science et industrie, Paris: E.Flammarion.
Lee, R.B. and De Vore, I. (eds) (1968) Man the Hunter, Chicago: Aldine.
Lemonnier, P. (1981) Le commerce intertribal des Anga de Nouvelle-Guine, Journal
de la Socit des Ocanistes 27:701.
Leplat, J. and Pailhous, J. (1981) Lacquisition des habilets mentales: la place des
techniques, Le Travail humain 44(2):27582.
Leroi-Gourhan, A. (19645) Le Geste et la parole, 2 vols, Paris: Albin Michel.
Lvi-Strauss, C. (1958) Anthropologie structural, Paris: Plon.
(1962) La Pense sauvage, Paris: Plon.
Lowie, R.H. (1934) An Introduction to Cultural Anthropology, New York: Farrar &
Rinehart.
Lubbock, J. (1865) Pre-Historic Times, As Illustrated by Ancient Remains and the
Manners and Customs of Modern Savages, London: Williams & Norgate.
Macaulay, D. (1979) Motel of the Mysteries, Boston: Houghton Mifflin.
Mahias, M.-C. (1985) Dlivrance et convivialit, Paris: ditions de la Maison des
Sciences de lHomme.
Malinowski, B. (1935) Coral Gardens and their Magic, 2 vols, New York, Chicago,
Cincinnati: American Book Co.
455
CULTURE
456
TECHNOLOGY
Perriault, J. (1989) La Logique de lusage, essai sur les machines communiquer, Paris:
Flammarion.
Perrignon de Troyes (1926) Le commerce des grains aux tats-Unis dAmrique,
Revue du Service de lIntendance Militaire 33:93044.
Pfaffenberger, B. (1988) Fetishized objects and humanized nature: towards an
anthropology of technology, Man (N.S.) 23(2):23652.
Pitt-Rivers, A.H.Lane Fox (1906) The Evolution of Culture and Other Essays, Oxford:
Clarendon Press.
Pouillon, F. (1976) La dtermination dun mode de production: les forces productives
et leur appropriation, in F.Pouillon (ed.) LAnthropologie conomique, Paris:
Maspro.
Presutti, W.D. Jr (1987) Productivity: questioning the assumptions for
improvementa role for applied history, History and Technology 3(4):34964.
Pryor, F.L. (1977) The Origins of the Economy, New York: Academic Press.
Rapp, F. (1974) Introduction, in F.Rapp (ed.) Contributions to a Philosophy of
Technology, Dordrecht: D.Reidel.
Reuleaux, F. (1865) Der Constructeur, Braunschweig: F.Viehweg.
(1877) Cinmatique, principes fondamentaux dune thorie gnrale des machines,
trans. A.Debize, Paris: F.Savy.
(1884) Eine Reise quer durch India im Jahre 1881, Berlin: Allgemeiner Verein fr
deutsche Literatur.
Reynolds, B. (1983) The relevance of material culture to anthropology, Journal of the
Anthropological Society of Oxford 14(2):181226.
Riboud, A. (1987) Modernisation, mode demploi, Paris: Union Gnrale dditions.
Rioux, J.-P. (1971) La Rvolution industrielle, 17801880, Paris: Seuil.
Sagant, P. (1987) Traditions enfantines. Lapprentissage des techniques au Npal
oriental, in B.Koechlin, F.Sigaut, J.M.C.Thomas and G.Toffin (eds) De la vote
cleste au terroir, du jardin au foyer, Textes offerts Lucien Bernot, Paris: Editions de
lEHESS.
Salthouse, T.A. (1984) The skill of typing, Scientific American 250(2):949.
Schnaiberg, A. (1982) Did you ever meet a payroll? Contradictions in the structure of
the appropriate technology movement, The Humboldt Journal of Social Relations
9(2):3862 special issue, Socially Appropriate Technology.
Schuhl, P.-M. (1938) Machinisme et philosophie, Paris: P.U.F.
Sebestik, J. (1983) The rise of the technological science, History and Technology 1(1):
2544.
Sigaut, F. (1975) LAgriculture et le feu, Paris/La Haye: Mouton.
(1977) Quelques notions de base en matire de travail du sol dans les anciennes
agricultures europennes, Journal dAgriculture traditionnelle et de Botanique
applique 24(23):13969; special issue, Les Hommes et leurs sols.
(1978) Identification des techniques de rcolte des graines alimentaires, Journal
dAgriculture traditionnelle de la Botanique applique 25(3):14561.
(1980) Un tableau des produits animaux et deux hypothses qui en dcoulent,
Production pastorale et socit 7:2036.
(1985) More (and enough) on technology!, History and Technology 2(2): 11532.
(1987a) Prface, in A.-G.Haudricourt, La Technologie: science humaine, Paris:
ditions de la Maison des Sciences de lHomme.
(1987b) Des ides pour observer, Techniques et Culture 10:112.
457
CULTURE
FURTHER READING
Journals
There are a number of journals devoted to the history of technics, the most widely known
being Technology and Culture. By contrast, there is only one journal specifically dealing
with technology from an anthropological perspective, Techniques et Culture (since 1983).
Other journals are more concerned with objects from a museological point of view, like
Objets et Mondes, published by the Muse de lHomme, Paris.
Recently, however, several journals have published special numbers on technology
where the main issues being discussed in this chapter are considered:
Des systmes techniques, Anthropologie et Socits (Quebec) (1989) 13(2).
Technology in the humanities, Archaeological Review from Cambridge (1990) 9(1).
Savoir-Faire, Terrain (1991) 16.
458
TECHNOLOGY
459