Best - Mick.
It's been a quite a while since I read Geertz's article. I remember at
the time that it seemed to make quite a lot of sense; and re-reading it
now, I think that it also makes some kind of sense, although for
Initially, as new anthropology student, Geertz seemed an attractive
alternative to existing stuffy approaches in social and cultural
anthropology. These other approaches, criticised by Geertz at the start
of his piece, seemed to be unsuitable for anthropological work, for a
number of reasons, including:
- colonialist mindset
- a fascination with remote/isolated/pristine (i.e. 'primitive' by
another name) cultures
- oversimplistic materialist exaplanations for cultural phenomena
- overcomplex systemic models
In addition, many anthropologists by default seemed to focus almost
exclusively on overseas field sites, and as a consequence devalued any
anthropological work done in the west, in particular (or so it seemed
in my case) any work done in technology and science or with
technologists and scientists. Obviously, not all anthropological
approaches could be criticised on all these grounds; but Geertz did
seem to provide a critique that articulated my own concerns with
anthropological theory. Does Geertz however in the end provide a
reasonable alternative to these approaches? I'm not so sure now.
First, I think that Geertz's clinical induction model of anthropology,
which he uses to support his claim that anthropology is a 'real'
science, is questionable. This is because, in my opinion, social
science is not equivalent to physical science, and the same terms have
different meanings in each arena (this is perhaps a kind of Geertzian
analysis of science). for instance, a theory in anthropology is not
really a theory in physical science; an anthropological theory is at
best a hypothesis, and more often than not just a claim; and
anthropological settings are vastly complex and have too many variables
to permit hypothesis testing and falsification. This is not to say that
a clinical inductive (or abductive) model for anthropology is wrong;
just that I'm not sure that it's going to produce the wider
generalizations that Geertz claims.
Second, and somewhat ironically given Geertz's introductory discussion
of how theories can be adopted enthusiastically but badly at their
introduction, the attractiveness of Geertz's critique has given rise to
a lot of bad interpretive anthropology. Especially, I think that there
is sometimes a confusion between *long* description, and *thick*
description. Geertz's method involves not a long description of a
phenomena, but repeatedly revisiting the same terrain, criss-crossing
it, viewing it from different angles, building up reflexive
juxtapositions of what might be going on. From my point of view,
observing something once is not participant observation; and writing it
down is not a thick description (but maybe a diary entry?). However
such actions are sometimes represented as ethnography and thick
description in the literature.
In what ways therefore does Geertz work for me now? I think his
emphasis on interpretation, Wittgenstein, and others, reminds us of how
little we actually *know*, interpretatively and otherwise, of what
other folks may be up to, and thus highlights the problems and effort
involved in bringing ourselves to understandings of what they might be
up to, regardless of whether or not these understandings are absolute
or accurate representations. It's difficult to 'find our feet' with
other people, and this uncertainty applies to my understandings of
designers and users, as much as it does to Trobriand Islanders. What
therefore does it take to really get a handle on what designers and
users might be thinking? What is the nature of their knowledge (is it
transparent or obvious to us)? Are designers and users obvious and
transparent to each other? If not, how may their knowledge and
communication, when working on a common project, be supported? And so
And this supports the contrast with the physical sciences, perhaps.
At least how we experienced them at undergrad levels. Certainly there
is a similarly unknowable aspect to physical science the more you get
into quarks et. al, but still a legacy of objectivity and
measurability that we don't share.
There's a certain confidence-of-humility that I take away from what
you are riffing on here....
A second book, Philosophical Investigations, was compiled from notes
and published after his death in 1951. This is the book that people
often refer to, when they refer to Wittgenstein, and concepts of his,
such as 'language-games.' Geertz credits Philosophical Investigations
with giving him the inspiration for his ideas about culture. The
specific idea of Wittgenstein's that Geertz draws on, is his theory of
the impossibility of a private language. I'm probably going to mangle
this a bit, but as I understand it, what Wittgenstein pointed out was
that while we think that it is possible to have a private language all
to ourselves - one that just plays in our head and which only we can
understand - in fact such a language is always built on, and always
maps to (dictionary/phrase book style) previously available public
meaning. So if I stub my toe and think, "Wuurhg blapooble
nitryqwgjghj##,' and also think, "This is my private language," while
this may appear to be a 'private' language, in that only I can
understand it, invariably I also understand it in terms of "Aaargh my
toe!" or something similar, which is a prior, public meaning. That is,
I can't help but think "Aaargh my toe!" as well as "Wuurhg blapooble
Anyway this lead Geertz to speculate that the locus of culture is not
therefore mentally internal in people's individual heads, but rather in
publicly shared meaning, and it's this which led him to look at the
thick description and interpretation of publicly available meaning as a
route to describing what (a) culture might be about. From Geertz' point
of view, it's these prior public meanings that can help us to identify
local cultural particularities.
So what might this have to do with users and design? One obvious way in
here, is that we can see designers and users as having different
cultures - which leads to the question of how particular groups of
users may make sense of and use a designed thing in context - which
leads to the question of how we (as designers) understand users in
order to inform our designs for them. Geetz quotes something from
Wittgtenstein about going to a different country and learning some of
the language, yet not 'finding our feet with them.' It's a bit like
going to France and learning to order a meal or a train ticket; but
this does not necessarily help us to understand the French. Similarly,
we have be designing (for instance) cool new technologies, but have no
idea of how potential users actually live, or of how they might want to
fit these new technologies into their daily lives. There's debate as to
whether such understanding can be gained from observation, or
observation and interviews, and so on. I do think that to be as useful
as possible, there needs to be some extended immersive
observation/interviewing going on. We need to get at not just what
individual users think, but the publicly shared frames/webs of meaning
that guide and shape their actions.
I should add that I'm no huge fan of Geertz. I suggested that we read
him, as it looked as if we were starting off with anthropology, and in
fact I do like this piece as a conversation starter. Plus, talk to any
anthropologist, and they'll have at least *some* opinion of Geertz.
I also find Wittgenstein very interesting but also obscure and
difficult to understand. Two intros that I like are Ray Monk's bio of
Wittgenstein 'The Duty of Genius' (because I like bios anyway), and A C
Grayling's summary in Wittgenstein: A Very Short Introduction (Oxford
University Press). Pelle Ehn's particpatory design theories include a
big chunk of Wittgenstein; I have a copy of an Ehn paper around
somewhere and may post that as a link later.
> I wasn't familiar so I used Wikipedia
> *the world consists of independent atomic facts ? existing states of affairs ? out of which larger facts are built.
> *Language consists of atomic, and then larger-scale propositions that correspond to these facts by sharing the same "logical form."
> *Thought, expressed in language, "pictures" these facts.
> *We can analyse our thoughts and sentences to express ('express' as in _show_, not _say_) their true logical form.
> *Those we cannot so analyse cannot be meaningfully discussed.
> *Philosophy consists of no more than this form of analysis: "_Wovon man nicht sprechen kann, dar?ber mu? man schweigen"_ ? whereof one cannot speak, thereof one must be silent.
> Some commentators [_citation needed_] believe that, although no other type of discourse is, properly speaking, philosophy, Wittgenstein does imply that those things to be passed over "in silence" may be important or useful, according to some of his more cryptic propositions in the last sections of the _Tractatus_: indeed, may be the most important and most useful.
Mick - this is fantastic! Thanks for the detail and interpretation -
the guy published one book and I referenced the wrong one :)
Well I guess it was 50/50 ;)
Just FYI, I am including a YSI link for a zip archive of some related
- 2 interviews with Geertz (gives you more of an idea where he is
- a copy of Ryle's "What is Le Penseur Doing?"
- a copy of Pelle Ehn's paper "Playing the language-games of design and
use: On skill and participation" which includes a discussion of the
relevance of Wittgenstein to participatory design
- a 'thick description' of a park bench by Molly Schuchat
You may have to double click the downloaded file to unzip it.
This link will expire on June 23.