The Rules of Sociological

parts of what there is, and in some respects they are best 
understood from a less detached standpoint' 
.59 
Then you will enter 
seriously into the problems of interpreting social life. You may 
then hold, with Clifford Geertz, that the enterprise is rather one of 
'thick description', essentially involving imaginative interpretation 
of actors' interpretations, where the observer is faced with 'a 
multiplicity of complex conceptual structures, many of them 
superimposed upon or knotted into one another, which are at once 
strange, irregular, and" inexplicit. and which he must contrive 
somehow first to grasp and then to render'. It is 'like trying to read 
(in the sense of "construct a reading of') a manuscript'. On 
Geertz's view, 'anthropological writings are themselves interpreta­
tions, and second and third order ones to boot': the anthropolog­
ist's business is to 'construct actor-oriented descriptions of [say] 
the involvements of a Berber Chieftain, a Jewish merchant, and a 
French soldier with one another in 
1912 
Morocco'. The line, he 
holds, 'between mode of representation and substantive content is 
as 
undrawable in cultural analysis as it is in painting'. 
60 
At all events, you will have to face the question of how to 
evaluate success at interpreting the meaning of actors' interpreta-, 
tions for them and how, in particular, to escape (or at any rate 
escape the viciousness of) the hermeneutic circle - the fact that 
particular interpretations or readings" only make sense within a 
total interpretative framework, which is in turn made up of 
particular interpretations.61 You will also have to face the inevit­
able question of whether, and if so hQw, relativism may 
be 
avoided 
- relativism being the thOUght that there may be multiple and 
divergent points of view but no way of relating them to any 
independent reality or to one another that it is not just the view 
from one of them. 62 
Durkheim, so far as I know, never entertained this thought. Nor 
did he ever in his methodological writings seriously consider what 
one might call the hermeneutic dimension of social inquiry: the 
problem of how to achieve 
and. 
evaluate success in the interpreta-

---

Introduction 15 
tion of the actors' world from within. This might seem strikingly 
odd, since, as we have seen, from 1895 onwards he laid so much 
stress upon the representational and psychological character of 
social life. He avoided the problem - indeed he avoided even 
raising it - by continuing to reason as though social facts and 
collective representations were brute facts, causally related to 
others, identifiable through 'external' definitions, making no 
reference to subjective or inter-subjective meanings. So it was that 
he could write a book on suicide which nowhere addresses the 
question of ho\\:' people in different societies, groups and situations 
view suicide, or the meaning of death, or life. In so far as he 
practised his own methodology, he sought, as in 
Suicide, 
to 
account for social facts, identified 'externally', by causal analysis, 
borrQwed from physics, chemistry, physiology and medicine. 
Yet such an approach, dictated by an obsession with an 'abso­
lute' conception of objectivity, can only be a sterile prescription 
for the human sciences, which must in large part be hermeneutic in 
character. The science of social facts cannot, as Durkheim recom­
mends in his review of Labriola and his debate with Seignobos, 
bypass actors' conceptions: it must seek to interpret them and 
relate them to one another, for they are in large measure 
constitutive of its very subject matter. Fortunately, indeed inevit­
ably, Durkheim's practice illustrates this very truth. Thus his 
definitions of crime, punishment, education, socialism, religion, 
and of types of suicide are, because they are bound to be,