re: Language and non verbal constructs
Thu, 6 Jun 1996 23:06:04 +0000

Bob Parks writes:

>Perhaps the process of avoidance/suppression of topics (e.g., sex) is
>comparable to the processes in intercultural interaction that Devi
>Jancowicz refers to. When one culture attends to a matter in its "stories"
>that lie behind words, perhaps the other culture neglects that matter in
>its stories.

I'm not sure. Yes, different cultures draw on "different stories that lie
behind words". But I was referring more to structure than to content. There
seem to be constraints in different languages, such that, for example,
those associations (set X) which are possible to a word W in language L are
rather different to the dictionary-translated word Y(X) in language M: for
language M, the associations might be (set X'), (set X + Z) or even (set
not-X). To the extent that associations determine meaning, we would thereby
say that the meaning of the same (dictionary-translated) word can be
extremely different in different languages.
And to the extent that we use words to encode experience, the meaning of
the same experience is thereby going to be different for people using a
different language.

My paper Jankowicz (1994) develops this argument in some detail, and
provides you with a simple technique for experiencing in English, as far as
is possible, the associations that a non-English speaker might have for
certain non-English words, just to exemplify the above argument while
giving you a taste of what bilingual experience is like.

But just to give you a flavour right now...

Take the word "to manage"
In English, the associations of controlling, administering, coping with
limited resources, bringing into consent; governance, and husbandry are
obvious to any English speaker who hasn't a cloth ear (or who looks up the
word in something comprehensive like the Oxford English Dictionary).

A really good English-Polish dictionary (e.g. Bulas et al. 1967) will give
you two Polish words for "manage", _zarzadzac_ and _kierowac_.

To a Polish speaker, the associations of _zarzadzac_ involve connotations
of authoritative commanding and governing. The associations of _kierowac_
involve connotations of steering and controlling.

And therefore it is much, much more difficult for the native Polish speaker
to construe such issues as our UK/US notions of management style
(authoritative yes, but also democratic/participative, and also
laissez-faire as stylistic alternatives) as _necessarily_ involved. To the
UK/US English speaker, the term management is open to a variety of
possibilities as regards style; to the Polish speaker, it isn't.

The US/UK English speaker then commits a catastrophic error when s/he
assumes that the Polish management style is limited to what we'd construe
as the "autocratic" management style.

That construal is possible to a UK/US English speaker precisely because it
represents a choice, in that the English language (so far as that word is
concerned) is _open_ to choice between autocratic, democratic, and

It is not possible to a Polish speaker because his/her language does not
_permit_ a choice between those three possibilities, but rather, offers a
choice between different associations, (steering-governing, if you like,
and a number of others which we'd expereince as rather "commanding" in
their flavour).

Clearly, to the extent that language encodes experience and therefore
individual instances of behaviour, the UK/US observer will certainly be
able to report a greater proportion of activities between Polish managers
and their employees which the UK/US English speaker will label
"authoritarian". But, d'you see, to the Pole that behaviour _isn't_
"authoritarian" because the structure of his/her language doesn't construe
that experience in that way!


Funnily enough, an easier way of expressing all that is in terms of good
ole psychometrics. Give a sample of UK/US managers a (western) management
style inventory, and do the same to a sample of Polish managers. Yes
(because the inventory has been standardised on a western sample), the
Polish mean will come out as more autocratic and non-participative than the
UK/US mean (see, e.g., Maczynski 1991).

But you can only validly provide a score to any individual Polish manager
in terms of his/her deviation from the Polish mean, and not with respect to
the (western) norms with which the test has been supplied.

Well well! What _very_ interesting ground we've suddenly strayed into:
group differences...
to dramatise my point: sex differences... racial differences...

And just to blow away the whiff of brimstone which that thought might
engender among some, let's follow my argument through to its conclusion.

The constructivist argument about differences in meaning, translated into
the pscyhometric argument about group differences,

(to the extent that we can model the usual group comparisons which engage
psychometricians, on the language-meaning issues I've outlined above)

suggests that it is _meaningless_ to make any sort of comparisons across
groups, using the mean of just one of those groups. I don't say
"inequitable"- that's surely obvious- I don't say "socially impolitic"-
that's an issue of politics; I don't say "immoral"- that's an issue of
ideology: I say, literally, _meaningless_.

Stay cool!

Devi Jankowicz

Bulas K., Thomas L.L. and Whitfield F.J. (1967) _The Kosciuszko Foundation
Dictionary (English-Polish; Polish-English)_ New York: Koscisuzko
Jankowicz A.D. (1994) "The new journey to Jerusalem : mission and meaning
in the managerial crusade to eastern Europe" _Organization Studies_ 15, 4,
Maczynski J. (1991) "A Cross-cultural Comparison of Decision Participation
Based on the Vroom-Yetton Model of Leadership" Report PRE 23, Institute of
Management, Technical University of Wroclaw