Abstract: During the period when positivism was dominant scientific methodology and its products were believed to rest on ahistorical principles. More recent work in the sociology, philosophy and history of science has undermined this belief. Modern science studies are based on the realization that the nature of any actually existing science will be profoundly determined by the historical conditions of its existence. Hence the objects of scientific research must be regarded, not as purely natural, but as historical facts. This even applies to empirical data because they are heavily dependent on theoretical preconceptions and methodological constructions. The historical understanding of a science like Psychology is thererfore no mere garish but is concerned with the essential nature of the discipline.
Half a century ago virtually everyone regarded the products of natural science,
its discoveries and findings, as being different from other products of human
activity in that they transcended the mundane circumstances of their production.
Although scientific results were of course produced by people of flesh
and blood, working under specific historical conditions, the results themselves
were generally seen as independent of these origins once they were received into
the canon of genuine science (see however Fleck, 1935).
Such beliefs were reflected in the philosophy of science which predominated in
the Anglo-Saxon countries at that time.
It had been a primary concern of the logical positivists to set down
criteria by which one could distinguish science from non-science.
A standard philosophical rationale for the purity of science was based on
an explication of the purity of its language.
It was said that in the language of science every statement could be
either derived by necessary deduction from first principles, and/or it could be
demonstrated to be true by reference to direct empirical observations.
By contrast, the mundane language of everyday life was sloppy and
ambiguous; most of the statements it contains were neither derivable by any
process of strict logical deduction, nor were they unambiguously verifiable by
American psychologists were understandably fascinated by a philosophy that
promised to end the persistent doubts about the scientific status of their
field. By spelling out the criteria
of scientificity in a very explicit way logical positivism gave psychology a
clear set of conditions whose fulfillment should result in broad recognition of
the psychologists' claims to the status of scientists (Mandler and Kessen,
1959). This philosophy seemed to
provide the discipline with a prescription for how to become a genuine science,
and in the next few decades this prescription was quite thoroughly internalized
by large sections of the discipline (Toulmin and Leary, 1985).
In the meantime the world has changed.
Attitudes to science have become, on the whole, more ambivalent than they
were half a century ago. (When I
say "science," throughout this presentation, I am of course using the term in
its usual Anglo-Saxon sense of "natural science").
This ambivalence is a consequence of the way in which certain negative
aspects of the application of science intruded on people's consciousness,
whether in the form of weapons of mass destruction or in the form of
environmental damage or threats to health as everyday facts of life.
Inevitably, such developments have tarnished the once unblemished image
of science in people's minds.
As might have been expected, these developments have been accompanied by a
change of fashions in the philosophy of science, a change that had already been
placed on the agenda by the internal difficulties of logical positivism.
In the post-positivist phase that has prevailed during the last quarter century
the differences between science and non-science have not seemed nearly so great
and so clear cut as they seemed before, and the need to make this distinction
has lost some of its urgency. A new
field of social studies of science has flourished, and this field takes
scientific activity as an object of sociological examination in exactly the same
way as any other human activity. [Knorr-Cetina (1984) provides one well known
example for such studies, but the English language literature in this field is
by now enormous]. Doing science is
now seen as being as much a matter of social organization, competition for
scarce resources, social interests, rhetorical persuasion and consensus building
as many more mundane forms of human activity.
Thus is science robbed of its moral exclusiveness.
Perhaps we should see this disenchantment of science as the last stage in
the "disenchantment of the world" that Max Weber diagnosed as being
characteristic of the modern period.
Within Anglo-Saxon philosophy of science the process of disenchantment
manifested itself in a turning away from the relentless formalism that had been
so characteristic of the period dominated by logical positivism.
The old philosophy had tried to establish the purity of science on the
basis of the purity of its language.
This was largely based on a strict separation of an empirically grounded
data language and a formal theoretical language.
Unfortunately, this distinction could never be rigorously justified and
soon became a casualty of powerful philosophical attacks (Achinstein, 1968;
Hesse, 1970; Hanson, 1959; Quine, 1953).
There is no need for me to cover this well known ground here.
Suffice it to mention that the empirical statements of experimental
science are never formulated in a language that simply refers to sensory
impressions but in a theory loaded language.
When they are talking as scientists, physicists do not talk about
displacements of light points in their visual field, they talk about measuring
the length of light waves. But
reporting on the length of light waves already presupposes a wave theory of
light as well as some theoretical understanding of the measurement process and
the instruments it relies on.
Psychologists do not empirically report marks on paper but scores on an
intelligence or personality test.
That presupposes a massive framework of theoretical presuppositions that make
possible the identification of certain pieces of printed paper as intelligence
tests and other pieces as personality tests, not to speak of the battery of
assumptions and decisions that lies behind the concept of a "score" (Danziger,
Thus, the so-called empirical statements that occur in scientific reports and
texts depend as much on particular theoretical frameworks as they do on the
sensory experience of individual scientists.
But it is not only that some kind of theoretical framework is necessary
for formulating an empirical result, it is also the case that the appearance of
the empirical result depends on prior methodological decisions.
Experimental facts are not usually discovered just lying around the floor
of the laboratory or blown in through the window by the wind.
They have to be painstakingly constructed with the help of complex
instruments and carefully thought out procedures.
But the use of one set of instruments and procedures rather than another
depends on decisions and assumptions that are certainly underdetermined by any
list of scientific facts known at the time.
In a very real sense, empirical data are not the starting point of
science, rather they are the yield that science ends up with as a result of an
elaborate process of production (Bhaskar, 1978).
It is clear that if we want to understand this process of production we will
have to pay a great deal of attention to the underlying theoretical assumptions
and the methodological choices that play such an important role in the
constitution of empirical data. No
longer is it permissible to relegate questions about the genesis of theoretical
assumptions to some limbo of irrationality, called the "context of discovery,"
that was created with the express purpose of preserving the purity of scientific
rationality in the form of a "context of justification."
Real science does not conform to this hopelessly idealized image.
The rational and the irrational are thoroughly entangled, or, to put it
another way, the rational side is not nearly so rational and the irrational side
not nearly so irrational as the older view would have had us believe.
What then is the origin of the implicit and explicit theoretical assumptions and
methodological choices that co-constitute the empirical yield of science?
Part of the appeal of the idealized view of science was due to its heroic
image of the individual scientist.
Scientific activity was the arena of a tournament in which the ingenious man of
science (generally it was always a man) pitted his wits against nature
(feminine, of course) who was reluctant to yield up her secrets.
The products of science therefore depended on three sources:
Nature herself, specific psychological qualities of the individual
scientist, and certain principles of scientific rationality and morality that
regulated the scientist's practice.
These last were conceived ahistorically - they were basic rules of logic, like
the rule of non-contradiction, and basic rules of ethics, like the rule of
honesty, for example, that had not changed since the time of the Ancient Greeks
Such an account could be regarded as useful only as long as its function was
regarded as prescriptive rather than descriptive.
Of course people knew that the actual practice of science included
features not covered by this account, but those features were relegated to the
history of science, a field of messy detail that was quite separate from the
lofty world of principle inhabited by the philosophy of science.
However, this separation has not lasted.
As the interest in science shifted from a prescriptive interest in an
idealized science to an interest in the actual practice of science as a human
activity so the study of the history of science has become crucial for an
understanding of the nature of science.
Once we see science, not as an idealized abstraction, not as a set of
disembodied propositions, but as an activity engaged in by real flesh and blood
people, then the heroic image of the individual scientist confronting nature
collapses. The timeless canons of
scientific method turn out to be not timeless at all but subject to massive
historical change as well as profound variation that depend on local conditions
and traditions (Danziger, 1990b; van Strien, 1990a).
The epistemic access to nature that science provides is always a
collective access, and the arena within which the ingenuity of the individual
scientist is allowed to operate is an arena constituted by social groups whose
life and whose struggles are subject to the same mundane constraints as are
those of other social groups.
Before the raw givens of nature can become data for science, can become the
kinds of things that science can actually work with, they have to be transformed
by the collective activity of human investigators.
This transformation is not only material - chemical substances have to be
purified, for example - but also conceptual.
I have to think of the white powder in front of me as a chemical compound
with a certain molecular structure before I can set to work on it as a chemist.
As P. van Strien has reminded us:
"Facts are always interpreted facts.
The development of science depends on the success of competing traditions
in constructing a plausible account in which these "facts" are interpreted in
the light of theory" (van Strien, 1990a, p.39).
In other words, insofar as objects are objects for science they exist
within a certain conceptual framework.
But the conceptual frameworks of science, as we know very well, have
undergone much change in the course of history.
They are historical products.
It follows that the objects of science which only exist within these
frameworks (and take their meaning from them) are also historical objects that
change in the course of human history (Danziger, 1992).
To quote the German philosopher of science, Kurt Hübner:
most of the objects that science has dealt with in the course of its history,
objects which appear ostensibly to be the same, really bear only a family
resemblance to one another. Whether
it be space, time, the starry heavens, the forces which move bodies, or some
other object of science, we would look in vain for some shared or common meaning
which might apply to any of these objects throughout their respective histories
and which as such . . . might serve as the common and continual ground for all
the scientific theories devoted to any such object.
It was hard enough for mankind to grasp that the same time does not tick
off in all parts of the world. It
may be even more difficult to grasp that when we investigate some scientific
object, both today and as it existed in the past, we are not necessarily
speaking about one and the same thing (Hübner, 1979, p.218).
Hübner was speaking of the objects of physics.
But if the objects of physics must be regarded as embedded in human
history, how much more obvious is this in the case of the objects of psychology.
The memory that a contemporary student of the area investigates is not
the same object as that which Ebbinghaus tried to study by means of nonsense
syllables, and neither of them has more than a tenuous connection with memory as
understood by Aristotle (Danziger, 1990c).
The individual differences that an Eysenck, for example, believes to
constitute objective features of the world have nothing in common with the
individual differences pondered by a Carl Jung.
The "behaviour" studied by the "behavioural science" of the recent past
is a very different object from that which inspired John B. Watson or Lloyd
Such historical changes are due to changes in the framework within which
different generations of scholars and scientists have operated.
But such changes of framework are embedded in a general historical
situation that includes the values, the implicit assumptions and the social
interests of groups of investigators as well as their placement in a broader
sociocultural context whose influence they cannot escape.
Thus, if the objects of science necessarily exist within some theoretical
framework, and if such a framework is always part of a broader historical
context, it follows that the objects of science are historical objects.
But in order to arrive at an adequate understanding of historical objects
we must engage in historical studies (Jüttemann, 1986).
That is why history can legitimately lay claim to a certain priority when
we try to understand what it means to study some topic scientifically.
The history of science has, as philosophers like Hübner have pointed out, a
propadeutic function. Among other
things, it provides us with "a standard against which to judge the scope,
validity, and applicability of the methods, principles, postulates, etc., that
have been worked out by scientific theoreticians."
It also has an important critical function that counteracts the
"degeneration" that often sets in when a particular scientific position becomes
generally accepted. The position is
soon considered "self-evident" and eventually becomes something that can no
longer be seriously questioned (Hübner, 1979, p.94).
In psychology one can most readily detect evidence for such a
degenerative process on the level of methodology (Gigerenzer and Murray, 1987).
Contemporary philosophers who stress the propadeutic function of history of
science had some notable predecessors, for example, Dijksterhuis (1961) and
Duhem (1954). The latter believed
that "the legitimate, sure and fruitful method of preparing a student to receive
a physical hypothesis is the historical method," because, as he put it, "to give
the history of a physical principle is at the same time to make a logical
analysis of it" (Duhem, 1954, p.268-269).
Duhem advocated the historical approach even though he believed that
physical concepts became "perfected" in the course of history, in other words,
that later versions were in some real sense better than earlier versions.
This is a position that is easier to maintain in physics than in
psychology. Duhem showed that even
though the more recent position might be more satisfactory scientifically one
could not really understand scientific development without introducing the
In psychology the superiority of more recent positions is often not as obvious
as it is in physics. How much more
reason then to turn to historical considerations when trying to assess the
status of recent "advances" in psychological theory.
Of course, this means running the risk of discovering for oneself the
truth of the old dictum that those who ignore history are condemned to repeat
it. But that seems preferable to a
state in which one lacks the means to distinguish real progress from the
illusion of progress.
But is it still possible to make such a distinction if one historicizes science
to the extent that I have? Let me
try to answer by invoking an analogy.
In the traditional view the process of science was seen rather like the
painting of a picture, a landscape perhaps, that was gradually being represented
on canvas. Both the landscape and
its pictorial representation were vast and complicated, but over a long period
of time more and more details would be filled in, until one day the picture
would be finished, and we would be able to enjoy the perfection of the final
To-day we know that we have to make some changes in this analogy.
The picture we are confronted with rather resembles one of those canvases
that has been painted over many times by different painters.
After the first one each of them has reacted both to the work of his
predecessor and to the landscape as he saw it.
Of course, they didn't all see it in the same way, and in painting it
they pursued different purposes.
They learned from the techniques of their predecessors, but they also reacted
against their apparent inadequacies.
Each of them thought of his predecessor's style as old-fashioned and of
his own work as advanced. This
process need never end. But from
the fact that there may always be a fresh artistic style, it does not follow
that there exists no landscape to be painted.
It also does not follow that one painting is as good as another.
Some are better. But that is
a judgment which is only possible on the basis of certain criteria which change
historically. Does this
change constitute progress? The
problem is one of definition. There
is no such thing as absolute progress, only progress in this or that sense.
And that means that progress is often ambiguous - which is exactly the
conclusion many people have arrived at with regard to scientific progress, or
economic progress for that matter.
In concluding my argument I cannot improve on the words of P.J. van Strien: "It cannot be denied," he writes, "that awareness of the history of current theories also leads to a sense of the relativity and temporality of all theories. Of course, this will only lead to resignation if we are still hoping for timeless Truth. Problem situations are always historically relative situations. Can we then expect the answers to be context free?" (van Strien, 1990b, p.313).
Originally appeared in
Psychologie und Geschichte,
Achinstein, P. (1968).
Concepts of Science.
Baltimore: The Johns Hopkins
Ben-David, J. (1971).
The Scientist's Role in Society.
Englewood Cliffs, N.J.:
Bhaskar, R. (1978).
A Realist Theory of Science.
Danziger, K. (1990a).
Constructing the Subject:
Historical Origins of Psychological Research.
New York and Cambridge:
Cambridge University Press.
Danziger, K. (1990b).
Die Rolle der psychologischen Forschungspraxis in der Geschichte: Eine
E.G. Wehner & A. Schorr (Eds.),
Danziger, K. (1990c).
metaphor and the history of psychological discourse.
In D.E. Leary (Ed.) Metaphor in
the History of Psychology.
York and Cambridge:
Danziger, K. (1992).
Objects, Practice, and History.
Annals of Theoretical Psychology, vol. 9.
Dijksterhuis, E. J. (1961).
The Mechanization of the World Picture,
trans. by C. Dikshoorn.
Duhem, P. (1954).
The Aim and Structure of Physical Theory.
Princeton University Press. (Trans. of 1914 edition of La Théorie Physique:
Son Object, Sa Structure).
Fleck, L. (1935).
Entstehung und Entwicklung einer
wissenschaftlichen Tatsache: Einführung in die Lehre von Denkstil und
Benno Schwabe & Co.
Gigerenzer, G., and Murray, D. (1987).
Cognition as Intuitive Statistics.
Hillsdale, N.J.: Erlbaum.
Hanson, N. R. (1958).
Patterns of Discovery.
Hesse, M. (1970).
Is there an
independent observation language?
The Nature and Function of Scientific
Theories, ed. R. G. Colodny.
Reprinted in M. Hesse,
Reconstructions in the Philosophy of Science.
Indiana University Press, 1980.
Hübner, K. (1979). Kritik der
wissenschaftlichen Vernunft, 2. Aufl.
Jüttemann, G. (1986) Hrsg.
Die Geschichtlichkeit des Seelischen:
Der historische Zugang zum Gegenstand der Psychologie.
Knorr-Cetina, K. (1984).
Die Fabrikation von Erkenntnis:
Zur Anthropologie der Naturwissenschaft.
Mandler, G., and Kessen, W. (1959).
The Language of Psychology.
Quine, W. (1953).
Two dogmas of
empiricism, in From a Logical Point of
View. Cambridge, Mass.:
van Strien, P. (1990a).
historical practice of theory construction.
Heymans Bulletins Psychologisch
Instituut R.U. Groningen.
van Strien, P. (1990b).
Recontextualization as a contribution of history to theoretical psychology.
In W. Baker, M. E. Hyland, R. van Hezewijk, and S. Terwee (eds.),
Trends in Theoretical Psychology.
Toulmin, S., and Leary, D. E. (1985). The cult of empiricism in psychology, and beyond. In S. Koch and D. E. Leary (Eds.) A Century of Psychology as Science. New York: McGraw-Hill.