The Structuralist Promise*
Zygmunt Bauman
If, as we believe to be the case, the unconscious activity o f the mind consists
in im posing form s upon content, and if these form s are fundam entally the
sam e for all minds — ancient and m odern, primitive and civilised (as the
study o f the sym bolic function expressed in language, so strikingly indicates)
— it is necessary and sufficient to grasp the unconscious structure underly­
ing each institution and each custom , in order to obtain a principle o f inter­
pretation valid for other institutions and other custom s, provided o f course
that the analysis is carried far enough (1). This program m atic statem ent by
Claude Lévi-Strauss, since it has been m ade public in 1949, excites conti­
nuous and keen curiosity o f the learned public; it has raised high, alm ost
millenarian hopes, and inspired intense, alm ost religious resistance.
The attractions o f the Lévi-Strauss’ program m e seem to be irresistible in­
deed. The pledge to get rid once and for ever o f the troublesom e ghost o f
relativism was only one, though the m ost obvious, o f its advantages. The
others are:
1. The chance to grapple, for the first time in a serious manner, with the
problem o f veritable cultural universals. So far the only approach available
was that o f M urdock’s (2): the peculiar mixture o f butterfly collectors’ (3)
m ethodology and a classified telephone directory’s inspiration. N ow it has
becom e clear that not only different cultural system s can be ‘classified’ expost-facto into the same institutionally discriminated divisions, but that they
are built up according to the sam e ‘transform ational rules’ or ‘generative
meta-grammar’. The search for universals m eans not so much stepping over
the borders o f cultural com m unities, as discovering construction principles
* A n a bridged v ersion o f this article w as p u b lish ed in The British Journal o f Sociology, 1973,
nr. 1.
90
common to the spheres o f the human praxis apparently belonging to entire­
ly different realms. In other words, universal principles instead o f com m on
denominators; theory instead o f endless rearrangem ent o f disarrayed empir­
ical records.
2. The new look on the problem o f function o f the cultural phenom ena.
By the time the structuralist m anifesto appeared anthropologists and sociol­
ogists were getting increasingly w eary o f diminishing returns and self-de­
feating tautological sterility com m on to all available varieties o f functional­
ism. The Parsons’ substitute o f ‘system ’s prerequisites’ has too much sm elled
o f anthropomorphism for som e, to o much tasted o f conservatism for others
and has been too rem ote and irrelevant o f daily problem s o f the profession
for others still. For these disenchanted with enforcing functionality o f every
single cultural item or institution taken apart, assum ptions like ‘a single
term-object has no meaning at all; any m eaning presupposes the existence
of a relation; it is on the level o f structure w here w e should seek the ele­
mentary meaningful units, not on the level o f elem ents’ (4) m eant a genuine
stroke of good fortune. If ‘discrim inating’ and ‘delim itating’ (5) are essential
functions accountable for what the cultural phenom ena are, several exciting
conclusions follow immediately. First, it is quite possible that in non-linguistic cultural sub-systems, like in language, the value o f each elem ent ‘depends
entirely on their opposition to other elem ents, on their being different from
other elements. They are therefore characterised not by any positive quality
of their own but by their oppositional quality and differential value’ (6). If
so, then a social scientist may avail him self at will o f the unquestioned
achievements of theory o f inform ation and sem iotics. From behind despair­
ingly chaotic diversity o f cultural form s, suddenly em erges an ordered struc­
ture of relationship.
3. It looks as if the controversial culture-social structure paradigm can
now find finally a satisfying solution. True, som e people hope still that
something reasonable can be said on culture-society relationship while the
paradigm remains in the analytical fram ew ork w here it was put m ore than a
century ago: this of ‘what determines w hat’ (whether on the societal or on a
single ‘social action’ plane is o f minor im portance). There seem s to be, how ­
ever, a growing understanding that if the indispensable analytical distinction
o f culture and social structure is to be salvaged at all (many a social anthro­
pologist seems to doubt, disillusioned, w hether the rescue operation is
worth the effort) it must be put on a m ore updated and less metaphysical
91
foundation (7). This new basis is offered by linguistical, or - m ore generally by the sem iotic departures. It is likely that in a socio-cultural act (like in
any other act o f sem iosis) the tw o intim ately linked though existentially and
analytically distinct sides, these o f ‘signifiant’ and ‘signifie’ (the fam ous terms
coined by Ferdinand de Saussure, but descending back to ‘sem ainon’ and ‘sem ainom enon’ o f the ancient Stoics (8)) can be located and organised respec­
tively into two isom orphic structures: the one, called usually ‘culture’, and
the second, dealt with under the nam e o f ‘social structure’. If the second is
the web o f energy-channels (it is related to availability o f resources which
determines degree o f the freedom o f action), then the first is the code
through which information on the second is articulated, conveyed and deci­
phered. The tw o aspects join together in the basic human endeavour o f re­
ducing incertitude o f the human universe, ordering it, making it m ore pre­
dictable and so m ore m anageable (9). If it is so, then the relation betw een
culture and social structure is one o f signification, and the exact methods,
elaborated for analysing isom orphic sets, can be em ployed for its study.
4. Common m isinterpretations, notw ithstanding a chance to bridge the
conceptual chasm betw een statics and dynamics, synchronic and diachronic
dimensions, is also built into the analytical equipm ent o f the modern lin­
guistics. The numerous statem ents to the contrary, frequent as they have
been, have been born out o f the understandable, though not necessarily con­
vincing, passion o f the devout preachers o f an undoubtedly revolutionary
idea. Since the heresy long ago turned into respectable routine, it has be­
com e manifest that the m ost sophisticated synchronic analysis does not re­
quire abandonment o f the diachronic perspective; on the contrary, ‘som e
connection betw een diachronic process and synchronic regularities must ex­
ist since no change can produce a synchronically unlawful state and all
synchronic states are the outcom e o f diachronic p rocesses’ (10). M oreover,
genetic and structural aspects are understandable only in their reciprocal
processual and analytical interdependence (11), and socio-cultural change as
w ell as the structure o f social and cultural system s are analysable with the
same conceptual set (12). The conceptual tool which m ost readily com es to
mind in this connection is this o f ‘unm arked’ and ‘marked’ signs (the ‘priva­
tive’ opposition o f Troubetzkoy betw een ‘m erkm altragend’ and ‘m erkm allos’
members (13). The ‘unmarked’ sign, usually simpler and m ore sketchy o f the
two, denotes initially the w hole class o f phenom ena indiscriminately; then
an attribute possessed by a sub-class only, becom es for som e reason impor­
tant, and then part o f the unmarked sign’s applications receive a ‘mark’ to
92
distinguish just this sub-class. The heretofore monopolistic unmarked sign
stands now in opposition to the new marked one; so far neutral toward the
marked feature, now conveys the information on its absence. V. V. Marty­
nov (14) has developed recently a fairly convincing theory employing the
concept of ‘markers’ showing how diachronic processes of change are con­
stantly generated by synchronic structure in virtue of its endemic rules.
There is no doubt that no serious consideration impedes substituting cultural
items for linguistic terms in the M artynov’s model. (See more of this later).
There is much more to the great structuralist promise than we have suc­
ceeded in showing by enumerating only some of its main points. No wonder
that in spite of the outspoken criticism voiced by the more traditional repre­
sentatives of anthropology and sociology (15) the ranks of scholars who try
to apply achievements of linguistics to socio-cultural analysis are getting wi­
der every year. In anthropology the application of structuralist ideas
brought remarkable accomplishments to which works by Edmund Leach
and Mary Douglas in Britain testify convincingly.
Still the case is being reinforced again and again against the linguistic
analogy and not all of it can be dismissed as a tribute paid to the conservat­
ism of institutionalised science. Those who tried it and those who did not
warn against attaching exaggerated hopes to applications of linguistical
methods to non-linguistical though human phenomena. As it is usually the
case, the onthological language is preferred to a methodological one; adver­
saries of the Lévi-Strauss programme make a point first of all of the qualita­
tive peculiarity of the non-linguistic cultural realms, which allegedly thwart
any attempt in extrapolating structuralist methodology to the general cul­
tural analysis.
•
Two issues are mixed up hopelessly in most of the criticism. The first
whether the non-linguistic realms of human culture are constructed in the
same way the language is, and so we proceed properly when trying to dis­
tinguish in them the same type of units and relationships which were discov­
ered by de Saussure, Jakobson, Hjelmsley and others in language. And the
second — whether all human culture, language including, stems from the
same universal human effort to decipher the natural order of the world and
to impose an artificial one on it, and whether in doing this all fields of cul­
ture are submitted to the same logical principles which have evolved to suit
the properties of the universe; and so we are justified in applying to the so­
cio-cultural analysis the general methodological principles, which have
achieved the highest level of elaboration and sophistication in structural lin­
93
guistics. It goes without saying that a negative answer to the first question
does not necessarily presuppose rejecting the second proposition. U nfortu­
nately, to many a critic it does.
There were so far only few cases o f defending the scientific relevance of
the first issue. One o f the m ost influential has been this o f Kenneth L. Pike
(16). Pike is concerned with exactly the op posite problem than the students
o f the second issue are: not with w hat is signified by cultural items, how cul­
tural items, how cultural phenom ena organise and order the cognitive and
operational field o f human behaviour etc., but with proving that — regard­
less their sem iotic function — there are, in all institutionalized human behav­
iour, elem entary units analogous to those o f language. The Pike contention
is that all culture is language in the form al m eaning o f the word. W hat Pike
chooses as a task to be solved is ‘the apparent irreconcilability betw een the
fact that a behaviour event is often a physical continuum with no gaps in
which the m ovem ent is stopped’ and the discrete character o f linguistic ele­
ments. The solution lies in the fact that ‘the human beings react to their ow n
behaviour and to that o f other individuals as if it w ere segm ented into the
discrete elem ents’. The part o f behaviour to which human beings react can
be taken as a ‘behaviourem a’ — an elem entary meaningful unit o f culture
analogous to a ‘sem em a’ in structural sem antics. A set o f behaviourem ae
which may be put into the sam e place (‘spot’) in the action-string are in par­
adigmatic relationship betw een them selves, exactly as sem em ae are; and
each behaviourema consists o f a peculiar com bination o f a limited amount
o f elem entary (constructing m eaningful entities, but m eaningless in them sel­
ves) building elem ents which Pike prop oses to name ‘ernes’ or ‘em ic units’.
So, Pike is convinced, the main ob stacle has been overcom e, w e have in all
human behaviour the tw o levels o f articulation which constitute the defining
feature o f any linguistic structure, and now the w ay is cleared to organise
the cultural facts into a system o f m any paradigmatic oppositions built up
out o f few phone-like bricks.
The trouble with Pike’s argument is that although language is a part of
culture (specialised in conveying inform ation alone), culture is not a lan­
guage. If not for other reasons — so at least because cultural phenom ena
perform many other functions besides informing som ebody about som e­
thing. What follow s is that it w ould be very odd indeed were the culture
built according to constructive principles made to m easure o f com m unica­
tive function alone. It is true that human beings, w hatever they do, always
build plenty o f different things out o f a limited am ount o f basic materials
(the endless variety o f each national cuisine, for instance, is achieved usually
94
with the help of relatively few basic ingredients). But stating this fact would
not bring us any closer to the understanding of human culture. The one pos­
sible result is likely to be a new version of the spurious classification-comparison feats of butterfly collectors: the ‘knowledge’ that, say, the ‘cuisine
language’ is built of salt, sugar and pepper ‘phonemes’, while the ‘language
of gestures’ is constructed of raising hands and lowering heads. It is doubt­
ful whether moving along this way we can achieve something other than
discrediting the very idea of the linguistic analogy. The fate of this analogy
does not depend, furthermore, on whether Pike will succeed in discriminat­
ing ‘emic units’ everywhere, or whether Charles F. Hockett is right when
declaring that ‘it can be demonstrated very easily that not all cultural be­
haviour consists of arrangements of discrete units of the kind that we find in
language when we analyse speech into arrangements of discrete pho­
nemes’ (17).
What seems to be really important and fruitful is the second issue of the
two mentioned above. This issue had probably Norman A. McQuown in
mind when stating that ‘the general principles which I cite are of such gen­
erality that they are probably attributes of the universe and not of human
beings in particular, or human culture in particular, or of the structure of
language in particular. . . After all, all things have structure of some kind,
and the elements within that structure contrast or complement each other,
or are in free variation with each other, or show pattern congruence, or
look elegant when we find out what the thing is like overall’ (18).
The chance offered by the structural principles discoverd by linguists con­
sists, briefly, of this: in search of the necessary general laws governing hu­
man culture we can now descend to the unconscious system which precedes
and conditions all specific empirically approachable, socio-cultural choices.
Thus we can grasp the necessary relations where they really are. The only
alternative available is the programme typified by M argaret Mead’s state­
ment: ‘More widespread similarities in cultural behaviour which occur in
different parts of the world, at different levels of cultural development’ —
should be made understandable by assuming hypothetically a possibly biolo­
gical organisation which no cultural imagination may overstep or ignore
(19). What we have been proposed here is to relate the ex-post-facto similar­
ities, located on the level of cultural usages and performances, directly to
the pre-human, universal biological nature. A procedure which can result
only in Murdock’s conviction of the biological foundation of the apparently
universal human interest in sun, moon, rain and thunder. Instead of trying to
discover the general cultural laws in the sphere of necessary, endemic and
95
generative relations, w e have been asked to locate them in the field o f the
accidental and external.
Having thus delineated the dim ensions o f linguistical analogy, w e can pro­
ceed now to enumerate som e o f the differences betw een non-linguistic and
linguistic sub-systems o f the human culture, which lay off the limits o f its
possible applications. The author is convinced that the maximum o f clarity
as to the limitations constitutes one o f the forem ost conditions to the analo­
gy’s fruitful application.
1. It is generally assumed that the linguistical process is a ‘pure communi­
cation’; the only reason why people use linguistical devices at all is that they
wish to transmit to each other som e inform ation they consider useful or im­
portant. The m ore radical version o f the above opinion says simply that
each speech event has no other function but transmitting a m essage; thus it
is a highly specialised activity and everything it consists o f can be interpret­
ed in the light o f intended com m unication or intention to elicit a specific re­
sponse.
N ot all linguists and psycho-linguists are prepared to sign this statement.
To give an exam ple o f rather forceful objections raised against radically
‘com m unicative’ image o f language w e can quote the A. T. Dittman and L.
C. Wynne list o f om nipresent attributes o f speech events which how ever
cannot be considered as parts o f the language system sensu stricto (20). The
authors distinguish, am ong others; vocal characterisers (voice breaking,
laughing background etc.), segregates (sounds which are not words), qua­
lifiers (crescendo or piano etc.), voice quality (tem po, rhythm, precision o f
articulation etc.), voice set (fatique).All these phenom ena cannot be treated
as parts o f the language proper (so w e can add) because o f their defective­
ness; instead o f being arbitrary signs, deserving their meaning to their rela­
tions with other signs, they are much closer to what was meant by Charles
Peirce when he spoke on ‘indices’; they can be read by the receiver, if he is
acquainted with som e kind o f psychological and physiological know ledge,
as information on the sender’s state; but the know ledge o f language would
hardly help in their decoding. W e w ould say with Karl Buhler (21), that
though they possess the Ausdruck quality (fonction emotive), according to
Giulio C. Lepschy (22), they have not been b estow ed with either conotative
or denotative intentions as have linguistical signs. But they do participate in
each act o f speech and thus make it much less hom ogenous than it would
seem at the first sight. Another departure o f natural languages from the
96
t
I
purely communicative model has'been pointed to by a distinguished Soviet
linguist S. K. Shaumian: ‘We would not expect to arrive at the causes of linguistical change through immanent exploration alone. The structure of lan­
guage is acted upon by psycho-physical and social factors, which are from
its point of view external; their influence cannot be taken into account be­
cause —as far as the linguistical structure is concerned —it is accidental’(23).
If even linguistical process cannot be looked upon as ‘pure communica­
tion’, doubly so the non-linguistic field of culture. With few exceptions (like
language of gestures and etiquette; it is not by accident that the word ‘lan­
guage’ has been spontaneously applied to these phenomena) the non-lin­
guistic culture operates with material which by itself is directly related to
non-informative, in some way ‘energetic’ needs. Although we can justly con­
sider the non-linguistic cultural events as information-transmitting, the ratio
information/energy is in their case much less favourable to information
than in the case of purely linguistic acts. Which means that the role of the
non-informative elements in these events is much greater than in speech­
acts, and so, almost by definition, much more influential in shaping the
events themselves. First, the ‘energetic needs’ set the limits of freedom in
adjusting the uses of a given material to semiotic purposes. Secondly, in
case of clash or a friction between informative and energetic functions it is
not always the informative one which gains the upper hand.
At least in one of his recent papers (in Linguistics at Large, ed. by Noel
Minnis, 1971, p. 139-158) Edmund Leach, whose views vary from one paper
to another, defeating any attempt at consistent reconstruction seems to im­
ply that a direct extrapolation from structural linguistics to analysis of hu­
man culture in its entirety is warranted by the fact that ‘the patterned con­
ventions of culture which make it possible for human beings to live together
in society have the specifically human quality that they are structured ‘like’
human language ‘and that the structure of human language and the struc­
ture of human culture are in some sense homologous’ (although it can al­
ways be argued what do the question marks in the word ‘like’ mean and
what is the sense of ‘in some sense . . . ’). Leach’s analysis avoids crucially im­
portant distinctive feature of non-verbal, though semiotical, sub-systems of
culture — that, to use Ronald Barthes’ words, they ‘have a substance of
expression whose essence is not to signify’; Barthes proposes to call ‘sign
functions’ these semiotical signs, whose origin is utilitarian and functional’
(Elements o f Semiology, 1969, p.41)
The most important point is that the non-linguistic branches of culture
cannot be exhausted by any description or modelling organised around the
97
informative function alone. Tw o autonom ous functions interfere constantly
with each other and no cultural phenom enon is reducible entirely to one
function only. Each cultural system , through choices it makes, orders the
world in which members o f the respective com m unity live; performs a clear­
ly informative function, e.g. reduces incertitude o f the situation, reflects
an d/or moulds the structure o f action through signalling/creating the rele­
vant portion o f the w eb o f the human interdependencies called ‘social struc­
ture’. But it also shapes the world o f concrete beings, w ho — to survive —
must satisfy their irreducible individual needs. This double aspect is clearly
discernible in shelter, dress, cuisine, drinking, m eans o f transport, leisure
patterns, etc. W e will try to elaborate this point later.
One more remark, however, is in place in this context. It is quite possible
that the basic materials which serve as the object o f human ordering activi­
ty have been in the first place pulled into the orbit o f the human universe in
virtue o f their ‘energetic’ applications. But the variety o f forms they subse­
quently acquire, the lavish abundance o f sophisticated and elaborate usages
which cluster around them, have little in com m on with their primary uses.
W e can risk a hypothesis that although the fact that artifacts o f som e kind
are being produced by human beings at all is likely to be accountable for by
basically non-informative human needs — the differentiation o f their form
and m ost of the intricacies o f their genealogical tree must be referred, to
be explicable at all, to sem iotic function they perform in relation to the so­
cial structure (i.e. in relation to the task o f ordering the human environ­
ment). The m ost recent illustration has been supplied by the wild and tech­
nologically (energetically) w asteful and senseless outburst o f im agination o f
the car producers. W ere there no stratifying function attached to the cars in
their role of signs, w e w ould hardly be able to understand the fact that so­
phisticated products o f the m odern industry becom e worn out after two
years o f use.
T o sum up — contrary to the case o f language, in analysing the non-linguistical sub-systems o f culture w e have to apply tw o com plem entary
though independent analytical fram es o f reference. N o single and qualita­
tively hom ogeneous m odel can account for all empirical phenom ena of
culture.
2. The second limitation concerns the ‘law of parsim ony’. It is frequently
assumed that in historical developm ent o f natural languages the m ost active
factors are those o f increasing econom y; not only the distinctions not
backed by isom orphic discriminations o f m eaning tend to shrink and gra98
dually disappear, but alternative types of expressive oppositions tend to
congeal thus diminishing the total number of oppositional patterns. Louis
Hjelmslev has even defined the language, in opposition to other cultural
phenomena but few (like art or games), ‘comme une structure où les élé­
ments de chaque catégorie commutent les uns avec les autres’ (24). The cen­
tral term ‘commutation’ means a correspondence between distinctions ap­
pearing on the level of ‘expression’ and those discernible on the level of
‘content’. It is Hjelmslev’s contention that expressive oppositions not backed
by isomorphic differentiations of meaning and vice versa are simply ‘extra­
model’ phenomena and are not linguistic facts proper.
Even in natural languages the amount of this type of redundancy (which
should not be mixed up with another, eufunctional type of redundancy safe­
guarding the proper deciphering of messages) seems to be however quite
impressive. B. Trnka, one of the founders of the famous Prague School,
points out that there are in each language plenty of phonomes which ‘are in
complementary distribution with each other and there is no environment in
which both of them occur’. This means that ‘their ever-present and potential
capacity for differentiating words remains unutilised’. Trnka goes as far as
concluding that, ‘strictly speaking, the true function of phonomes is not
keeping the meaning of words from each other, but only distinguishing
phonemes between each other’ (25). Much of the phonemes’ potential distin­
guishing power remains unused in every living language. Which means, that
whenever facing an opposition on the level of expression, we are entitled to
suspect a ‘commuting’ opposition on the level of content, but we cannot be
certain that there is one. Harry Hoijer has attacked the same issue from the
point of view of relics and archaisms abundant in every language: ‘There are
structural patterns like that which, in many Indo-European languages di­
vides nouns into three great classes: masculine, feminine, and neuter. This
structural pattern has no discernible semantic correlate . . . Whatever the se­
mantic implications of this structural pattern may have been in origin, and
this remains undetermined, it is now quite apparent that the pattern survives
only as a grammatical device, im portant in that function but lacking in se­
mantic value’ (26).
Whatever can be said in this connection in relation to language, the
exemptions from the ‘law of parsimony’ are much ampler in the case of nonlinguistic cultural sub-systems. Discriminating capacity of cultural items
available at any given time to any given community overgrows as a rule
their actual use. The empirical reality of each culture can be said to be full
of ‘floating’ signs, waiting for meanings to be attached to. This is, at least
99
partly, determined by the particular situation o f non-linguistic codes: while
every geographically condensed com m unity uses basically one language on­
ly, it is exposed to many criss-crossing cultural codes, institutionally separat­
ed but em ployed by the same people, though in different role contexts. The
signs float freely over institutional boundaries, but when cut off from their
intra-institutional system ic context they lose the ‘com m uting’ bond with their
original meanings. The only set available as a com m on semantic fram e of
reference for all sub-codes used by the m em ber o f a given com m unity is the
social structure o f the community as a w hole. It is true that som e signs
meaningful inside specialised ‘institutional’ sub-codes acquire also an addi­
tional discriminating quality in the com m unal ‘over-code’ (as it happens, for
instance, to the signs originated in the fram ew ork o f ‘professional’ sub­
codes, usually indicative also o f the position occupied in the overall societal
stratification) — but it is by no m eans a general rule. On the other hand,
though the human creativity is to a very great extent inspired by the de­
mand for new signs to replace the older ones, worn out because o f their fre­
quency, it could not be reduced to this cause alone. D ue to its, at least in
part, spontaneous and unm otivated character, the human creativity prod­
uces cultural items in numbers exceeding the actual sem iotic demand. These
are ‘would-be’ signs, potential signs, which for the time being do not ‘com ­
mute’ with any real distinctions in the structure o f human reality. Thirdly,
there is also the tremendous role played by tradition — by the delays in the
cultural ‘forgetting’. The developm ent o f any culture consists as much in in­
venting new items as in selective forgettin g o f the older: o f those, which in
the course o f time grew out o f their meaning, and having not found any
new sem iotic function linger as an inexplicable and m eaningless relic o f the
past. Som e o f the item s how ever refuse to disappear long after they have
been shorn o f their m eaning. Surviving som etim es only because o f de-syn­
chrony between system ’s change and socialising institutions, they defy the
functionalists’ belief in universal utility o f everything real and feed the
Durkheimean myth o f collective soul.
In short, not all elem ents in a cultural empirical reality are explicable by
referring to their sem iotic role. O nce again, what may be said on a culture
from the point o f view o f its actual sem iotic function does not exhaust the
richness o f its empirical existence.
3. One further conclusion from the com m unicative nature o f language is
that speech acts can be defined as events arising from an intention to con­
vey a message. The French team o f linguists led by André Martinet w ent far
100
enough to define the language as on e o f the ‘very wide, and so far not very
w ell delimited, kind o f social phenom ena which define them selves through
intention to com m unicate, which can be checked with behavioural criteria’.
Though the above sentence suggests that according to the authors’ opinion
the intention to com m unicate does not discrim inate the language alone,
another sentence testifies to the contrary: ‘Before it will be decided that the
art is a language, it is reasonable to investigate carefully whether the artist
has in the first place sought to com m unicate, or only to express him self’
(27). The idea o f intention to com m unicate as the defining feature o f lin­
guistic phenom ena has been so deeply entrenched in scholars’ minds, that
Lévi-Strauss, when trying originally to exp ose the linguistic nature o f the
kinship system, seem ed to assume that w hat this system is an attempt, in its
ow n sym bolic way, to achieve, is transm itting w om en or exchanging them
by men (28).
N ow it seem s doubtful whether the com m unicative function is indeed the
m ost general one, to which all m ore specific functions pursuable in human
society remain in the relation o f subordination and particularity. It might
be, but on condition that w e had defined com m unication more in the spirit
o f the m odern system theory than in the ‘exchange’ tradition o f ‘passing
som ething to som ebody by som ebody’. The modern system theory relates
the notion o f ‘com m unication’ to the concepts o f ‘dependence’, ‘orderliness’,
‘organisation’. These concepts in their turn have been defined as som e kind
o f lim itations im posed on the otherw ise unlimited (e.g. unorganised, chaotic)
space o f events (29). Tw o elem ents are mem bers o f the same system ( =
they com m unicate with each other) if not all states o f a one are possible
while the second remains in a given state. In a m ore descriptive language
w e can say that one elem ent ‘influences’ the values the second may assume.
In short, w e speak o f com m unication w henever there are som e limits im­
posed on what is possible or what can happen and what the probability of
its occurrence is. W e speak o f com m unication w henever a set o f events is
ordered, which m eans — to som e extent predictable. If w e now start from
the sociological perspective to structural linguistics and not the other way
round, w e look at the totality o f human activity as an endeavour to order,
to organise, to make predictable and m anageable the living space o f human
beings, and the language discloses itself to us as one o f the devices devel­
oped to serve this over-all aim: a device cut to m easure o f the com m unica­
tion in the narrower sense. Instead o f all the culture being a set o f particu­
larisations o f the com m unicative function em bodied in language, the lan­
guage turns into one o f the many instruments o f the generalised effort o f
101
ordering, laboured on by the culture as a w hole. This sociological approach
to language and its functions is not alien to the original intentions o f de
Saussure himself, at least according to som e o f his follow ers, A. M eillet (30)
in the first place.
It seem s that to avoid m isunderstandings caused by equivocality o f the
term ‘com m unication’, it is better to speak o f ‘ordering’ as the superior func­
tion o f the culture as a w hole. The direct effect o f a linguistic act is to order
in a way the cognitive field o f the recipient o f the m essage; as a result som e
other behavioural acts can follow , which organise the action space itself —
but these acts, though consequences o f speech, do not belong with the
sphere o f the language proper. On the other hand, the cultural events in the
broader sense (o f which purely linguistical acts can be a part) are accom ­
plished only when the particular ordering has been achieved. The culturally
institutionalised cerem ony o f addressing and greeting organises the behav­
ioural space for the interaction which follow s — through signalling what
patters o f behaviour are appropriate and stimulating the participants to
choose these patterns instead o f others. Each participant is aware o f the
fact that particular patterns are likely to be chosen by his partner, and this
know ledge enables him to plan his ow n actions and to manipulate the glob­
al situation in the fram ework o f the options which are open to him.
The specific socio-cultural way o f ordering-through-lim itation is intimately
correlated (31) with on e paramount characteristic o f the human condition;
the link between an individual’s position inside the group and his biological,
‘natural’ equipment is mediated. W hich m eans that the ‘social’ status o f any
individual is not determined unam biguously, if at all, by his natural attri­
butes in general, and his physical p ow er and prow ess in particular. Which
m eans in turn that the inherited or developed, but in both cases biological
indices o f an individual’s quality in the fram ew ork o f Nature becom e social­
ly irrelevant if not misguiding. Im pressive brawn o f a docker w ould surely
guarantee him a m ost respected status w ere he a m em ber o f a herd o f deer
or o f a birds’ pecking order. They are, how ever, utterly m isleading as signs
o f his position in a human society.
The mediation began with production o f tools. Ever since human beings
have surrounded them selves with artifacts not to be found in natural condi­
tions, products o f their m odelling activity. Once created and appropriated,
these artifacts destroyed the previous h om ology betw een the natural and
the social order by changing entirely the action-capacity o f individuals and
so creating a new arrangement o f environm ental opportunities and probabi­
lities. Thus a decisive adaptive value w as conferred on ordering of and
102
orienting in the w eb o f specifically social (which in this context means pri­
marily ‘non-natural’) relations (32).
These two requirements o f the specifically human condition — ordering
and orientation — are as a rule subsumed under tw o separate headings: so­
cial structure and culture (33). A historical study o f circum stances which led
to petrification o f tw o inseparable faces o f one coin into two, for a long
time unconnected, conceptual fram ew orks — remains to be written. W hat­
ever the reasons, however, a disproportionately time-consum ing effort has
been invested by scholars into solving o f what under closer scrutiny appears
to be a sham and artificial problem. In keeping with the notorious human
tendency to hyposthasize purely epistem ological distinctions, the tw o analy­
tical concepts coined to describe the tw o indivisible aspects of the human
ordering activity have been taken for tw o on tologically distinct beings.
The primary fact w e propose to start from is that substituting an artificial
environment for the natural one m eans that an artificial (not natural, not
created independently o f human activity) order is substituted for the natural
one. ‘Order’ is a graded notion: the level or orderliness is measured by the
degree o f predictability, e.g. by the discrepancy betw een probability indices
o f events admitted by the system and those which the system is an attempt
at eliminating. In other words, ordering m eans dividing the universe o f ab­
stractly possible events into tw o sub-sets o f — respectively — events which
occurrence is highly probable and those which hardly can be expected at all.
Ordering dissipates a certain incertitude as to the expected course o f events,
which existed heretofore. It cannot be accom plished but through selecting,
choosing a limited amount o f ‘legalised’ options from unlimited multitude o f
sequences. This understanding o f the w ay the orderliness o f a system is be­
ing achieved stands behind the classic, though forgotten, Boas’ remarks on
the intimate link betw een statistical and moral m eanings o f the ‘norm’ in the
order-generating and order-maintaining process: T h e sim ple fact that these
habits are customary, while others are not, is sufficient reason for eliminat­
ing those acts that are not customary . . . The idea o f propriety simply arises
from the continuity and automatic repetition o f these acts, which brings
about the notion that manners contrary to custom are unusual, and there­
fore not the proper manners. It may be observed in this connection that bad
manners are always accompanied by rather intense feelings o f displeasure,
the psychological reason for which can be found in the fact that the actions
in question are contrary to those which have b ecom e habitual’ (34). Let us
turn our attention to the fact that Boas d oes not distinguish betw een orderestablishing and orientating-in-order faculties, probably assuming tacitly that
103
w e som ehow like and evaluate favourably the habitual and expectable while
disliking and rejecting the unusual and sudden (a conjecture which was
granted a full corroboration by psychologists); and that this single human
capacity is accountable for both need o f order and efficiency o f the culture’s
guiding function. A single vehicle is enough to achieve both aims — as or­
dering (structuring) means making the ordered sector meaningful, e.g. arriv­
ing at a situation in which som e concrete events fo llo w usually a particular
condition, and secondly som e beings to whom the sector is meaningful
know that these events do follow it indeed. In other words, the sector is
meaningful to those to w hom it is if and only if they possess som e informa­
tion on its dynamic tendencies. The divergence betw een the information ac­
tually needed to determine the sector com pletely and the amount o f infor­
mation which would be necessary w ere the sector entirely ‘unorganised’
measures the degree o f its ‘m eaningfulness’.
W e have arrived this far w ithout having distinguished conceptually the
tw o aspects o f the human ordering effort: introducing meaning into the
otherw ise m eaningless universe and supplying it with indices able to signal
and reveal this meaning to those w h o can read. Both sides of the tw o­
pronged endeavour — it looks — can be described and understood in a sin­
gle analytical fram ework. The question arises, w hether any other frame o f
reference or conceptual set, besides the on e necessary to analyse the order­
ing activity itself, ought to be brought in to explain the social structure —
culture relationship. Orderliness o f the world they live in is so vitally impor­
tant to human beings that it seem s entirely justified to ascribe to it an autothelic value. It is hardly necessary, if not redundant, to seek a further expla­
nation to the above need by pointing to a purpose which ‘making the world
meaningful’ allegedly serves.
Consequently, it seem s that the logic o f culture is the logic o f the self-re­
gulating system rather than the logic o f the code or o f the generative gram ­
mar o f language — this latter being a peculiar case o f the form er rather
than the other way round. The m ost im portant conclusion which follow s is
the following: w e are justified in extrapolating (to the non-linguistic spheres
o f culture) o f the m ost general features o f language only; exactly these fea­
tures, which characterise the linguistic interaction in its capacity o f a case o f
a more inclusive class o f self-regulating system s. T herefore w e had better
turn for inspiration directly to the system theory. W hich does not necessari­
ly mean that borrow ing from the im pressive achievem ents o f the linguistical
analysis o f the nature o f signifying should stop. W hat it does mean is that
while allowing ourselves to be inspired by linguistics’ achievem ents we
104
ought to be aware that they have no more proving power than analogies
usually do.
4. In its everyday use the term ‘sign’ means simply aliquid stat pro aliquo
and the attention of the students of ‘meaning’ was traditionally turned to
conditions on which a something ‘stands’ for something else. Closing — in
the light of the theory of learning — the long line of development of the behaviouristic interpretations of sign, which began with Watson and passed
through classic works of C. K. Ogden and I. A. Richards and Charles Mor­
ris — Charles E. Osgood defined in 1952 the sign as something which
evolves in an organism a mediating reaction, this a. being some fractional
part of the total behaviour elicited by the object and b. producing distinc­
tive self-stimulation that mediates responses which would not occur without
the previous association of nonobject and object patterns of stimulation’
(35). Thus from the behaviouristic perspective to solve the problem of
meaning is to show that a ‘nonobject’ through its association with ‘the ob­
ject’ evokes responses akin to those stimulated by the object. For a psychol­
ogist whom a behaviourist would call ‘mentalist’, ‘standing for’ means ‘send­
ing to’, which differs substantially from the behaviouristic definition in terms
it employs, but remains very well within the framework of the only question
any psychologist bothers himself with: just what a sign is to somebody to
whom it is already or becomes a sign. As we have seen before, for a sociol­
ogist or ‘culturologist’ the main question is different: how it comes, that a
‘something’ acquires a non-natural, non-intrinsic power to stand for some­
thing else and so serve in the role of a sign. That is why — technically —
the problem of socio- and culturo-logists is much closer to this dealt with by
structural linguists who try to solve exactly the question of conditions which
have to be met by a ‘non-object’, not to evoke responses ‘natural’ for a con­
crete ‘object’, but to be able to evoke any responses at all.
Some linguists went so far as to distinguish between two entirely different
types of information which stay allegedly behind the two questions. Thus
according to Bertil Malberg a message ‘may be said to contain information
in a twofold way. It has its ‘meaning’, which is the traditional popular inter­
pretation of the concept. The message ‘gives us information about some­
thing’. But information also may imply what we can call here the distinctive
information; i.e. the distinctive characteristics which make it possible for the
receiver to identify the signs — or more exactly their expression level, for
this information does not necessarily imply understanding of the message’
(36). The secret of signifying, conveying information etc. lies in the first
105
place in relationships between sign-bodies them selves (syntactic relations,
according to the classic three-fold classification by Charles Morris). V. A.
Zvegintsev thinks it proper even to define the language, the m ost developed
and specialised natural sign-system, by the qualities o f peculiar inter-sign re­
lations. It is due to these relations that the language plays the role o f an ‘in­
strument o f discreteness, a classification system which em erges in the course
o f human speech activity . . . D issecting the perceived and sensed continuum
o f the world into discrete units, the language supplies m en with m eans en­
abling them to com m unicate through sp eech ’ (37).
Here w e com e across first im portant feature o f signs: they are discrete,
separate, different from each other, and their being different is the very con­
dition o f playing the role o f signs, o f being perceived as signs, o f ‘standing
for’ or ‘sending to’. It becom es clear at this m om ent how misleading it can
be to confine discussion o f signs to their relation to the signified object.
N othing can be learnt on the nature o f signs from studying a relation be­
tween a single sign and a single signified object. Certainly discreteness and
differentiation o f signs, which appears to be their first defining feature, can­
not be discovered in the frame o f a ‘single sign — single object’ correspon­
dence. This correspondence to be possible at all — the signs must enter be­
forehand som e determined relations betw een them selves.
Roman Jakobson asserts repeatedly that it w as Charles S. Peirce w ho dis­
covered these initial conditions o f any signifying, e.g. meaningful phenom e­
na. It was him w ho decided that ‘pour être com pris, le signe — et en parti­
culier le signe linguistique — exige non seulem ent que deux protagonistes
participent à l’acte de parole, mais il a besoin, en outre, d’un interprétant. . .
La fonction de cet interprétant est rem plie par un autre signe, ou un ensem ­
ble de signes, qui sont donnés concurrem m ent au signe en question, ou qui
pourraient lui être substitués’ (38). Typical for the m ore recent statem ents of
the problem is the rather blunt form ulation o f A. J. Greimas: ‘La significa­
tion présuppose l’existence de la relation: c’est apparition de la relation
entre les termes qui est la condition nécessaire de la signification . . . C’est au
niveau des structures qu’il faut chercher les unités significatives élém en­
taires, et non au niveau des élém ents’ (39). André Martinet and his disciples
are even more explicit and definite: ‘l’inform ation n’est pas donnée par le
m essage lui-même, mais par sa relation avec les m essages auxquels il s’o p ­
p ose’ (40).
Insightful conjectures o f Peirce w ere in the course o f tim e reinforced and
corroborated by the m odern theory o f inform ation and have turned into un­
shakable foundations o f the contem porary understanding o f signs and sig­
106
nifying function. One sign taken by itself has no m eaning at all; what it does
mean is a difference between signs which could be alternatively used in the
same place. What follows, is that any inform ation is and can be conveyed
by presence or absence o f a particular sign, not by the immanent qualities
o f the sign itself. Which means in turn that the m ost important, defining at­
tributes of a sign are exactly those w ho discrim inate it from alternative
signs — and this discriminating capacity is the only thing which m atters in
conveying information — e.g. in turning a chaos into a m eaningful system,
or — in more general terms — in reducing the level o f incertitude.
N ow if human culture is a meaningful system (and it must be, if ordering
human environment and patterning human relations is one o f its universally
admitted functions) what w as said heretofore on the nature o f signifying is
entirely relevant in its context. W hich m eans that trying to establish the
meaning of a cultural item by analysing this item apart from others, in itself
— is sometimes totally misleading and irrelevant and always inexhaustive
and partial. But that is exactly w hat functionalists from M alinowski on
usually do. They either try, as M alinow ski him self did, to explain cultural
phenomena through relating them to individual needs they allegedly satisfy
(to this habit G eorge Balandier rightly retorted, that “la place que Mali­
nowski accorde aux besoins, dits ‘fondam entaux’, peut inciter à trouver l’ex­
plication des phénom ènes sociaux par un procédé (très aléatoire et très sus­
pect du point de vue scientifique” (41) — which it surely does) or — this
time faithfully to the Durkheimean tradition — they mould an anthropo­
morphic concept o f the ‘need o f the system ’ to allege a reasonable function
for every single cultural pattern. Both approaches quite obviously contra­
dict the m ethodological im perative to relate m eanings to oppositions be­
tween signs and not to each sign taken in isolation. The meaning o f a sign
becomes transparent not in the context o f som e non-sem iotic entities, but in
the context of other signs to which the sign under analysis is system atically
related.
Having focused our attention on differences betw een cultural item s and
patterns in their semiotic (inform ation-conveying) role, w e should not how ­
ever conclude that every difference in physical shape o f the items is neces­
sarily meaning-loaded. Significant are these differences only, which exist be­
tween alternative items, e.g. items which can substitute each other in the
same situation, in the same place o f the string o f human interaction. To this
semiotically important category belong different behavioural patterns em ­
ployed by two persons in addressing each other, evening and casual dresses,
mini and ‘simple’ skirts, doors with and w ithout ‘N o entry’ label or tw o sides
107
o f the same door with the ‘no entry’ label on one side only. These are ‘paradigmatically opposed’ items, e.g. item s mutually replaceable in the sam e sec­
tion o f behavioural string. W henever tw o cultural item s are paradigmatically opposed, w e can, conversely suspect that they con vey an inform ation on
som e non-semiotic reality. B efore any o f the paradigmatically opposed
item s or patterns w as em ployed, the situation had been uncertain as each o f
the items was to som e extent likely to appear; after on e o f the items did ap­
pear instead o f the others, the uncertainty w as reduced and so an order was
achieved.
According to the well-know n typ ology o f N. S. Trubetzkoy (42) the tw o
members o f a meaningful opposition may differ from each other in three al­
ternative ways: each member may possess, apart from the part com m on to
both, also an elem ent which does not appear in the other; these are ‘Equipollente Oppositionen’. Or every m em ber possesses the sam e quality but in
different degree; these are ‘G raduelle O ppositionen’. And there is also a
third category called ‘privative’: ‘O ppositionen sind solche, bei denen das
eine O ppositionsglied durch das Vorhandensein, das andere durch das
Nichtvorhandensein eines M erkm ales gekennzeichnet wird’. This type o f op­
position, which members are correspondingly ‘m erkm altragend’ and ‘merkm allos’ (Trubetzkoy), ‘marked’ and ‘unm arked’ (‘Le langage peut se contenter de l’opposition de quelque ch ose avec rien” — Jakobson (43) ‘intensive’
and ‘extensive’ (L. H jelm slev), though statistically less frequent than equipol­
lent opposition, is b estow ed with som e particular features which should fo­
cus attention o f any student o f culture. The m ost im portant feature consists
in a ‘double meaning’ o f the unmarked member: it ‘represents’ either the en­
tire category, or one part o f it — the on e left after the marked m em ber had
‘cut o ff the other. Thus the unmarked m em ber is indicative o f certain cate­
gory o f entities, but says nothing about presence or absence o f a certain
feature which appearance is signified by the m arked member (is neutral to­
ward this feature). Joseph H. G reenberg is fascinated by this ‘pervasive na­
ture in human thinking o f this tendency to take one o f the m embers o f an
oppositional category as unmarked so that it represents either the entire ca­
tegory or par excellence the op posite m em ber to the m arked category’ (44)
to the point o f declaring the privative opposition one o f the m ost pertinent
‘language universal’.
There are reasons to assume that the ‘unmarked -marked’ opposition, be­
ing much more a general form o f human ordering activity than a specific
linguistic device, plays a crucial role in functioning o f culture in general and
in its dynamics in particular. It seem s, by the way, that this peculiar type of
108
opposition caused generations o f anthropologists to overlook distinctive
functions of cultural entities and induced them to concentrate on analysing
separate items. It happened this way, because — by its very nature — the
unmarked category discloses its ‘unm arkedness’ only when confronted deli­
berately with a marked one. Usually how ever we do not perceive it in terms
of distinguishing; it denotes to us a ‘norm al’, universal state of things, a
‘norm’ in the statistical sense whose very prevalence inspires a tacit- assump­
tion that there must be some ‘general hum an needs’ which make the particu­
lar unmarked category required and inevitable. The unm arked is a back­
ground rather than a distinctive feature. W e had a special name for ‘mini­
dresses’, but had no name for the rest of ‘just-so-dresses’; we w ere ready to
admit that the mini-dresses som ehow distinguish their wearers, that they
convey a specific message, are loaded with a particular symbolic value etc.
— but at the same time it would hardly occur to m ost th at since ‘minidress­
es’ have appeared the same can be said on ‘just-so-dresses’; as to the latter,
we have kept convinced that they perform some purely psychological func­
tion (body tem perature protection) and — may be — some vaguely moral
one, too diffused and universal to arouse suspicion as to its sectarian-discrim­
inating character. It took some time fo r the mini-dresses to become so wide­
spread and ‘normal’ as to turn into a new semiotically neutral background
and by their very frequency, bordering on ‘norm alcy’, to seem to be stripped
of any distinguishing capacity. Thus the ground was prepared for trium­
phant appearance of the maxi.
In an extremely stimylating treatise by V ictor M artynov (45) we find the
following hypothesis. If the nuclear structure of a semiotically relevant sen­
tence is SAO (Subject-Action-Object), then we can pass from one sentence
V’ to another V” through modifying one of the three m em bers of the struc­
ture. ‘Modifiers’ are those new signs which are being added to one of the po­
lar members; ‘Actualizers’ are signs added to the central element. Let us
notice that both modified and actualised m em bers are related to their pre­
vious versions as ‘m arked’ signs to ‘unm arked’: S” is the m arked m em ber of
the opposition S” - S’ etc. This is actually the only way of creating new
meanings; it always leads through cutting off some p art of form erly undis­
criminated category, through extracting a specified feature of a particular
subset of a larger set. Sometimes the older signs absorb their modifiers or
actualizers (when they are frequently used together) while transform ing
their own shape; this process was called by A. V. Isatchenko (46) the ‘se­
mantic condensation’ and it seems to be responsible, a t least in part, for the
difficulties which any attem pt to trace back the com m on roots of diversified
109
signs regularly encounters. Still on e is inclined to suspect that ‘adding
marks’ (modifiers or actualizers) to already existing signs ( = introducing
finer, more subtle and m ore discrim inating distinctions into a heretofore un­
divided category) provides the main, if not the only road to ramification and
enrichment o f any sem iotic code. It w as noticed also by M artynov that the
marks can be characterised by their peculiar ‘w andering capacity’ inside the
nuclear structure: modifiers can turn into actualizers and vice versa (Man in
office should respect his elders — men should respect their elders in office
— man should respect in office his elders), which m eans that the sam e or
affine ‘marked’ meaning o f the relation in toto can be expressed interchan­
geably through marking the subject o f action, its object or the action pat­
tern itself.
N ow there is a striking h om ology betw een the nuclear structure o f a sen­
tence, analysed by M artynow, and the nuclear structure o f social relation
( = relation betw een socially institutionalised roles), as analysed, say, by S. F.
Nadel (47).
Behavioural pattern and a corresponding social role are not only intrinsi­
cally interconnected; they are, as a m atter o f fact, tw o com plem entary ways
o f conceptualising the same repetitive and recurrent process o f interacting.
The changing relation betw een tw o individuals (or, m ore properly, tw o cate­
gories o f individuals) com es into relief through changing the social defini­
tion o f role and changing the ascribed behavioural pattern at the sam e time.
Practically — w e can hypothesize — appearance o f a marked, discriminated
sub-pattern o f behaviour leads consequently to distinguishing, inside the
broader role, o f a new, marked and narrow er subcategory. N ew roles in a
ramifying social structure appear to be categorised institutionalisations o f a
new, more specialised and specific function. The basic device operative in
the process leading from the nuclear structure R’jA ’R ^ to the m ore specific
nuclear structure R’j A ”R”2 are on ce again ‘modifiers’ and ‘actualizers’ — in
short, markers and marking.
5. N ow , one o f the basic axions o f the structural linguistics is that the
form o f expression is basically arbitrary toward the content denoted. In
terms proposed by de Saussure, the ‘signifiant’ is ‘unm otivated’ by the ‘sig­
nifie’. N ot all linguists o f stature w ould agree with this contention. One o f
the first to protest against the extrem ity o f de Saussure’s attitude has been
Emile Benveniste: ‘Entre le signifiant et signifie, le lien n’est pas arbitraire;
au contraire, il est nécessaire.. .Ensem ble des deux ont été imprimés dans
m on esprit; ensem ble ils s’évoquent en tout circonstance’ (48). Today the
110
same arguments are being exposed by R om an Jakobson. The essence o f the
argument is the intimate tie betw een a ‘thought’ or an ‘idea’ on the one
hand, and the cluster of phonem es through which this idea is expressed and
conveyed, on the other. The uttering o f particular sound evokes, if deci­
phered rightly, a particular idea; and this idea cannot exist but in its accept­
ed expressive form; its existence is m ediated and accom plished by the ‘signifiant’.
However controversial the issue has been in the field o f linguistics, there
is no doubt that in the socio-cultural phenom ena the ‘cultural signs’ and the
corresponding social relations are in m ost reciprocally m otivated and not
arbitrary towards each other. Their mutual relations can o f course assume
all shades of the spectre from entirely accidental genetically to interwoven
to the point of identity. But the frequency o f relations close to the second
pole of the continuum caused innumerable trespassings o f analytical borders
between sociology and ‘culturology’ (w hatever its institutionalised name),
and — the worst o f all — plenty o f efforts w asted on phony problems
of whether the ‘ultimate essence’ o f the society is cultural or social. A s a
matter of fact all phenomena o f human life seem to be socio-cultural in the
Benveniste’s or Jakobson’s sense: the w eb o f social dependencies called ‘so­
cial structure’ is unimaginable in any form but cultural, while m ost o f the
empirical reality of culture signals and brings into existence the social order
accomplished by the established lim itations. The fam ed U ngeheuer’s princi­
ple “Im ‘Kanal’ fliessen nur Zeichenkorper” (49) — is obviously irrelevant in
case of the communication in the broader sense, which accounts for the
overwhelming majority of socio-cultural phenom ena. W hile choosing a par­
ticular cultural pattern w e create in the sector o f a given social action the
web of dependencies which can be generalised into a total model o f the so­
cial structure. And we cannot arrive at anything generalisable into this con­
cept in any way but the one made possible by the available resources o f cul­
tural patterns. The social structure exists through the ever continuing pro­
cess of the social praxis; and this particular kind o f existence is rendered
possible by the fact that the praxis is patterned by a limited amount o f cul­
tural models.
If asked to express the ‘structuralist prom ise’ in one brief sentence, I
would point to the unique chance o f overcom ing the notorious duality o f so­
ciological analysis while avoiding sim ultaneously the tem ptation to slip into
one of its two extremist alternatives. There w ere recently attempts to adopt
structuralist method to traditional spiritualist idioms through a single device
111
o f postulating the realm o f m entalistically interpreted ‘m eaning’ as the se­
mantic field o f cultural signs. It is my conviction that the structuralist
prom ise can be transferred from possibility into reality only if it is under­
stood that the role played in linguistical analysis by the sem iotic field is as­
sumed, in the world o f human relations, by social structure.
N otes
1. Lévi-Strauss, C lau d e, Structural Anthropology, N e w Y o rk , 1967, p. 21-22.
2. C f. M u rd o ck , G e o rg e P., ‘C u ltu ra l U n iv e rsals’, in: Readings in Anthropology, ed. by E. A dam ­
son H oebel e t al„ N ew Y o rk , 1955. A lso b y th e sa m e a u th o r ‘T h e C o m m o n D e n o m in a to r in
C ultu res’, in: The Science o f Man in the World Crisis, ed. by R alph L inton, N ew Y o rk , 1945.
T h e w ell-nigh o p p o site ep istem o lo g ical s ta tu s o f th e M u rd o c k ’s-like u niversals an d the type
Lévi-Strauss is a fte r is w ell illu stra te d by th e B lo om field-C hom sky c o n tro v e rs y in linguistics.
In keeping w ith th e n o m in alistically p a rtic u la ris t tra d itio n o f B oas-inspired study o f languages,
Bloom field coined in 1933 th e p h ra se w hich w as to b e co m e th e unc ritic a lly acc e p te d idiom fo r
m ost linguists aim ing a t g e n eral sta te m e n ts : ‘T h e o n ly useful g e n eralisa tio n s a b o u t la nguage
a re inductive g e n eralisa tio n s' (Language, N ew Y o rk , p. 20). T his m inim alist p ro g ra m m e w as
ch allen g ed im plicitly, if n o t explicitly, b y C h o m sk y ’s c o n c e p t o f ‘d e ep -ro o te d fo rm al u n iv er­
sals’, based o n a th e o re tic al assu m p tio n th a t ‘all la n g u a g es a re c u t to the sam e p a tte rn ’, and
th a t real p ro g ress in linguistics c o n sists in th e d isco v e rin g th a t c e rta in fe a tu re s o f given lan­
gu ag es can be red u ced to u n iv ersal p ro p e rtie s o f lan g u ag e, a n d ex p la in ed in te rm s o f these
d e e p e r asp ects o f linguistic fo rm ’ (Aspects o f the Theory o f Syntax, C am b rid g e (M ass.), 1965,
p. 30, 35). T he new a p p ro a c h h ad b een a n tic ip a te d , th o u g h s o m e w h a t incon seq u en tially and
vaguely, in K lu ck h o h n ’s id ea o f ‘u n d e rly in g d e te rm in a n t fa c to rs ’ o f la n g u a g e e n titie s lo c a ted
o n an en tirely d ifferen t p la n e th a n em p iric a lly accessib le p h e n o m e n a o f lan g u ag e p ro p e r. T he
idea w as sw eepingly v ast o n an alm o st L ev i-S trau ssian scale: in v a ria n t, u nderlying fa c to rs ‘a re
3.
4.
5.
6.
to b e found in th e n a tu re o f social system s, in th e b io lo g ical an d psychological n a tu re o f the
c o m p o n e n t individuals, in th e e x te rn a l c o n d itio n s l n w hich th e y live and act, in th e n a tu re o f
actio n itself, in th e necessity o f its c o rre la tio n in social sy stem s' (‘U n iv ersal c a te g o rie s o f cul­
tu re ’, in: Anthropology Today, ed. by A. L. K o re b e r, C h icag o , 1953, p. 513).
T h e term ap p lied by E d m u n d L each to th e a d m ire rs o f ‘c lassificatory m e th o d s’, in sp ired by
R adcliffe-B row n. C f. his Rethinking Anthropology, L o n d o n , 1961, p. 2-3.
G re im a s, A. J., Semantique structurale, Paris, 1966, p. 19, 20.
Cf. ‘D istinktive’ und ‘d elim itativ e’ ‘F u n k tio n e n ’ — N . S. T ru b e tz k o y , Grundziige der Phonologie, G o ttin g e n , 1967, p. 241.
L eroy, M aurice, The Main Trends in Modern Linguistics, O x fo rd , 1967, p. 55
7. S om e less m o d e ra te e n th u siast o f ‘e th n o m e th o d o lo g y ’, k een o n d ra w in g o n to lo g ic a l c onclu­
sions fro m w h a t w as initially co n ceiv ed as a m e th o d o lo g ic a l choice, p ro v id e re c e n t e xam ples
o f revived in te re st in o ld p a ra d ig m . C f. ‘T h e m e m b e r’s sen se th a t h e lives in a real w orld
s h ared in co m m o n w ith o th e rs is th e fo u n d a tio n o f his b ein g in th e w o rld ’ (T hom as P. W ilson,
in: Understanding Everyday Life, ed . by Jack D. D o u g las, L o n d o n , 1971, p. 19); ‘T e x tu re o f
the scene, including its a p p e a ra n c e as an o b je c tiv e , re c a lc itra n t o rd e r o f affairs, is th e a cc o m ­
plishm ent o f m e m b e r’s m e th o d s fo r d isp la y in g an d d e te c tin g the settin g ’s fe a tu re s ’ (D o n
112
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
26.
27.
28.
29.
H. Zimmerman & Melvin Power, ibid., p. 195). W hich rem inds me again of the innocent zeal
of a verse printed in the times of La grande révolution:
Les grands ne nous paraissent grands
Que parce que nous sommes à genoux.
— Levons — nous!
The rhyms were used in 1846 by M arx to ridicule Bruno Bauer.
Cf. A la recherche de l ’e ssence du langage, 1965, 51 p. 22.
Cf. Bauman, Z., “Marx and the C ontem porary T heory o f Culture', in: Marx and Contempora­
ry Scientific Thought, The Hague, 1969, p. 483-497.
Greenberg, Joseph H., “Language U niversals’, in: Current Trends in Linguistics, ed. by Tho­
mas A. Sebeok, vol. Ill, The Hague, 1966, p. 61.
Cf. Goldman, Lucien, ‘Introduction G énérale’, in: Entretiens sur la notion de genèse et de
structure. The Hague 1965, p. 12.
Cf. Bauman, Z., ‘Semiotics and the functions o f culture’, in: Social Science Information, 1968,
5, p. 69-80.
Op. cit., p. 67.
‘Kibernetika, semiotika, lingvistika’ in: Nauka i Technike, Minsk, 1966, p. 118 ff.
There was a whole spectre of reactions ranging from contem ptuous silence, through as tren­
chant as ignorant summary sneer, up to bringing into relief the French structuralists’ one­
sidedness without m entioning one’s own. F or a sociologist of science in general, of scientific
revolutions in particular, it may be illuminating th at even some converts to the Lévi-Strauss
programme (brought up initially in the Boas-Malinowski or Durkheim ean tradition) thought it
necessary to voice in public their restraint and to pretend that no qualitative shift has hap­
pened.
The fullest version of his theory is contained in the three volumes o f his Language in Relation
to a Unified Theory o f the Structure o f Human Behaviour, Sum m er Institute o f Linguistics,
Glendale, Cal., 1954-1960. The quotations which follow have been taken from Pike's paper
‘Towards a Theory of the Structure of Hum an Behaviour’, in: Laguage in Culture and Society,
ed. by Dell Hymes, New Y ork 1964, p. 54-62.
In: Language in Culture, Conference in the Interrelations o f Language and other Aspects o f
Culture, 23-27 March 1953, ed. by H arry H oijer, Chicago, 1960, p. 163.
Ibid., p. 162.
Cf. 'Anthropological Data and the Problem of Instinct', in: Personality in Nature, Society and
Culture, ed. by Clyde Kluckhohn and C. M urray, Knopf, New York, 1949, p. 111.
Cf. ‘Linguistic Techniques and the Analysis of Em otionality’ in: Interviews, J. Abner. Soc.
Psych.', 1961.
Karl Bucher, Sprachtheorie, Jena, 1934.
La linguistique structurale, Paris 1968, p. 28
Strukturnaja Lingvistika kak immanentnaja tieoria jazyka, Moscow, 1958, p. 29.
Le langage, Paris, 1966, p. 135.
Trnke B. et al„ ‘Prague Structural Linguistics,’ in: Classics in Linguistics, ed. by Donald E.
Hayden et al., New Y ork, 1967, p. 327.
‘The Sapir-Whorf hypothesis’, in: Language in Culture, p. 97-98.
La linguistique, sous la direction d’A ndré M artinet, Paris, 1969, p. 165.
Cf. Structural Anthropology, p. 4445.
Cf. for instance W. R. Ashby, R. W. Sperry and G . W. Z opf Jr. in: Principles o f Self-Organisa­
tion, Transactions o f the Illinois Symposium on Self-Organisation, 8-9 June 1961, ed. by Heinz
von Foerster and G eorge W. Z opf Jr., London, 1962.
113
30. Cf. ‘Le développem ent des langues’, in: Linguistique historique et linguistique générale, vol. Il,
Paris, 1936, p. 75 ff.
31. I wish to stress that 1 use the term ‘co rrelated ’ instead of ‘determ ining’ o r ‘determ ined’. The
discussion on ‘determ ination’ seems to be as barren as it has been long. The real relationship
between the two factors reminds much m ore o f w hat has been nam ed by cyberneticians the
‘positive feedback’.
32. A part from the tight bond betw een tools and emergence of socio-cultural order — there is al­
so an intimate link betw een the level of tools’ developm ent and the types of the socio-cultural
regulating system. A good m odern illustration has been pointed to by William G. Elliot Jr.:
‘W ithout the m otor vehicle, highway signs m ight well have rem ained primitive, local and high­
ly individualistic. The m otor vehicle th at trem endously expanded the range of travel and
brought an era of individual travel for m asses also created new hazards and a need for vastly
improved guidance fo r the strangers who w ere following new highways into distant places’
(‘Symbology of the Highways of the W orld’, in: Symbology, by A rt Directors Club of New
York, 1960, p. 50).
33. W hat follows is one of the many recent m anifestations of the traditional paradigm : T. O.
Beidelman discusses ‘the interplay betw een culture and society’ (‘Som e sociological implica­
tions o f culture’, in: Theoretical Sociology, ed. by John C. McKinney & Edward A. Tiryakian,
New York, 1970, p. 500) as this ‘betw een ideology (as exhibited in cosm ology and moral
norms) and social action (as exhibited both in adherence to and divergence from such norms)’.
34. Introduction to the Handbook o f American Indian Languages, Sm ithsonian Institution, 1911;
reprinted in: Classics in Linguistics, ed. by D onald E. H ayden et. al., New York, 1967, p. 220.
35. Osgood, Charles E., ‘On the N ature of M eaning’, in: Current Perspectives in Social Psycholo­
gy, ed. by E. P. H ollander and Raymond G. Hunt, New Y ork 1963.
36. Malmberg, Berzil, Structural Linguistics and Human Communication, Berlin, 1963, p. 31.
37. Zvegintsev, V. A., Teoreticheskaja i prikiadnaja lingvistika, M oscow, 1967, p. 421.
38. Jakobson, Roman, ‘Le langage commun des linguistes et des anthropologues’, in: Essais de lin­
guistique générale, Paris, 1963, p. 40.
39. G reim as, A. J., Sémantique structurale, Paris, 1966, p. 19, 20.
40. La linguistique, p. 155.
41. Balandier, G., ‘L’expérience de l’ethnologue et le problèm e de l’explication’, in: Cahiers inter­
nationaux de sociologie, Décembre 1956.
42. Cf. Trubetzkoy, N. S., Grundziige der Phonologie, G ôttingen 1967, p. 67.
43. Zero, Signe, in: Mélanges de linguistique, offerts à Charles Bally, G eneva, 1939, p. 144.
44. In: Current Trends in Linguistics, ed. by T hom as A. Sebeok, vol 111, 1966, p. 72.
45. Cf. Martynov, V. V., ‘Kibernetika, sem iotika, lingvistika’, in: Nauka i technika, Minsk, 1966, p.
72.
46. Isatchenko, A. V., K voprosu o strukturnoj tipo logii slovarnovo sostava slavianskich jazykov, 1958, 3.
47. Nadel, C. F. S. F., The Theory o f Social Structure, London, 1957, especially pp. 22-26, 60.
48. ‘N ature de signe linguistique’, in: Acta linguistica, 1939. R eprinted in: Readings in Linguistics,
vol. II, ed. by Eric P. Hemp et al., Chicago 1966, p. 105-106.
49. Ungeheuer, G., ‘Einfiihrung in die Inform ationstheorie unter Berücksichtigung phonetischer
Problème’, in: Phonetika, 1959.
114