300 Linda R. Waugh Definition of markedness In a letter of July 31, 1930 to Roman Jakobson, Nikolaj Trubetzkoy outlined his reflections on the interconnection between correlative phonemes, and the concept of markedness was overtly defined in linguistic analysis for the first time: Statistics has nothing to do with it. And the essence lies in the so-to-speak intrinsic content* of the correlation. Apparently any (or might it not be 'any'?) phonological correlation acquires in the linguistic consciousness the form of a contraposition of the presence of a certain mark to its absence (or of the maximum of a certain mark to its minimum). Thus, one of the terms of the correlation necessarily .proves to be 'positive*, 'active', and the other becomes 'negative', "passive'. For instance, the tonality differences between correlative consonants are objectively confined to a 'maximally high tonality vs. maximally low tonality', but subjectively they are always transformed into an opposition 'heightened tonality vs. non-heightened tonality' (—'maximally high vs. minimally high') or into an opposition 'unlowered vs. lowered tonality' (='minimally low vs. maximally low'); the former type is implemented for instance in the correlation 'palatalized vs. non-palatalized' consonant and the latter for instance in the North Caucasian correlation 'labialized vs. non-labialized' consonant (probably also in the Semitic correlation 'emphatic vs. non-emphatic' consonant). In both cases, only one of the terms of the correlation is conceived of as actively modified and positively endowed with a certain mark, while the other is merely conceived of as non-endowed by this mark and thtds passively modified. [Trubetzkoy 1975: 162f.; italics mine]. In his reply, Roman Jakobson underscored the utility of the concept for conceptual domains in general: I am coming increasingly to the conviction that your thought about correlation as a constant mutual connection between a marked and unmarked type is one of your most remarkable and fruitful ideas. It seems to me that it has a significance not only for linguistics but also for ethnology and the history of culture, and that such historico-cultural correlations as life ** death, liberty"~ non-liberty, sin <*> virtue, holidays ** working days, etc., are always confined to relations a ~ non-a, and that it is important to find out for any epoch, group, nation, etc., what the marked element is. For instance, Majakovskij viewed life as a marked element realizable only when motivated; for him not death but life required a motivation. Cf. the wa y the relation of life and death differs for the two heroes of Tolstoy's 'Master and Worker'. Another example: the Chekists said that everyone is a man of the White Guard, and if not, it must be proved in every separate case. Here the Soviet allegiance is a marked element. At present in Soviet print there has emerged a slogan; they used to say that 'all those who are not against us are with u$% but now they say 'all those who are not with us are against us'. That points to a shift of tlie elements, i.e. to a generalization of the Chekist standpoint. I'm convinced that Marked and unmarked 301 many ethnographic phenomena, ideologies, etc. which at first glance seem to be identical, often differ only in the fact that what for one system is a marked term / may be evaluated by the other precisely as the absence of a mark (Trubetzkoy ; 1975: I62f.) , Subsequent work on grammatical meaning in language (cf. Jakobson L 1932, 1936, 1939; van Schooneveld 1978a; Waugh 1975, 1977), and to a I certain extent on lexical meaning and word formation (see van Schooneveld 1978a, 1978b; Waugh 1976a; Howden 1979; Salus and Salus 1978), has shown that in paired grammatical categories there exists the 1 same type of asymmetry between correlated elements, with one of the elements conceived of as endowed with a mark and the other conceived of r as nonendowed with that same mark. In particular, it was found that | there is a constraining, focusing characteristic for the marked term of any grammatical opposition: the marked term necessarily conveys a more narrowly specified and delimited conceptual item than the unmarked. As a consequence, the marked element signals a certain grammatical concept ; that the unmarked leaves unsignaled: the marked term of any grammatical _ opposition specifies a particular unit of information x, while the un-\ marked term does not necessarily specify that particular unit of infor-f mation. Thus, to take an example that will return again, in most European languages (e.g., English), with the grammatical distinction past tense ~ present tense, the marked past tense specifies 'past time'3 (time that is past I with respect to the speech situation); the unmarked present tense specifies .. peither 'past time' nor 'present time', nor does it deny 'past time*, if we : take into account the full range of its usage. The present tense may be used '| in such instances as two and two are four\ I teach at Cornell University; the * rain in Spain stays mainly in the plain, etc. where both 'past time1 and 'present timeV(and 'future time') are covered by the present tense. We see ; from these examples that the characterization of the present tense as meaning either 'nonpast time' or 'present time' does not conform to tense usage. In fact, in many contexts, the present tense conveys neither the ~j presence nor the absence of x (=*past time5). So, while the marked term necessarily conveys the 'presence of the unit of information xf the ; signalization of x\ the unmarked term (if we take into account all its • usage) conveys 'the presence and/or absence and/or nonpertinencc of the unit of information x\ Since the unit of information x is in fact (he •■ marked pole of the opposition, we may say then that the Opposition is 'í between the presence of x and the nonnecessary presence (presence or absence or even nonpertinence) of x. : The fact that the marked term is more narmwly specified than the