TOULMIN AND THE DISCOVERY
OF HISTORY
Steve Bayne
Abstract: History unlike science is not only a field of
inquiry but a subject matter. Stephen Toulmin's
philosophy of science has not only revealed
new approaches to the relatively young field of
philosophy of science but in this domain has
motivated the discovery of a relatively new
subject matter. I will discuss both achiements,
touching briefly on work by Norwood Russell Hanson,
Dudley Shapere, and others.
First, Toulmin's examination of the nature of
scientific inference leads us to challenge the
distinction between contexts of discovery
and contexts of justification. Second, Toulmin's
own historical examinations uncovers new
history, subject to conceptualization.
In the first instance, I will examine the fate of
the "logic of scientific discovery." I will
maintain that the program failed because its
proponents came to believe they were
doing "superscience." What they were doing,
along Toulmin's lines, was discovering
history - a repository of information that
awaits the opportunism that accompanies
scientific curiosity. In the second instance,
I examine aspects of the influence of Wittgenstein
on Toulmin's own characterization of his
philosophy of science and to some degree
I examine how, along with June Goodfield,
Toulmin pursued the subject matter of history
itself.
I conclude that history is a science of the
future, but a science nonetheless. Throughout I
will stress how sharply Russell's view on the
relation of common sense and science contrasts with Toulmin's
own.
TOULMIN AND THE DISCOVERY OF HISTORY by Steve Bayne (2001)
Awareness of certain facts of biology and physics has always been part
of the human experience. Death as well as matter in motion pre-exist the
species. History, by contrast, didn't come into existence until about four
thousand years ago; and I don't mean the study of history. I mean certain
actual processes of an essentially human nature that must be recognized
in any ultimately satisfying account of the present human condition. R.
G. Collingwood whom Toulmin has frequently cited, points out that it is
not that mankind was before then oblivious to history, or insufficiently
reflective, but rather because "History did not exist." (The Idea of History.
p. 12). There is a very real possibility, I believe, that the accumulation
of history has yet to be discovered applications. Here I shall only lay
out a few road signs based on an understanding of Toulmin's views on scientific
explanation and try to point these road signs in the direction I think
we want to go. So what I am going to do is first take a look at what Toulmin
has contributed to the theory of scientific explanation and then try to
draw some conclusions about that comparatively new discovery which is history.
TOULMIN AND SCIENTIFIC EXPLANATION One of the defining themes of twentieth
century philosophy was the relationship between common sense and science.
Toulmin is explicit about what he takes common sense to be: 'Common sense'=df
'recognizing those regularities with which we are familiar from everyday
experience' (Introduction to Philosophy of Science. 46) Among the logical
positivists, I believe, there were primarily two ways of viewing that relationship.
Some were inclined to believe that science was a complex extension of common
sense; while others felt that common sense led to science and science showed
that common sense could be wrong. This later view was held by Russell,
and Einstein wrote admiringly of Russell's critique of naive realism as
self defeating by appealing to the very science it spawned. Stephen Toulmin,
I contend, found a third way relating common sense and science. It consists
in rejecting both the purely Russellian view as well as the first view
making science an extension of common sense - and here I associate Hempel.
More particularly Toulmin maintains that the relation between common sense
and science will take neither the form of deduction, where no valid connection
can go wrong, nor the form of induction where probabilities don't exclude
the possibility of error. Both views assume the possibility and even the
desirability of formal representation, but it is formalization in the characterization
of scientific explanation that Toulmin attacks. When we explore this third
way relating common sense and science, we see curious parallels between
Russell, who works from within science itself, and Toulmin whose views
are largely a rejection of many of the conclusions of those who don't.
Toulmin has been very much influenced in his program by Wittgenstein and
since Russell and Wittgenstein are frequently compared in order to appreciate
Toulmin's tie to Russell we need to look at how he makes use of Wittgenstein,
while appearing to resist the sort of excesses we find in J. L. Austin.
TOULMIN AND WITTGENSTEIN For Wittgenstein while all explanation comes to
an end (PI:1) it need not. It need not because as long as we require an
explanation to avert misunderstanding further explanation is in order (PI:87).
Toulmin can be read as a studied contrast when he suggests that scientific
explanation not only does but must come to an end. There must always be
some point in a scientist's explanation where he comes to a stop: beyond
this point, if he is pressed to explain further the fundamental basis of
his explanation, he can only say that he has reached rock bottom. FU 42
No one denies that laws are somehow involved in scientific explanation.
Scientific laws for Toulmin are "inference tickets." This conception is
not far from Wittgenstein's who averred that a rule stands "like a sign
post" (PI. 85). Broadly speaking what they have in common is that both
are saying "This (the rule) will get you from here to there." There are
other important similarities between Toulmin and Wittgenstein. For example,
Toulmin places a great deal of emphasis on the use of diagrams in science.
One reason is the power they have for describing relationships that formalisms
either can't show or else presuppose. This is reminiscent of Wittgenstein's
assertion that "we must do away with all explanation, and description alone
must take its place" (PI 109). But there is another similarity between
these two philosophers, one in which Toulmin presses ahead of his teacher,
using his statements like a ladder in order to climb to a higher perspective.
Wittgenstein remarks that When philosophers use a word ...one must ask
oneself: is the word ever used in the language game which is its natural
home. (PI 116) Wittgenstein, I believe, is addressing the metaphysicians
who take words like 'acquaintance' and make them into tools for tasks outside
the language games of every day life. But consider Toulmin's description
of the scientist who extends the notion of 'traveling' in propounding the
principle that light travels in straight lines. Let's take a brief look
at how the scientist extends a word already having a use in the natural
home of ordinary discourse. We are to contrast Robinson Crusoe's discovery
that there is another man on the island he inhabits with the scientific
discovery that light travels in straight lines. Unlike Robinson Crusoe
who, never having seen a foot print on "his" island before, might have
shouted "Eureka! I've just discovered there is another man on this island"
the scientist who theorized for the first time that light travels in straight
lines upon observing the facts from which he drew his conclusion did not
say: "Eureka! I have discovered something accounting for shadows traveling
in straight lines." In the scientific case, the optical case, the actual
inference was from something familiarly commonsensical, such as our experience
with shadows, to something novel -- that is something that can be said
to "travel.' The novelty introduced by something unseen is augmented by
the fact that it is said to 'travel' and in this way the term 'travel'
receives extension. By extending the terms of ordinary language systematically
we see in science a new way of looking at the common sense world from the
perspective of emerging paradigms. But is this so different than Russell's
and Broad's determination to extend expressions for primary qualities of
experience to the unseen causes of veridical perception? Just as Toulmin
accepts extending terms of ordinary language to do "a new job in the service
of physics," don't metaphysicians do the same thing in the service of philosophy?
And isn't this extending of terms just what James was talking about when
he spoke of what we do when we "project words primarily connoting our affections
upon the objects by which the affections are aroused" (Essays on Radical
Empiricism. Nebraska. 143-144). I think the source of Toulmin's greatest
discontent is with the idea that the task of the philosopher of science
is to provide the logical form of a scientific explanation. For Toulmin,
theories that explain are much like maps. One is tempted to say that a
map might be thought of as a set of possible experimental itineraries just
as a drawing of a machine for Wittgenstein can be thought of as expressing
a set of possible motions. Representation is one thing, representation
another. One might ask if scientific laws limit the representational possibilities
available for diagramming in the exactly the same way they limit our interpretation
of a diagram of some machine? I doubt this, but can say no more. What interests
Toulmin is the relation of the data to scientific discovery. Formalized
theories of the "logic of the language of science" will not illuminate
this region; and it's not just formalization, per se. It's cousin in the
pragmatic area, predictability, shares the guilt by association (Foresight
and Understanding p. 24). But for now we need to look at just what it is
about the relation of scientific data to theory that formalism cannot capture.
Here is how Toulmin describes the situation: The fact of the matter is
that we are faced here with a *novel method* of drawing physical inferences*
-- one which the writers of books on logic have not recognized for what
it is. (PS 25) One, perhaps the most important, method of drawing the best
theoretical inference from a theory is the use of diagrams. First, like
Russell, Toulmin is distancing himself from common sense; not entirely,
but in a way that will lay scientific discovery at the feet of paradigms
rather than the feet of common sense from which the scientific ones are
departures. Second, Toulmin excludes common sense appeals to simplicity
(PS33). In his optical case, involving inferring that light travels in
a straight line on the basis of shadows cast and elementary trigonometry,
what he says about diagrams is that ...by the use of diagrams of this kind
it has been found possible to show, and so explain over a wide range of
circumstances and to a high degree of accuracy, what optical phenomena
are to be expected. PS 33 This is a topic in its own right, one I believe
to be directly related to Reichenbach's Kantian treatment of axioms of
physical geometry. But what I need to focus our attention on is this: There
is no deductive methodology for arriving at theories. Indeed, the defenders
of formalization, like Popper, had long argued for a distinction between
discovery and justification. With such a distinction in mind the positivists
could pursue their syntax of the logic of the language of explanation and
the historians could pursue the more interesting but thankless task of
drudging up historical details. But Toulmin by challenging the positivists
on the matter of inference had positioned philosophy of science to dispose
of the justification vs. discovery distinction. But having done that, so
what?! By calling the distinction into question the possibility was raised
that there might be a logic of scientific discovery which could be comprehended
if we examined the history of science, including its errors. What I contend
is that those who posited a logic of scientific discovery were participants
in the discovery of the history of science not, as some of them supposed,
super scientists. Toulmin enriched the nonformalist approach by introducing
paradigms, while Russell went straight forward towards the works of Heisenberg,
Einstein, and Schroedinger. Within the "context" (something very important
to Toulmin) of the arguments in philosophy of science, particularly during
the sixties, Toulmin and Russell share in a studied disregard of problems
such as the confirmation paradoxes of Hempel and the "mindless" axiomatizations,
if such there can be, of biology and relativity. I said I would draw some
conclusions about history based on Toulmin's characterization of science.
Toulmin provides reason to believe that there is continuity in moving from
an attack on the formalist distinction between contexts of discovery and
justification to a reexamination of the way we look at history. He says,
A purely chronological history of science and a purely formal philosophy
of science thus have the same deficiency: each of them neglects to place
the scientific ideas which are in question into their intellectual environment,
so as to show what, in that particular context, gave these ideas and investigations
their merit. (FU p. 109) What better way to access the structure of history,
if it exists, than looking at the history of how we have gone about gainfully
describing the nature of our world? It may be that there is little to be
gained through chronology alone, but just as in Toulmin's optical case
we had to say what it is that does the traveling, our novel move; so too
in history, we have to ask a somewhat analogous question, What is our chronology
to be a chronology of, and even more to the point, why do we need anything
for which there is to be a chronology; just as we ask why must there be
in the optical case anything at all that "travels" in straight lines? My
answer to the first question is that history is composed of trends. My
answer to the second question, why we need individuatable entities like
trends at all, relies on thinking of history much as we would a moving
stream, but not just any kind of stream. The metaphor of the stream will
serve much the same purpose as diagram's in Toulmin's philosophy of science.
Imagine history as a fluid in motion, as the stream of which Heraclitus
speaks. Now let's assume the very opposite of what we want to show. Let
us assume that history is not made up of individuatable phenomena, such
as what I've called "trends." Indeed let us suppose, much like Emile Meyerson
did, that there is no such discreteness as would distinguish historical
processes. In other words, let us assume history is much like what P. W.
Bridgman in his book _The Nature of Thermodynamics_ calls a "uniform fluid."
One thing that Bridgman noticed about such fluids is that there are certain
things that may be going on about which we can have no awareness and therefore
no understanding. Says Bridgman, ...we cannot think of the velocity of
a uniform fluid without imagining the "particles" of which the fluid is
composed...(Harper. p. 10) What Bridgman doesn't call to our attention
is that there is much else we cannot know about a uniform fluid, such as
the direction of any part of the fluid with respect to any other. I have
said that history consists of trends; these are the vortexes of of history.
Philosophers will surely ask, "In the case of a uniform fluid, can we be
sure that there can be a fact of the matter of there being such a fluid?"
Would we suppose the same thing of history; that is, that it might be without
distinction and pass leaving an eventless presence in its wake? This is
the absurd consequence that concludes my reductio. I am going to conclude
with a few polemical remarks on the nature of history, remarks which bear
Toulmin's imprint. Toulmin has pointed out that in science we often extend
the meaning of a term of ordinary language in order to capture an explanation.
His example was 'traveling' in the case of light, the only way to make
sense of his diagram that portrayed a shadow cast from light at an angle.
The sort of extension Toulmin has in mind is very much like extending the
term 'memory' to include history, encompassing a past which is beyond any
living recollection. But history is not just a recollection embedded in
a living culture; it is not just the ordering of past facts, particularly
in the case of intellectual history, which is not simply an ordered list
of past ideas. Nor is history simply a natural history of the human race.
History is a much more complex story of overlapping influences and tendencies.
Unlike physics, the events by which we determine the flow of history, and
here recall the metaphor of a uniform fluid, are the creations of people
and made relevant as phenomena for study by unfolding events, themselves
part of the investigator's own physical reality. Collingwood said it best
when he said that "the historical development of the science of human nature
entails an historical development in human nature itself" (The Idea of
History. p. 84). Vico observed that history is a confused memory. As history
accumulates, I believe that as in the case of science our commonsensical
albeit confused recollections of our own lowly origins will add dignity
to who we once were. Broad believed he discerned a competition between
physics and death and psychology and life; Toulmin offers us physics as
life and places mind in nature in living historical perspective.