Comparative Anthropology and Action Sciences

An Essay on Knowing to Act and Acting to Know

S.N. Balagangadhara

Introduction
Should action sciences exist, it is obvious what they would have to study; the
nature of human action, the kind of knowledge that actions generate, the
process of learning to perform different types of action and so forth.
Comparative anthropology, as a discipline, studies and contrasts, where such
contrasts are possible, the different ways in which human beings organize
their lives, think about and experience both themselves and world around
them. A culture is a form of life and, as such, it is a way of going about the
world. One would therefore expect to find a deep and intimate relation
between action sciences on the one hand, and studies of culture on the other. A
perusal of literature, however, shows the fact of the matter to be different.
Different in which way? It is not necessary to speak here about the attempt
of anthropologists to relate both their theoretical and ethnographic work to
theories of action; many anthropologists and few anthropological quasi-schools
have felt and continue to feel the need to integrate theories of human action in
their approaches to the study of society. The same, it would appear, cannot be
said of action theories. As a rough way of establishing the interest that action
theorists show in this matter, let us notice the kind of themes they take up for
investigation; the relation between action and intention, between actions and
events; the nature and properties of efficient and inefficient action, the rational
and irrational action etc. A great deal of familiarity with the literature in this
tradition is not required to observe that neither anthropological theories nor
ethnographic studies play any role in their discussion.
Why is this so? Here is one possible answer; the problems that are taken up
for scrutiny are invariant across cultures and, consequently, there is not much
that anthropology can contribute to this enquiry. This stance is implicitly
assumed by most; perhaps, if called for, they would also defend it explicitly.
There is, however, a second way of looking at this issue; the themes that action
theorists address themselves to are the deep and ground intuitions of one
specific culture, viz. The West. To the members of this culture, it is obvious
that human action, by virtue of being human, exhibits some typical, speciesspecific properties such as being intentional, goal directed, rational or
whatever else that you may want to attribute. Not only is it natural speak
about human action in these terms, but to look at it any other way would be so

deeply counterintuitive that it does not appear plausible. In so far as most

western anthropologists share this basic intuition, there is not much in their
field work that could contribute to the discussion. After all, they too are framed
within the descriptive possibilities open to this culture.
While both these answers shed some light on why action theories are
indifferent to anthropology, the second carries a rather disturbing implication:
It suggests that what the action theories study are not so much human actions
(in their species typical generality) as a culture-specific mode of acting. And
that there are other modes of acting and other ways of going about the world
which may not exhibit any/most of the properties that human action are
supposed to have. Consequently, if action sciences are to exist as sciences at
all, it is advisable that they look at the way other cultures act and at other
forms of life.
In this paper, I want to examine whether the second answer could turn out
to be both intelligible and sensible. (I shall leave the question of its truth
aside.) I shall do so by very briefly looking into a notion that pertains to the
domain of action theory viz., action-knowledge. I will suggest that the notion of
action-knowledge within the western tradition, by and large, has come to mean
knowledge about actions. My claim will be that this is not what actionknowledge is. In such a case, we need to understand why action-knowledge
has come to mean what it means. Lending credibility to this claim requires that
we develop a more adequate notion of action-knowledge. Both purposes could
be fulfilled, I shall suggest, by looking into the practices across cultures. In the
following pages, I shall endeavor to lend some plausibility to these suggestions
by looking specifically into one culture, viz., India.
To provide a degree of intelligibility to these claims is the aim; detailed
arguments and elaborate defences are as much outside my ken as they are
beyond the province of this essay.

The structure of the paper

This paper has six sections. In the first, I introduce the problem that, more or
less, forms the thread that runs throughout: Why is there an absence of
theoretical treatises about social practices in a culture like India, when it has
got what is manifestly the most complex form of social organization that
human history has ever known? This question is not answered directly, but
functions as the background within which other issues are raised. In the
second section, I suggest that learning is related to culture and examine one of
the elements, viz., the experience of order, which plays an important role in the
kind of learning process that a culture develops. In the next two sections I
outline a notion of learning and claim that as a mode of learning it is specific, if
not unique, to Indian culture. The properties of such a learning process, which

creates action-knowledge, are examined in the third section. While at it, some
empirical hypotheses are generated to show the heuristic potential of this
approach. In the two parts that comprise the fourth section, I take up the
problem of comparison: How culture specific is the notion of actionknowledge? I examine one domain that belongs to the realm of human action,
viz., the ethical in the first part. In the second, I compare the notion of actionknowledge with the most familiar notions of knowing how and knowing that.
The problems that arise as a result of such a contrast function as the theme of
reflection in the last section.
As a part of an unconcluded project I am working on, this article merely
expresses some working hypotheses that appear rather productive. As a
project, it has the ambition of wanting to formulate the intuitions of one
culture within the language of an other. As an article, its aim is not to convince
but to persuade. Could it be, I want to ask, that we may fruitfully look at the
issues in a way we are not accustomed to? What would it be like; I want you to
ask yourselves by the end of this article, if cultures could learn to really look at
each other?

1. Raising a problem: knowledge and social life

To any one who has some first hand knowledge of India, the conjunction of the
following three phenomena must appear both extraordinary and striking:
Firstly, its social organization (viz., the so-called caste system) exhibits
enormous complexity, manifests some kind of an order and touches every walk
of life. It is a social organization under whose scope falls not only the mundane
and minutiae, but also the deeper and the lofty. And yet, no Indian could tell
you much about the principles of this system, leave alone the dynamics of its
reproduction. Theories about this social organization, one that has had a
history of more than two thousand years, are not to be found within the Indian
tradition. Of course, many treatises about the caste system have been
authored by Indian intellectuals during this century. My point is that theorizing
about the caste system is neither indigenous to the tradition nor inherited
from it, but a learned and acquired mode of treating the subject. Secondly,
each of the caste groups in India appears to have an enormous repertoire of
dos and donts. Again, were you to dip into the literature of the Indian
tradition expecting to find complex reasonings and justifications to support any
one set from this variegated, hardly overlapping sets of injunctions, you are
likely to draw a blank. That is, individuals perform a great variety of moral
actions and there appears to exist many action alternatives without a
correlated corpus of justificatory literature. Thirdly, neither the social
institutions nor the moral practices are founded upon centralized authorities:
be they religious, moral or political. In other words, absence of knowledge
about practices is not supplanted by the presence of authorities whose
cogitations could make debates and arguments superfluous.

What, you may wonder, is problematic about the coexistence of these three
phenomena in any one society? The answer would be evident, if we were to
indulge in a comparison. Contrast the situation sketched above with that of the
West regarding some similar domain, say, that of the ethical. Consider, for
example, the total number of moral principles (or injunctions) in the West
with the amount of literature produced about them. These injunctions, each of
which is formulated and defended as a general principle, do not (probably)
number more than twenty. But the literature about them stretches to infinity.
With respect to India, it would be difficult to make any such claim; it is not
clear whether the 'injunctions' are general principles or not; one does not
know how many injunctions exist and, as I said, there is hardly any justificatory
literature.
What are we to make of this? Here are two possible responses from
among many:
(a) Because Indians have no comparable moral theories (comparable, that is, to
those of the West) they really have no notion of good and bad. Therefore,
they are immoral. Not many would say it in quite this fashion, even though
many still think so. This does not merit an answer.
(b) They are arrested at a primitive stage of moral development, even if they
have stumbled upon some fundamental moral principles. This is a selfdefeating response: if a culture has stumbled upon such principles and every
individual from each generation over thousands of years continues to stumble
upon it, moral practice does not require to be supported by ethical theories. By
the same token, the existence of theoretical treatises about moral practices in
a culture is hardly indicative of its maturity.
It is only by leaving such or similar responses aside that we are better able
to formulate the issue that a culture like India raises. Let us do so by looking at
some facts. The Indian caste appears to have survived (in whatever form, and
through whatever mechanisms) many fundamental upheavals: through
challenges posed by internal movements (like Buddhism etc.) to externally
imposed economico-political reorganizations (Islamic invasion, British
colonialism and subsequent integration into a world capitalist system). That is
to say, caste system appears to have adapted itself to changing environments
over the course of the centuries. This is a statement that one could make
without having to prejudge the desirability or otherwise of such an
organization.
Clearly, the minimum that is required for such an adaptive social
organization is that some kind of knowledge present in society and that it be
available to its members. Knowledge must be present, because the actions that
reproduce a way of living are knowledgeable actions and cannot be either
random or purely explanatory ones. Furthermore, because these actions

enable a social organization to adapt itself, they will have to be informed

actions. And because what is adaptive over thousands of generations is one
and the same organization, which is a creation of infinitely many actions of
indefinitely many individuals, each member of such a form of life must have
access to this knowledge. Otherwise, there would be no continuity between
generations and hence no culture either. Given that Indian society has some
such thing called culture and some kind of a history, the question must now be
fairly obvious: If knowledge about these practices is not what there is, what
other kind of knowledge is it? How is such knowledge transmitted through the
generations? How do individuals learn it? How does this enable them to
sustain a very complex form of social interaction?
These questions have rarely been raised in the literature, be they
philosophical, sociological or anthropological. Most of the studies about the
Indian caste system, for example, have assumed the presence of some or
other principles underlying and supporting the system. These studies have
also excavated and brought to light many such principles: from non-egalitarian
principles through hygienic principles (albeit in a metaphysical form); from
transactional rules through rules of coalition formation to the very propensity
of human beings to maximize fitness through extended nepotism, etc. This is
not the place to discuss the veracity of any of these studies. And yet, I mention
these attempts because it is relevant for the concerns of this paper to ask why
this approach has been the dominant one at all. That is, why is it assumed that
knowledge about human practices is the best way to understand human ways
of going about the world? In this paper, I begin to answer this question as well,
by trying to link notions of knowledge and ways of knowledge acquisition to
the nature of cultures. Thus, a philosophical quest, viz., that of clarifying some
conception of knowledge begins and ends as a project in comparative
anthropology.

2. Culture, learning and paradigms of order

Is it plausible to accept the idea that the way the members of a group learn is
intimately connected to the culture of that group? In this part, I shall provide a
brief positive answer by looking at one element that plays an important role in
linking the mode of learning to the culture of a group.
As human beings, we are socialized within the framework of groups. In its
broadest sense, socialization refers to the process of living with others. Who
these others are, and what it means to live with them are things that a
human organism learns when it gets socialized. That is, socializing process
involves transmitting this knowledge, which is practical in nature. The
customs, lores and traditions of human groups not only preserve this
knowledge but are also the mechanisms of its transmission. What a learning
process is, when viewed from the point of view of he organism that is being

socialized, is a teaching process from the point of view of those responsible for
socializing that organism. The teachers, thus, draw upon the resources of the
culture to which they belong.
Consequently, the methods of teaching an organism will teach only to the
extent they dovetail with the process of learning. Because we, as human
beings, are not genetically determined to learn in any one particular way, it
could be safely held that the teaching processes give form to the way an
organism learns about environment. That is, the way one learns is non-trivially
dependent upon one's culture. We may, therefore, accept the idea that not only
the what but also how of learning is connected to the culture of an organisms
group. All of the above claims appear extremely reasonable. As we shall soon
see, it also appears possible to generate a culture-specific notion of learning,
which is able to shed light on a diverse set of phenomena that are
characteristic to a culture.
Several elements are of direct relevance to the way a culture teaches, and
to the way its members learn: the experience of self and others; the
experience of body and space; etc. Unable as I am to discuss any of these, I
shall restrict myself to just one element from the set, viz., the experience of
order.

If construed rather broadly, learning can be seen as the way an organism

makes its environment habitable i.e., it is an activity of making habitat. The
fashion in which this takes place, however, depends on the experience the
organism has of the environment that has to be made habitable. One of the
fundamental differences between cultures is the manner of structuring this
experience and that, I suggest, take place through the mediation of some or
other paradigm or root model of order. Let me explain.
Almost all cultures inculcate and preserve a sense and feeling of order
within their members. It is almost as if each generation teaches the following
truth to the next generation: cultural systems are not the intended results of
the actions of its members. Both the order in ones culture and the order that
the cosmos is share the property of not being and planned products of the
actions of its members. The awareness that actions of its members are
necessary to maintain the cultural order and that actions, somehow, can either
sustain or disrupt these orders is also present in most cultures.
However, what distinguishes one culture from another is, among other
things, the way this sense and feeling of order is preserved. Quite obviously,
some way of preserving a sense of order can continue to preserve only to the
extent it ceaselessly structures the very experience of order itself. That is to
say, preserving the sense of order and structuring the experience of order

are both descriptions of the same process but from two different points of
view; each generation preserves its experience of order by structuring the
experience of the next. A way of preserving a sense of order can structure an
experience of order, it appears to me, if it itself models or represents or
explicates that very order whose experience it is supposed to structure. In
other words, that which structures the experience of Cosmos as an order must
itself, in some way or another, exhibit or express the order of the Cosmos. I call
such a prototype of order, the primary root model of order. What distinguishes
cultures from one another, under this construal, would be the nature of their
paradigms of order.
My suggestion is that religion has been the paradigm of order for the
western culture. According to the account given above, it would mean that
religion both structures the experience of the universe and is itself an
instance of that order. Now, whatever religion might be, it is without doubt
true (If, that is, Judaism, Christianity and Islam are what religions are) that it is
explanatory. That means to say, the structure and the strategy of religious
explanations both structures and sustains an experience of order that that
universe is. Because the finer and subtler points about the nature of religious
explanations are not relevant for this paper, let me crudely outline the three
basic phases which culminate in the sense of order that I consider typical for, if
not unique to, the West.
There is the first phase, I suppose, of a pre-religious experience of the
world. In this phase, there is an experiencing of the constancy and regularity
of the universe. The second phase denies this experience of order and draws
the attention to the chaos the world actually is. In the third phase, there is a
discovery of a deeper and underlying order beneath the chaos of the
phenomenal world. What happens in this process is that the first experience,
the naive experience of order, is totally bracketed away, leaving behind only
the chaos that the phenomena experience is supposed to be until the deeper
order is discovered. The obviousness and the all pervasiveness of this attitude
is expressed in the way child development was conceived until recently, and in
the way the emergence of mythical and magical thinking is accounted for. A
child's world, it was surmised, is a bloomin, buzzin confusion until it learns
to conceptualize, categorize and speak. And we are told that our ancestors
attempted to find explanations for natural phenomena because they were
confronted by a chaotic world i.e. by the occurrence of random events. Even
the arguments for the existence of God took the form of arguments from
design in the hands of the medieval naturalists and theologians. The
fundamental attitude, in other words, is that if there are no laws or
principles underlying an experiential world, such a world will have to be a
chaotic and unordered world.
Consider now another culture, where such or similar paradigms of order do
not exist. To its members the above suggestion would appear strange: neither

chaos nor deeper laws are experiential units to this culture. The universe is
experienced as an order: neither a creator nor His design requires to be
invoked as the reason for this order. Explanations may try to say what the
order consists of; but the truth or otherwise of these explanations arc
irrelevant to the fact that the universe is an order. That is to say, the world
does not embody or express an order (or anything else): to speak of the world,
to experience it, is to experience order. For this to be true, it would require
that such a culture have a root or primary model of order which is not
explanatory in nature. Nevertheless, it must be ordered and that order must be
visible on the surface. The truth or falsity of the many different explanations
(interpretations would, perhaps, be a better word in this context) must be
irrelevant to answering the question whether such a root model is an order.
That is to say, in such a culture, the root or primary model of order must be
pre-linguistic (or non-linguistic) in nature. A set of actions, with a beginning
and an end (neither of which need be an absolute), following each other
contiguously in time would constitute such an order. India is one each culture,
I want to claim, and its paradigm of order is not religion but something else.
Call it, for the sake of convenience, Ritual.
Under this construal, Religion and Ritual are two fundamental paradigms of
order or two fundamental ways of structuring the experience of the universe.
This does not require that the other mode (ritual for one, religion for the other)
is absent in either of the two cultures: they exist, but as subordinate moments
and hence as almost unrecognizable counterparts. A religious ritual is as
different from Ritual as ritualistic religions are from Religion. We shall have an
opportunity to relive this problem again, but in another form later.
Given this, the problem that occupies the intellectual energies of one
culture is: given that the universe is an order, how to perform actions that
better fit the order? In an another, it takes a different form: What is the nature
of the order such that one may perform the action that require to be
performed? How, asks one culture, to perform actions and continue to improve
them such that the order does not disintegrate? In an another, the order is not
visible: it is a design; a law that lies hidden beneath. Knowledge of this
order would allow one to decide about the right action that requires to be
performed in indefinitely many contexts. These two paradigms of order, that
distinguish these two cultures, shed some light on the way the problem of
action is treated: to know one has to act in one culture; to act one has to know
in another.
My suggestion is that these paradigms of order, in their turn, enable the
emergence of culturally dominant modes of learning. In a culture like India,
this leads to a learning process whose main focus it is to develop the ability of
performing and improving actions. The knowledge that the members of such a
culture have is what I shall call action-knowledge. In the next two sections, I
shall go a bit deeper both into the kind of knowledge that action-knowledge is

and into the nature of learning process required to acquire such a knowledge.
Contrast and comparison with the West will be attempted subsequently.

3. Action-Knowledge and exemplary learning

The proposal that I shall outline in the rest of this paper is the following:
Action-Knowledge (or Practical Knowledge) is a species of knowledge that is
distinct from theoretical knowledge. The process of acquiring this knowledge
involves mimetic learning i.e., learning through exemplars. Exemplars are
different from examples and, therefore, the process of learning through
exemplars is not the same as learning through examples i.e., It is not some
kind of inductive learning. Action-Knowledge, on this ac- count, is not
knowledge about actions; neither is it the same as acquiring some skill or the
other. In the next section, I shall look at the properties of practical knowledge
and the consequences that follow when proposed as a culture specific mode of
learning.

3.1 Stories as models

One of the characteristic properties of Indian society, something which strikes
everyone with more than a vague acquaintance of this culture, appears to be
its incredible variety and stock of stories. Inundated with stories, its people
appear to relish the act of telling stories: there are stories for every situation;
all reasons and explanations require stories and their presence is as
ubiquitous as the very air itself. Clearly, if a child is constantly exposed to
stories at all levels of social interaction, they end up playing an important role
in the very process of socializing itself. So, whatever be the roles that stories
play, it is reasonable to assert that they are vita] ingredients in the process of
learning to live with the others. That is, in. such B culture, stories would have
to function as units of a learning process.
By virtue of which property can stories play the role that I suggest they do?
I am rather hesitant to ascribe any internal or intrinsic properties to stories
that enable them to be units of learning. I do not think that it makes a great
deal of sense to speak of unite of learning outside a specific learning process.
However, given a specific kind of learning process stories could be its units
because they do have some properties. For the moment, two such merit our
attention.
Firstly, stories are a way of representing the world. Cognitively speaking,
they are models of the world in a broad sense of the term. As models, they
portray, stand for or represent some small part of the world. Let us briefly see
how they do it.

Take, for instance, a group performing some ritual or the other, say, a rain
ritual. When asked about the significance of their actions, one gets to hear a
story. Such a story depicts a set of events which includes the performing of the
rain ritual in conjunction with some other events. Now it is not the case that
causal efficacy is attributed to the performance of such a ritual. That is, the
members of the group do not believe that their singing and chanting in some
specified fashion and the pouring of ghee into the fire altar cause the rains to
come. They are not justifying this belief by telling a story. What then are they
doing?
Because stories are models of a situation, as models they are neither true
nor false; it is only in models that statements come out as true or false. When
the group performs such a ritual and no rains come, all that can be said about
the story is that it is not a model for such a situation. If, on the other hand, we
look at the way the members of the group experience the situation, then quite
a lot could be explicated. When the ritual is performed and the rains do not
come, the group experiences this situation as something having gone wrong
somewhere. When the rains do come, it is experienced as everything is as it
should be. What are these experiences signalling?

Recollect, if you will, the suggestion I made earlier about experiencing the
world as an order. The truth or falsity of the explanations about the nature of
the order, I said, is irrelevant to the experience that the world is an order.
There is a question embedded in that suggestion that can be answered now: In
any such culture (including the Indian culture that I am talking about), at any
moment of time, hypotheses float around which purport to explain the order of
the universe: some or other-account of the pattern that ones culture and the
cosmos exhibit, and some explanation of the role of individual actions with
respect to sustaining or disrupting that order. Many such explanations have
come and gone: Why does the sense of order not follow suit?
This is best answered, if we ask how such cultures manage to sustain this
feeling in the absence of knowledge about the pattern. What mechanisms
preserve the sense of order without requiring the presence of knowledge about
the order, which the culture and cosmos exhibit? One such mechanism
admirably suited for the job is the stories and legends that a culture possesses.
Stories preserve patterns without saying what these patterns are. They
depict partial aspects of an order without specifying what the order consists of.
Performing the ritual, the coming of rains, etc., is a sequence of events
described in a story without specifying a relation between them. The
experience of something having gone wrong somewhere and that of
everything is as it should be are expressions of disturbance/appropriateness
accordingly as whether the story is not/is a model of the situation. Stories do

not explain anything because they do not model relations (causal or otherwise)
between events. In very simple terms, they just model a set of affairs.
What I have so far said about the stories allows roe to propose the following
idea: the representational aspect of stories is what makes them continuous
with other representational products known to us like philosophy, scientific
theories, etc. But, of course, there are also differences between them: whereas
theories claim to explain, stories make no such claim. Theories may justify
some belief that you have, stories do not. Nevertheless, stories are pedagogical
instruments in so far as they have the representational (or cognitive)
property. Therefore, I will now make a mild claim, which I hope to strengthen
later on in this article, that stories embody some kind of knowledge. Or,
slightly differently, they are units of learning.
What kind of knowledge is it that these stories embody? And in which kind
of a learning process are they its units? In order to begin answering this
question, I need to look at the second property of the stories as well.

3.2 Stories as exemplars

Apart from exhibiting a cognitive, property, stories possess a practical one too.
By describing a way of going about the world, they are a way of going about
the world. They are models in a practical sense i.e. they can be emulated.
Stories are pedagogical instruments par excellence because of this additional
property. But how can stories leach us to do anything? How can they be
instructive, i.e. instruct us to do anything at all? How can a description of a
way of going about in the world be itself a way of going about in the world?

As stories, they do not come with any explicit morals attached: they do not,
for example, say that the moral of this story is... They are not structured as
manuals for practical action either: do X in order to achieve Y. If they can
teach, then it is because of the way learning occurs in India. Consequently, the
question becomes: What kind of a learning activity is required, if stories are
how one learns? My answer is that it is mimetic learning. As stories, they are
a set of propositions. What they depict are actions. Between these actions and
those of one's own, what obtains is a practical relation of mimesis. Only as
such can stories function as instructions for actions.
Stories combine this double function: they are theoretical and practical at
the same time. They are not straight forward instructions; nor are they only
representational. They entertain too: but not the way the The Little Red

Riding Hood does. Understanding and imitation fall together: to understand is

to imitate and to imitate is to understand. Stories are oblique instructions
disguised as representations depicting actions. One learns while one is not
aware that one is learning. Mimesis is a sub-intentional learning.
In the previous section, I suggested that methods of teaching will teach only
to the extent they dovetail with the processes of learning. Though not a
controversial suggestion by itself, it must be clear what it implies: Stories can
be used to teach if, and only if, the process of learning is such that its units are
exemplars. Consequently, in a culture where mimetic learning is not a
dominant mode, there the stories do not play the same role. They may
entertain, take your fancy, capture your imagination; but instruct, they cannot.
Stories, in such a culture, become a genre of literature and, mostly, remain at
that level. Narration would be different from instruction.
If, on the other hand, narration is to become coextensive with instruction,
either the narrative or the narrator must carry instructional authority. Neither
the stories nor the story tellers carry this label on their sleeves. It is not even
the case that stories about Gods or respected figures (say a Buddha or a
Shankara) are sufficient to lend instructional authority to the narrative. That is
to say, stories are not emulated because they portray action executed by either
divine or human authorities. (This is both a factual statement about India and a
consequence that flows from the suggestion that mimesis is a sub-intentional
learning.) What, then, lends Instructional authority to the stories? This is the
same as asking what makes stories into units of a learning process. I have
partially answered this question already.
But, there is a more intriguing point lurking in the background:
instructional authorities need not be coextensive with religious, moral,
political or divine authorities. That is to say, learning through mimesis, looked
at from the point of view of the learning subject, involves the activity of
constructing the authority of the other. This must be an active process
because the other does not carry identifying marks on its forehead. This has
several interesting consequences as a result, not all of which are of equal
importance to this paper. Let me, therefore, restrict myself to just identifying
two.
If the process of going about the world involves the activity of constructing
instructional authorities; and if it is the case that there is no necessity, say, that
religious or political authorities, by virtue of their position, are also
instructional authorities, the consequence is obvious: their ability to give form
to your learning activities is very much reduced. Or, put in broad and historical
terms, where mimesis obtains, there neither religion nor politics would have
a dominating and decisive impact on ways of going about the world. The
converse would also hold true: the foundation for forms of interaction
between members of such a society would have to be fairly independent of

both the religious and political authorities. Sociologists have often puzzled
over the fact that in India, religion, commerce, politics and education did
not fuse, for any length of time, into one centralized authority. I suggest to you
that the way learning process occurs in India might well begin to shed light on
this issue.
The second point that I want to make regards the public or social process of
mimesis. Both the process of learning and the knowledge that is acquired
presuppose social interaction, instead of arguing for it in the abstract, let me
illustrate by referring to, say, the moral domain once again. To be moral, in the
West, is to follow some or other moral principle. The relation is between an
individual, isolated subject and some injunction or the other. How the
community in which such an individual finds himself is? Is a question of no
moral significance to his moral behavior? By contrast, in a culture dominated
by mimesis (like India) the relation is between individuals (be they the really
existing community or the fictitious individuals portrayed in the stories). A
moral individual, in other words, presupposes a moral community.
Reformulated in different terms, the general point is this: Because mimetic
learning has a public dimension, epistemic problems will have to refer to the
community of learning subjects.

4. Learning through exemplars: properties and consequences

Mimetic learning, to briefly summarize, is learning through exemplars.
Exemplars, as unite of such a learning process, have a representational
property and can be emulated. My claim is that the dominant mode of learning
in India is mimetic or exemplary learning.
There is a kind of circularity in what is said, if you take it as a definition:
mimesis is exemplary learning and exemplars are units of mimetic learning. To
avoid this circularity entirely, more requires to be said both about exemplars
and mimesis than I could possibly say now. However, it is possible to minimize
the circularity in two, ways: Firstly, take what I have said so far not as a
definition, but as an attempt at explication; secondly, by speaking a bit more
about both exemplars and mimesis, I shall suggest that it is possible to
understand them in relative independence. Quite apart from that, something
else requires to be done: the claim that this way of learning is a dominant
mode of learning in India has not yet shown to be the case. I shall try to do that
as well. However, the intention is not so much to convince you of the truth of
that claim as to make it appear plausible. That will be done by drawing out
some consequences that follow from such a claim.

4.1. Properties of action-knowledge

Let me begin by exploring the properties of action-knowledge. Because I want

to talk about this at a rather general level, let me take an example some or the
other variant of which will be familiar to you.
Suppose that you are a social worker, working, say, amongst immigrant
children from the neighborhood. You have undertaken many activities to
integrate them better into the neighborhood social life and, for the moment,
you are at a loss about how to proceed further. That is, you do not know what
to do next. While wrestling with the despair arising out of the desire to do
something new and not being able to know what it is that yet should do, you
meet an old friend of yours. In the process of conversation, he casually
mentions about some activity that an acquaintance of his undertook in, say, an
old peoples home. As soon as you hear about it, you suddenly know what your
next action would be with the immigrant children. And, in all probability, your
action is totally different from the activity that took place in the old people's
home. What exactly happened here?

I would like to suggest that the activity performed at the old peoples home
functioned as an exemplar for you: that is, it functioned as a generative action.
This, in the first place, is what an exemplar is: it creates new, original actions.
If you are willing to accept this suggestion, I should now like to elaborate
further on this point.
There are two ways in which you could understand the statement made
above. You could say that the linguistic description of the action gave you a
new idea about the action that you can execute. But, there is also a second way
of construing the statement, which is what I am proposing. I am not saying
that the activity performed at the old peoples home gave you an idea of
another action that you could perform in your situation. No. You must read me
literally: that persons action generated a new action; you are now able to
perform a new action, something that you could not do before. While you may
be able to describe the activity, which you intend to perform, this
description/conception of the action is parasitic upon the ability to execute it.
In exactly the same way someones idea/thought/theory can create/generate
new idea/thought/theory in your head, someone's action, as an exemplar,
creates a new action. You may complain that this is spooky. Fair enough. But, I
put to you, it is no more spooky than the fact that ideas can give birth to new
ideas. Actions give birth to new and original actions in the way ideas create
novel ideas.
You have now, in other words, acquired action-knowledge. That is, when
described from your point of view, your ability to produce a new action now is
what it means to speak of you having acquired action-knowledge. In this case,
the exemplar happened to be a linguistic description of someones action.

Quite obviously, it need not be so in all cases. You may see someone doing
something, and that enables you to execute a new action as well. In both cases,
the learning process is the same: learning through an exemplar or mimetic
learning. As a consequence, it must be clear that mimesis, as a learning
process, is creative and dynamic.
Consider, now, the second property of action-knowledge. In our case, the
exemplar happened to be a very specific action undertaken by, a very specific
individual in a very specific context. And yet, it was able to create in you,
another specific individual in a totally different context, the ability to execute
an entirely different action. All exemplars are always context bound. But, they
are generative in totally different contexts. Or, put even more clearly, it is
precisely the context dependence of an action that makes it fertile in different
contexts.
Our general intuition regarding the nature of guidelines or decision
principles is that the more general, abstract and context independent they are,
the more useful and true they are. In fact, this is a demand that we make upon
all knowledge-claims: a moral principle, a. legal statute is acceptable if and
only if it enables you to choose and execute the right action in all/most
contexts. For that to happen, a moral principle requires to be context free.
However, for action-knowledge it appears to go the other way: Inter-contextual
applicability is directly proportional to contextual embeddedness and inversely
to context insensitivity and generality.
Consider, now, the third property of action-knowledge. Even though you
knew what it is that you wanted to achieve, the knowledge of this goal,
together with the knowledge about your present state, did not help you in
producing a new action. Also, your learning through the exemplar occurred
when you were not aware that you were learning. Even though each of these
two points enables me to argue the point independently, their conjunction
gives it a greater force: your ability to execute a new action is indifferent to
the presence or absence of goals. Or action-knowledge is not goal dependent.
One of the basic beliefs in the Western tradition is that human action is goal
oriented action, and that this constitutes an intrinsic property of human
actions. It I am right, just the opposite is true: intrinsically, human action is
goal-less. This does not, quite obviously, prevent you from finding a goal for an
action when you have action-knowledge. But mimesis, as a sub-intentional
learning that involves the ability to execute actions, does not require the
presence of goals. Practical activity, practical knowledge - as species of
knowledge - is not intentional and it is not goal-directed.
Consider the fourth and final property of action-knowledge. Here, I will be
brief because to argue the case would require bringing in other considerations
extraneous to the paper. The situation you were in was one where you did not

know what to do. It is not as though you had difficulty in choosing between the
alternatives that were open to you. It is not even the case, I would like to
suggest, that you chose between the action that you can now execute and
those action-alternatives that existed previously. Rather, acquiring the ability
of executing a new action was the same as knowing what to do.
In slightly more general terms, what I am driving at is this. Within the
western tradition, the dominant approach is to treat epistemic problems in
decision-theoretic terms or as decision problems. I believe that this is not the
case with respect to practical knowledge: epistemic problems regarding
action-knowledge ore not decision problems, but learning problems. Problems
of social interaction and social organization, under this construal, are problems
of learning.
By saying this, I do not deny that it is possible to describe them as decision
problems. (Why it is possible to do so is a theme I reflect upon in the last
section.). Besides, I am aware, that social sciences are increasingly turning
towards game theory and choice theories in their attempts to understand
social interactions.

4.2. Action-Knowledge and cultural specificity: some hypotheses

Let me now turn my attention to outlining some empirical hypotheses that
seem to follow if we accept the idea that mimesis is a way of learning
characteristic to India. In fact I believe that it is typical of the Asian culture
and not restricted to India alone. Consequently, here and there I talk of Asia
and India as if these two terms are interchangeable, which they are in the
context of speaking about learning processes.
1. If socialization involves mimesis and families are the primary units of
socializing a human infant, the success of the socialization process
depends very much on what the family exactly models. That is, an
individual can be taught to live with others if, and only if, the family
stands for or represents the significant details of the social
environment. The family, in its important details, must be continuous
with the moral community at large. And, I submit, it does.
Not only this. In a peculiar way, this sheds light upon the
sternness or harshness considered typical of both family life and
teaching situations in India. One is being prepared for life when one is
brought up as an offspring and a pupil. The parents and the teachers,
between them, prepare the child to act morally when it goes out as an
adult to meet the world at large. That can only be done if the child,
during its maturational phases, faces a wide variety of situations and
sees the way in which the others are going to construe its actions.
Parents and teachers must, in the full sense of the word, stand for and

represent the rest of the community. Consequently, ones family is also

ones sternest and harshest critic. If one passes this test, the belief is
that one can pass any other test. Hence the descriptions of an ideal
father or teacher: harder than the diamond, softer than a flower.
The family as a moral arena as Indian culture sees it, and family
as a Heaven in a Heartless World (as Lasch titles his book on the
family) cannot be sharper. In Western families one is to experience
love, one learns to be oneself. The socializing or the educative role of
the family is secondary, it is derivative. Its primary task is to protect
the child from the cruel world out there. If it prepares the child to
face up to the cold and indifferent world, it does so by providing that
love and understanding which gives the child the courage to "go
and get" what it wants. Family is ones only oasis in the desert of
social life.
2. Mimesis, though not a blind imitative learning, entails a reproduction
of existing actions i.e. it essentially conserves. A culture dominated by
mimetic learning must, perforce, exhibit a very strong pull towards
conservatism. Indian culture is essentially conservative. Tradition, the
past etc., must weigh heavily on all those who are members of such a
culture. Again, I submit that it does.
3. The other side of the same phenomenon is what happens when such a
culture meets with that of the West. There is a partial exchange of
authorities, not their disappearance. The tendency is towards an
imitation of these new authorities. Whether we look at the shallow
westernization of the youth, the clarion call of the intellectuals to
follow the West, or even at the fact that the Japanese have earned the
label, often used pejoratively, of being very good imitators - the
phenomenon is the same. We imitate the West not because there is
some iron law of capitalism that compels us, willy-nilly, to be like
them, but because it is our way of learning.
4. Learning through mimesis requires that one develops the ability to
discriminate finely. One has to sort out, so to speak, situations and
actions in such a way that one distinguishes between to emulate in
this situation from to emulate in that situation. Not all aspects of an
event or action should be emulated. In other words, one grows to be a
member of such a culture by acquiring a finely tuned set of
discriminating criteria.
How is this acquired? Again, the answer cannot be other than to say
by mimesis. However, events and actions must loose their clarity and
simplicity when multiple and often incompatible models are said to
model the same situation. They must become complex and essentially
ambiguous. Indeed, I claim, they do. One expression of this situation

is the extraordinary productivity of Indian culture with respect to

religions.
5. In the previous section, I pointed out that instructional authority is
not coextensive with religious, moral or political authorities. Here, I
will simply state some things explicitly, which were said there
implicitly. Nothing about the learning process prevents that they be
the same; the point is that it is not required.
The first thing to notice is that if there exists a learning process
one of whose moments
requires constructing an authority, this
orientation is bound to spread into or spill over into other spheres. That
is, the very idea of what it means to be an authority will begin to get
affected. Consequently, what it means to speak of political
authorities, rule of law or even religious authorities in a culture like
India is anything but clear. To appreciate the force of
this
difference, takes a look at the history of the West: each challenge to the
authority (be it clerical, biblical, political or juridical) has precipitated
a deep crisis in the culture. Challenges of similar nature, of which there
have been many, have hardly rippled the fabric of social life
in India.
There is a second point, about which I can afford to be briefer.
Mimesis is a way of learning to live with others and go about the
world. The nature of this learning process severs any intrinsic
relationship between instructional authorities and political or
religious authorities. If the Indian caste system is the result of such
a learning process, it follows that it rests on neither religion nor
politics. Not just that. It must be an organization that resists any
centralization of 'political' and 'religious' power where the latter, by
virtue of such centralization, could begin, to function as an
instructional authority. (I say it must be, because it could not have
survived for so long if it had tied itself to any one religious or
political authority.) What its implications are, I will leave it to you to
reflect about.
6. Consider another implication of my suggestion. I am informed by
people active in the area that some philosophers of mathematics are
beginning to turn their attention to the processes of constructing a
proof in order to say what it is for something to be a proof. They
believe that the notion of proof is somehow related to, or requires
ineluctable reference to, the activity of constructing a proof. Consider
now another fact: Immigrant Asians in Europe and America turn out
to be better in such disciplines as Mathematics and Engineering than
any other ethnic minority. This difference between Asians on the one
hand, and other groups (including the native white population) is
statistically significant enough to engage the attention of

psychologists, pedagogists, etc., and to initiate research into this

question in a concentrated way. Suppose that there is some truth to
the proposal made by some philosophers of mathematics. In such a
case, the fact I drew your attention to appears to follow as a
consequence of the notion of learning through mimesis: A culture
whose dominant mode of learning develops action-knowledge in its
members, predisposes the latter to become proficient in domains
involving action-knowledge. When members of such a culture take to
mathematics or engineering, the number of those who become
mathematicians or engineers will have to be significantly higher when
compared to members from another culture. Prima facie, it appears to
me, such is the case.
7. Consider, finally, another kind of issue that emerges if we relate what
is said about actionknowledge with the suggestion I made about
order and learning.
If the 'action' of action-knowledge is neither intentional nor goal
directed, the problem of action could hardly be one of relation
between intending: and acting However, if the perception of order is
such that it is not possible for it to be there without underlying laws
or principles, then it is obvious that action (typically human action,
that is) could not be without a conception guiding it. Action becomes
nothing but applied conception; practice nothing but applied theory,
and technology nothing but applied science. However, this view is one
cultures way of looking at the issue. It is in this sense that I said in
the introduction that action theories, when they talk about action, are
not so much talking about the nature of human action as much about
one culture-specific -notion.
In what I have said above, there is a problem that requires to be
made explicit. If action (the typically human action) does not rest
upon conceptions; if it is not an execution of an idea, what could be
said about the results of such an action? By the same token, and
extending it further, it could be asked, Could an ordered phenomenal
world exist without being law-governed? The answer will have to
be yes, but requiring of some qualification.
Were we to think of the Natural world, the positive answer
provided above appears quite incomprehensible. The natural world is
ordered precisely because it is law governed. What would a negation
of this statement amount to? Frankly speaking, I doubt whether one
could put it in words at all. I do believe that the natural world is lawgoverned and an ordered phenomenal world without laws is simply
incomprehensible. We could not survive in such a world.

The matter, however, takes on a different light when the world in

question happens to be the social world. A social world is the creation
of human actions, the knowledge of creating it is action-knowledge
and this action is not an. execution, of conceptions. Incredibly
complex forms of social organization can exist, continue to adapt and
expand themselves without being governed by any laws. That is,
there is no prima facie reason why it is not possible. Not just that. If
you are willing to assent (however tentatively) to what I have said so
far, it might appear that it is the nature of social organizations that
they are based upon no laws or principles. A social organization is
accumulated practical knowledge. To seek to understand a social
organization by looking for its laws (or the principles upon which it
is based) might be as absurd as the denial of the law-governed nature
of the Natural world. It must be obvious where I am driving at: Indian
caste system is based on no principle. This is obvious to an Indian, but
quite incomprehensible to indologists. The way of creating a social
world is different from the way of understanding a natural world.
These are but a few of the consequences that follow from the notion of
mimetic learning. As I said in the beginning, these consequences do not prove
my claim. But, they ought at least make my suggestion appear plausible.
Let me summarize what I have said so far: Action-knowledge is not a
knowledge about actions, but the ability to execute new actions. Actionknowledge is acquired through mimesis, which is a process of learning through
exemplars. Actions generate actions - not ideas about actions - in exactly the
same way ideas generate ideas. Action and action-knowledge require total
contextual embedded ness and are intrinsically goal-less.
A culture like India must now begin to appear in a different light: inundated
with exemplars, it must be dominated by mimetic learning. Spheres such as
morality, law, social organization, human interaction etc., belong to that of
practical knowledge. Practical knowledge is cumulative perhaps to a greater
degree than knowledge in the theoretical sphere. And the form of social
organization, the so-called caste system, is one such cumulative result. And
that is why; no Indian could tell you what its principles are. Yet, it
reproduces itself because there is knowledge available - action-knowledge - to
reproduce it. Its ability to adapt itself to changing environments is merely the
ability of human beings to execute actions in different environments. Several
intriguing results flow-from this (re)description of the Indian caste system,
but exploring them is beyond the province of this paper.

5. Knowing how, knowing that and action-knowledge: a contrast

How much of what I have said so far is unique to Indian (Asian) Culture? Is
there no mimetic learning in the West? Or is mimesis merely a variant of a

learning process that is really not culture specific? The answers to these
questions are complex. In what follows, I merely try to take the first step in
exploring them.
At first sight, it would appear that mimesis is omnipresent in the West as
well. Children learn through imitation; an adult learns to eat with chopsticks or
learns to dance through imitative actions; an academic imitates his more
successful colleagues by trying to publish as many articles as he possibly can
etc. The list, it seems, is quite huge.
Not quite. Instead of arguing for the details, let me show what I take to be
the case in two ways. You would be willing to accept, I suppose, the
.suggestion that the moral domain is a domain of human actions. Let us,
therefore, look briefly at the presence or absence of exemplars in this domain
for the West. After having done this, let us see what the relationship is between
action-knowledge on the one hand, and the forms of knowledge theorized
about in the West on the other.

5.1. Action-Knowledge and moral action

If stories about individuals have to function as exemplars, as I have said
before, either the narrative or the narrator must embody instructional
authority. At first sight, it would appear that such moral authorities do exist
within the Western culture: some saints; perhaps the figure of Jesus Christ
himself; priests; or figures like Martin Luther King or even someone like
Gandhi. These people call them moral ideals, seem to play the same role as
those played by the exemplars: one is inspired by their lives, one takes them as
an ideal to emulate, one is exhorted to follow in their foot steps etc.
On the other hand, it is also the case in the West that moral learning i.e.,
acquisition of practical knowledge is conceived to consist of two steps:
(a) making some set of moral principles one's own; followed by
(b) an attempt to apply them as, consistently as possible. The first phase
alone involves what
could reasonably called a learning process: in the
process of maturation from childhood to adulthood,
one
learns
these
principles. Once this phase is traversed, the rest of one's life consists of
successes and failures in the application of these principles.
The link between these two, between the existence of instructional
authorities and the notion of moral knowledge as a knowledge of principles, is
quite clear: these moral ideals embody some or other set of principles.
Consequently, if none of the above figures embody the moral principles that
some individual happens to subscribe, these ideals cease having any
instructional authority. Further, it is not even necessary that there be

embodiments of moral principles at all: acceptance of moral principles is not

parasitic upon the existence of ideals.
Nevertheless, let us look at those to whom such or similar figures are worth
emulating. (Because, it is only with respect to these kind of people that we
could speak of mimesis in the field of the ethical.) What could they emulate?
Either they could follow what these moral ideals have said, or emulate what
they did. To do the former is the same as accepting some principle as your
own: the only difference, in this case, is that either its foundation or its
legitimacy derives from the person in question. Let us look at the latter
possibility. What does emulation mean here? It means a simple,
mechanical and blind imitation of these ideals. Each of these ideal figures,
as individuals, were born into and confronted situations and events that are
totally different from that of those who want to emulate them. Not only that. As
embodiments of principles, they are indifferent to contexts. Consequently,
these ideals are exemplars only in this sense: either one reproduces their
lives and their actions, which is impossible (not quite, think of the early
Christian martyrs), or one accepts the impossibility of such a reproduction.
These two possibilities are preserved in the moral talk of the West thus: One
ought to be moral, but one never is, or that the ought is different from
is; and that these individuals are (somehow) exceptional figures. That means
to say, moral ideals have instructional authority only in the sense that they
embody or represent the longing and desire to be moral. More, they are not.
As exceptional individuals, they are other than and different from the ordinary
mortals who strive to be like them (i.e. be moral persons themselves). But,
they represent a goal that is worthy, but quite unreachable. If they are relevant
to daily practices, it is only in a negative way: daily practices ought to be
other than what they, in fact, are. What one does not learn from such ideals is
how to reach that goal i.e. these ideals do not help improve daily practices.
If we look at the stories in India, on the other hand, they are tales about
real or fictitious figures performing actions- in-contexts. Stories of kings, long
lost life styles and of actions ages ago are used as though they have something
to say to us in today's world. That they do say something makes it obvious that
whatever might be required, it could not be blind emulation. These stories
enable you to execute the action in your situation without there being any kind
of similarity between these two contexts.
The point of this contrast is the following: morally exemplary figures (in the
West) are not exemplars in the sense in which I am using the term. An
exemplar is a unit of learning and moral ideals of the West are never that.
There is a disturbing consequence to this thought: in the field of morality, the
West does not have the process of learning that I call mimesis, i.e., because
there are no exemplars, there is no possibility of learning through exemplars.
If moral knowledge is practical knowledge (not knowledge about principles

that allegedly guide actions), and practical knowledge requires exemplary

learning, the conclusion is as obvious as it will be unpalatable: the
voluminous literature about moral activity in the West hides an abysmal
poverty in moral life. Those millions of treatises, tracts and articles that exist
are not expressions of the sophistication or the advance made by the West in
its quest for the moral order, but expressions of unease and absence: absence
of morality and the concomitant unease.
Is this consequence true? I do not think that the issue is one of veracity or
otherwise of the claim. I, for one, do not personally like descriptions of other
cultures that make them out to be immoral or non-moral. There is a more
substantial problem at hand: Why does the moral domain of one culture when
described from within the culture of another appear immoral or non-moral?
Unfortunately, I cannot take up this problem here.
Be as it may, it is time to look at the issue of the presence of mimesis in the
West from a more general point of view.

5.2. Skill and action-knowledge

Consider the following example, original to Ryle, about a skilled mountaineer
stuck on a mountain. At a loss to negotiate the obstacle that is in front of him,
he observes a monkey performing some sequence of actions. That inspires him
to explore new strategies; new ways of circumventing the problem, the result
of this strategizing is a successful climb.
A Rylean would see this improvisation as a part of being a skilled
mountaineer. Consequently, the suggestion about learning through exemplars
appears assimilable within the Rylean knowing how and knowing that. That
is to say, what I have called as a process of learning through exemplars may be
conceptualized as a process of acquiring the skill to identify an exemplar. As a
result, the notion of skill could be reintroduced at a meta-level in such a way
that knowing how to identify an exemplar and knowing how to use it on the
one hand, and knowing that something is an exemplar emerge as the two
primary divisions within the sketch that I have provided. An added temptation
or philosophical incentive for indulging in such an exercise would be that some
of the counterintuitive properties may be dispensed with by pleading
ignorance: because we are not quite clear how human beings acquire the
various skills that they do acquire, until such time as we are clear, there is no
need to postulate a new species of knowledge with queer properties.
Action-knowledge, as an ability to produce new actions, could not possibly
be a skill because any skilled action has a history of execution with respect to
the organism in question. The characteristic property of action-knowledge is
precisely that the action) which can be executed, is novel and original. Even

though this appears a decisive argument, it could be easily met thus: what it
means to speak of a skilled chess player, a skilled surgeon or a skilled driver is
precisely their ability to improvise in new situations and come up with new
actions. After all, skill is not an execution of drilled actions, even if some kind
of drill is required to acquire some skill.
However, it appears to me, that the new action that a skilled practitioner
can execute arises by a combination of actions, which he had already executed.
That is to say, it is like generating a new word out of an existing repertoire of
words. Secondly, even more importantly, the skill that one has in combining
familiar actions to produce new actions is not sufficient, in at least some
circumstances, to improvise. A skilled mountaineer stuck on the mountain
absently watching a monkey move from place to place is able to do something
as a result of taking the actions of the monkey as an exemplar. Prior to this
learning episode, there was an action that he could not generate out of the
repertoire of actions he had. If he could have, he would have not been stuck in
the first place. To be sure, he needed his skills in order to learn what he
learned from the monkey's movement: you and I watching the same monkey's
action would not be able to what the skilled mountaineer did. But, this is no
problem: what we can learn depends to a very great extent upon what we have
already learned. That one and the same kind of learning process can occur
both within the framework of exercising a skilled activity and elsewhere (that
does not involve an exercise of skill) suggest very strongly that learning
through exemplars cannot be seen as a skill to identify and use an exemplar.
There is a third reason, purely linguistic in nature, that suggests that
mimetic learning cannot be identified with knowing how. In the statement
skill to learn through exemplars, the word skill can be replaced by the word
learning without any obvious loss of meaning. This suggests that skill is
coextensive with learning. However, the same substitution does not work
elsewhere as far as I can see: He is a skilled tennis player, He is a skilled
surgeon, He is a skilled problem solver etc., would lose some of their
intended meaning if learning replaced skill everywhere. Consequently, we
could not be using skill with the connotations usually attached to it, when we
say the skill to learn through exemplars. Fourthly and finally, the knowing
how in its normal usage picks out a cluster of actions that are related to each
other. To do so with respect to learning through exemplars, it appears to me,
would not come easily and with the same degree of familiarity.
Even though these are some of the reasons for wanting to distinguish
between a learning process that is indifferent to what is learned and a skill
that is not so, none of them, as I see the situation, are decisive. With suitable
modifications, it is possible to absorb action-knowledge into the relatively more
familiar knowing how.

Why, you may wonder, is it important? The answer would become obvious
when we look into the way action-knowledge could be absorbed into
propositional knowledge.

5.3. Action-knowledge and knowing that

Is there any reason to limit the process of mimetic learning to the domain of
actions and social interactions? No reason apart from an arbitrary fiat comes
to my mind. Besides, if learning through exemplars is indifferent to what is
learned, there is no good reason to say that it is limited to practical actions
only.
Consider a surgeon who intends to use radiation therapy on his patient in
order to destroy a tumor. Though a concentrated high dose is required, to use
it would be fatal to the patient. While at a loss as to what to do, he hears of an
army general who conquered a fortress by sending in his army in several very
small units and have them converge upon the fortress from different
directions. The result of this hearing enables our surgeon to execute a new
action: he now radiates the tumor away by sending in several mild doses and
have them converge upon the tumor in different directions.
There is a reason why I choose this story: it is one of the examples used to
study the role analogies and metaphors in human problem evolving processes.
I do not want to go into the details of the arguments, but the claim of the
cognitive scientists working in this domain is that analogies and metaphors
play a role in discovering a solution in the way the story of the general inspired
a possible solution to the doctors dilemma. In other words, what I would have
called an exemplar is what they call an analogy.
If exemplars are nothing but analogies or metaphors, both the process of
mimetic learning and the notion of action-knowledge can be dispensed with.
With that also go the so-called counterintuitive consequences. We could safely
suggest that the stories, for example, give you an idea about the action that
you could perform; learning through exemplars' is really the more familiar
process of solving problems using analogies etc.
How can this challenge be met? Firstly, the point of analogies and
metaphors, in so far as they are a part of the learning process, is this: learning
is learning through examples. Examples are illustrations, or instances, of some
general principle or the other. To learn through examples is to discover the
general principles that are instanced. Analogies and metaphors, as a
consequence, turn out to be good examples, i.e., they are a subset of
examples. It appears to me that this exactly is what exemplars are not: they
are not examples because they do not instantiate anything.

Secondly, such a process of learning is what we call inductive learning, i.e.,

a process that induces general principles from a set of particular instances.
The kind of knowledge gleaned is the knowledge of general laws and, as such,
pertains to the domain of theoretical knowledge. However, it appears to me
that one does induce any principle while learning through mimesis. And the
kind of knowledge that one acquires is an ability to perform or execute some
kind of an action. It is because of this that I would like to look at actionknowledge as a distinct species of knowledge.
Neither of the two arguments are decisive, much less convincing. It appears
to me that it is possible to assimilate the notion of learning through exemplars
into some or other variant of analogical problem solving.
However, if such an assimilation of mimesis into either skills or
knowledge were to occur, the following appears to happen: you could not
shed light on, say, caste system (or any other fact) of a culture by speaking of
either skills or analogies. In other words, the explanatory potential of the
heuristic would be lost if such assimilations were to occur.
The problem must be obvious by now: Action-Knowledge and the idea of
mimetic learning, which have some counterintuitive consequences, are
proposed as culture specific modes of learning. When seen in this light, they
appear to be productive because they are able to shed come light on
phenomena which otherwise appear puzzling. However, as a mode of learning
it appears capable of extension to areas other than practical learning. Not just
that: any refusal to do so appears arbitrary. However, when this extension
takes place, it merely appears as a mild variant which can be absorbed by
other notions of learning. These other notions of learning, however, are unable
to shed much light precisely about those phenomena which mimetic learning
was able to illuminate. How should this be addressed?

6. Comparative anthropology as a philosophical quest

In the first section, I suggested that the paradigms of order differ between
cultures: Religion for one, Ritual for the other. There I hinted that each knows
of the other paradigms of order as well: religious rituals and ritualistic
religions. But a religious ritual is and is not ritual; a ritualistic religion is and is
not a Religion. This division has now reproduced itself in the dominant modes
of learning: mimesis is and is not inductive learning; it is and it is not a skill; it
is present and is not present in the West.
I am not going to plead dialectics in order to understand the situation.
There is, I believe, a more substantial issue at stake.

In so far as learning is a process of creating a habitat, several kinds of

activities are involved: the ability to build and sustain social interactions, the
ability to think about Nature and society etc. Each cultural group, it appears to
me, develops one of these kinds of activities into a dominant mode. Other kinds
and ways of learning continue to exist and develop: but they do so within the
dominant framework, as modified by it and in a subordinate mode. This is, of
course, obvious: even if mimesis is the process of developing actionknowledge, a culture dominated by it requires theoretical knowledge.
However, the mode of theoretical learning emerges within the overarching
frame of practical learning. The other way, for another culture, goes as well.
Consequently, the issue of extending one mode of learning to encompass
other modes is actual not because of the propensity of the theoreticians to
raise it: Rather, it is an actual extension that a culture has already
accomplished. That is to say, cultures have developed other modes of learning
apart from the dominant mode of learning but within the latter: subordinated
to it or modified by it. Formulated differently: A culturally dominant mode of
learning extends and adapts itself to other areas of learning. However, as we
have seen, such extensions are not productive: Mimetic learning docs not
appear to tell us much about theoretical learning; theoretical knowledge has
even less to say about practical knowledge. This state of affairs mirrors the
historical development that has actually taken place: A culture like India that
had specialized in developing action-knowledge did not give us the Natural
sciences. By the same token, the culture that did develop the natural sciences
exhibits abysmal poverty in social, practical and moral life.
When we therefore ask questions about the existence of different kinds of
learning in other cultures, we will have to answer in the affirmative: Yes, for
example, mimesis exists in the West as well. But its relation to the dominant
mode of learning in India is akin to some kind of family resemblance', these
two are not the same; they are not identical learning processes. Consequently,
the second answer: if mimesis refers to the way learning occurs in India and to
the domains it includes, then mimesis is absent in the West. Mimesis is and is
not present in the West.
In the first section, I linked modes of learning to the experience of order. In
this paper, I speak of the Indian and Western cultures. There are, however,
other cultures and civilizations: the African, the American-Indian - to mention
just two. How do they experience order? What kinds of learning
processes have they developed as their dominant modes of learning? To
really and seriously begin asking questions about human learning, leave alone
developing a theory about it, we need to have culture-specific answers to
issues that we cannot even formulate properly today. Before we could say what
knowledge is, we need to know how cultures survived, and what they have
learned.

Such studies, undertaken by members of different cultures against the

background of their cultures, have hardly started to emerge. Instead, what we
do have are some universal theories which are neither universal much less
theories - that merely legislate universal scope to one culture-specific
conceptions. In the same way a religion does not become a universal religion
because it calls itself Catholic, a dominant mode of learning does not become
the way of learning simply because it is called scientific. To debate today
whether theoretical knowledge is more basic than action-knowledge or
whether the former is parasitic upon the latter is to indulge in a senseless
struggle. However, given the kind of dominance that the West enjoys and the
tragic manner in which it has acquired this dominance, the questions set by its
intellectuals have become the questions of inquiry: the problem of action is the
relation between intending and acting, between mental states and actions
and so forth. These are not the problems of action or action-knowledge, but
those that arise when you attempt to treat a dominant mode of learning as a
subordinate one.
In this article, I have taken a very hesitant first step in executing a project
that I believe is necessary. As a first step, it is unbalanced and uncoordinated.
However, the tentativeness of the first step is not an indication of the weakness
of the project of learning to walk; instead, it is a prelude to tread firmly. In the
same way, I hope that you will not prejudge the kind of project I am pleading
for by the merits (or their absence) of this paper. As we well know, experience,
here as elsewhere, will bring the requisite skills.