II. ANALYTICAL APPROACHES
An examination of substantive theories of historical
causation shows that
many of the problems raised by
them tend to shade into others that concern,
not so
much matters of fact, as the meaning and implications
of
various key terms or ideas. This aspect of the subject
first began to
attract widespread attention during the
latter half of the nineteenth
century, when a certain
skepticism regarding the feasibility of large-scale
in-
terpretations of the historical
process on allegedly
scientific lines set in, and when it was felt to be
neces-
sary to consider more carefully the
actual structure of
the concepts by which the explanation of
historical
phenomena was customarily attempted. It was clear,
for
instance, that the program of elevating history to
scientific status
presupposed that historical events
could be subsumed beneath laws and
hypotheses of the
type that had been employed with success at the
level
of natural phenomena. How far, though, was such an
assumption
really justified? Was it not conceivable that
the whole notion of
explanation and understanding
within the field of the human studies
precluded the
adoption of such an approach, with the consequence
that
some of the grandiose attempts which had been
made to set history upon the
“sure path of a science”
could be regarded as
mistaken in principle, the prod-
ucts of a
profound categorial confusion? In any case,
was it not reasonable to
investigate the logical charac-
ter of the
explanations which historians in practice used
before embarking on projects
whose relevance and
applicability to the subject matter of historical
enquiry
had been taken for granted not critically ascertained?
The Notion of Cause.
The original stimulus of much
modern controversy concerning the
nature of historical
thought and understanding derived from the work
of
writers who were themselves deeply opposed to the
suggestion that
the explanatory procedures appropriate
to the human studies (or Geisteswissenschaften) were
not significantly
different from those typically adopted
in the natural sciences. For both
the German philoso-
pher, Wilhelm Dilthey,
and the Italian, Benedetto
Croce, it was the dissimilarities rather than
the simi-
larities between historical and
scientific conceptions of
enquiry that were important and striking. Thus
each
tended to emphasize the individuality or uniqueness
of historical
phenomena and to lay stress upon the need
to regard the events of history
in a light that presented
them as integrally related to the specific
periods and
social milieus within which they took place: human
nature
was not conceived to be something static, the
subject of omnitemporal laws or regularities, but was
viewed as
involved in a continual process of change,
finding expression in forms of
thought and feeling that
were in turn reflections of diverse patterns of
life or
culture. Moreover, the concept central to history was
identified as that of human agency, and the under-
standing or inner recognition of what it is to be
an
agent, pursuing purposes or adhering to practical prin-
ciples and beliefs, was held to bring into play a
variety
of explanatory concepts and modes of exegesis that
were
without analogue in disciplines whose province
was the nonhuman world. Many
of these themes were
adumbrated with exceptional clarity and force by
the
British philosopher, R. G. Collingwood, and it was
largely through
his persuasive writings that they subse-
quently came to achieve wide currency in Anglo-Saxon
circles.
Collingwood was especially insistent upon a
proper appreciation of the role
played by thinking in
determining historical phenomena: “all
history,” he
once affirmed, “is the history of
thought” (Idea of
History, pp. 214-15).
He went on to argue in accord-
ance with this
dictum that the notion of cause was
employed in the historical sphere with
a distinctive
meaning, a meaning that rendered misconceived and
futile
attempts to assimilate history to natural science.
For him, the cause of a
historical event was “the
thought in the mind of the person by
whose agency
the event came about,” the historian's
understanding
of such an event therefore consisting in the reconstruc-
tion or
“reenactment” of the process of thinking from
which
it issued. Such a model of the historian's proce-
dure was to be sharply contrasted with the mode of
explanation
Collingwood attributed to the scientist,
whereby particular occurrences
were shown to be
intelligible in virtue of their exemplifying generaliza-
tions correlating them with
other events of a specific
type.
The conception of historical understanding illus-
trated by the work of Collingwood has not been with-
out its critics, many of whom have considered his
claim
that such understanding is radically different in form
from
scientific kinds of explanation to be unwarranted.
What has come to be
known as the “covering law
model,” which was
developed in the thirties and forties
of the present century by a number of
logical empiri-
cists and
which—so far as history is concerned—has
achieved
classic expression in the work of Carl Hempel,
explicitly rests upon the
contention (resisted by Col-
lingwood) that
any adequate explanation of a causal
type must necessarily exhibit the
event to be explained
as instantiating some general law or laws; when
strictly
interpreted, this was held to imply that the explanan-
dum should be deducible from a set
of premisses con-
sisting, on the one hand,
of statements descriptive of
initial or boundary conditions and, on the other, of
further
statements expressing well-confirmed universal
hypotheses. Proponents of
the view in question, which
in its general outline conformed to the account
of
causality originally suggested by Hume, argued that
historical
explanations, when their basic structure was
fully revealed, displayed no
significant divergences
from those used in other fields of enquiry; indeed,
it
was only on an analysis along the lines proposed that
the
historian's manner of making the past under-
standable could be appreciated as implying an inher-
ently rational procedure, subject to the sort of
check
and verification that was a precondition of the respect-
ability of any empirical
discipline. Talk of empathetic
projection into the minds of historical
agents, such as
was indulged in by theorists like Dilthey and Colling-
wood, might have value as signifying
a heuristic aspect
of the historian's method of arriving at his
explanations:
so far, however, as the problem was one of elucidating
the fundamental logic of these, it amounted to little
more than a
mystifying irrelevance.
Despite its initial plausibility, and notwithstanding
the ingenuity and
pertinacity with which its supporters
have sought to defend it, the
covering law analysis has
in its turn encountered criticisms, two of which
may
be mentioned here. Thus from one standpoint it has
been contended
that the explanations historians actu-
ally
provide simply do not measure up to the stipula-
tions embodied in the proposed model: the average
historian
would be hard put to cite the universal hy-
potheses upon which the meaning and validity of his
causal
propositions allegedly depend. The model would
therefore seem to require
(at best) considerable quali-
fication and
amendment if it is to serve as an adequate
framework within which to
characterize how historians
in fact proceed. From another standpoint it has
been
urged that the presentation of historical events as ra-
tionally intelligible in the light of the
motives, aims,
and beliefs of the agents involved constitutes an in-
trinsic and ineliminable feature of
historical under-
standing, and that this
feature cannot be satisfactorily
accommodated within the limits set by the
covering
law theory. Hence there has been a tendency on the
part of
some recent writers—notably, W. H. Dray—to
try to
reformulate what they hold to be the essential
points of Collingwood's
antipositivist position in a way
that shows them to be both
epistemologically un-
mysterious and
empirically sound. Among other things,
it has been suggested that a crucial
characteristic of
the causal accounts offered by historians consists
in
demonstrating that the actions of which they treat were
rationally
justified or required from the point of view
of the agents concerned,
rather than that they were
occurrences to be expected or predicted on the basis
of inductively established uniformities. Whether, if it
is to
be finally adequate, an interpretation of this kind
can really dispense
with an appeal—at least at some
level of the
analysis—to the notions of law or general-
ization remains a disputed question, the answer
to
which would appear partly to depend upon the resolu-
tion of wider and still controversial issues in the
philos-
ophy of mind.
Freedom and Determinism.
An advantage that is
sometimes claimed for the approach favored by
Col-
lingwood and his modern followers
is that it does not
imply that there is any incompatibility between re-
garding an action as explicable and treating
it as a free
one. For it is argued that to explain what a historical
agent did by referring to the good, or even compelling,
reasons that he had
for doing it does not commit the
historian to maintaining that the agent's
recognition
of these reasons rendered his action inevitable. And
this
result may be contrasted with the consequence of
adopting a covering law
analysis. In the case of the
latter, it is held, an action is said to have
been explained
if and only if it has been shown to have followed
necessarily, as a law-governed effect, upon the fulfil-
ment of specific initial conditions. Thus the
covering
law theorist, insofar as he considers historical events
to be
capable of explanation, cannot avoid adopting
a deterministic position.
One reply to this has been that accounts of historical
causation often take
the form of indicating certain
necessary, as opposed to sufficient,
conditions for the
occurrence of an event, and that a covering law theo-
rist can without difficulty adapt his
analysis to accom-
modate explanation in
this sense. On this (modified)
view, his adherence to the postulate that
historical
events are explicable need not be interpreted as carry-
ing deterministic implications. But it has
also been
suggested that there is in any case no justification for
holding that the acceptance of determinism logically
excludes belief in
human freedom; even if a historian
assumes all human behavior to be
susceptible in prin-
ciple to explanation in
terms of sufficient, and not
merely necessary, conditions, he is not
thereby de-
barred from supposing that the
subjects of his enquiries
sometimes acted as free and responsible agents.
Thus
the traditional fear of determinism, which (as was seen
earlier)
often helped to inspire resistance to certain
speculative theories of
history, is based upon an illusion
that largely derives from an
illegitimate identification
of causation with such notions as those of
coercion and
external constraint: it is possibly with this argument
in
mind that the contemporary British historian, E . H.
Carr, has written that
the “logical dilemma about free
will and determinism does not
arise in real life,” human
actions being “both free
and determined” (What is
History?, p. 124). There are, however,
others—among
them Isaiah Berlin—who have found it
unconvincing
and who have felt that all attempts to analyze concepts
like freedom and moral
responsibility in such a way
as to make their employment compatible
with a thor-
oughgoing causal determinism
ultimately fail to do jus-
tice to the
implications of ordinary thought and lan-
guage. In their view, moreover, the fabrid of the
historical
studies, as we customarily know and under-
stand them, is shot through with libertarian and evalu-
ative conceptions to a degree that has not always
been
adequately appreciated. Hence those who have con-
tended that a commitment to unrestricted determinism
in human affairs would entail sweeping revisions of the
vocabulary and
categories the historian normally brings
to the interpretation of his
material have been substan-
tially right.
Though this is not of course positively to
affirm that a thoroughgoing
determinism is untenable,
it is to claim that for the most part historians
habitually
write and think as if it were untenable. And that is
a
point which, if correct, cannot be lightly brushed
aside.