Saussure’s reception by language reformers in Turkey

: It is surprising that the Turkish public had to wait until 1976 to read a translation of the Cours de linguistique générale [ Genel dilbilim dersleri ], an influential work on the national language reform of decades earlier. Ordered by the Turkish Language Society and published in two volumes (appearing in 1976 and 1978), the translation was followed in 1984 by a second edition with an updated vocabulary. Founded in the decade following the proclamation of the new republic, the society’s mission was to conceive and implement a language reform on a national scale. The translator, Berke Vardar (1934-1989), was an active theoretician and supporter of the language reform, and a pioneer in introducing the discipline of linguistics in Turkey. The local context of the Cours ’ publication in Turkey being closely related to this controversial but extensive language reform, my contribution will be dedicated to Saussure’s reception by language planners. Tahsin Yücel (1933-2016) made use of the arbitrary sign thesis to counter conservative objections against new words introduced by modernist language planners. Massive elimination of Ottoman words, quickly replaced by new ones – be they rediscovered “pure Turkish” or freshly “made-up” ones – provoked considerable indignation in more conservative circles. These objections were mainly targeted against linguistic intervention, seen as a destructive attack against the natural evolution of the language, besides being ultimately unproductive. Taking up the defence of the language reform, Yücel reclaimed individuals’ freedom in parole , referring to the Saussurean distinction between parole and langue . Another Saussurean theme that he adopted is the dis- tinction between synchronic and diachronic, which accounts for present-day words’ remoteness from their earlier ancestors without discrediting language change. Final- ly, by recalling the arbitrary character of the linguistic sign, he refuted the claim that old words were more genuine, or that they held a special relation to the concept that they signify. This essay examines the influence of ideas found in Ferdinand de Saussure’s Course in General Linguistics ( CGL hereafter) on Turkish language planning activists. Despite its relatively late publication in Turkey in the 1970s 1 , CGL ’s pioneering theses have been widely used to refute conservative objections to language reform even before its translation was achieved. a major role in the insti-tutionalisation of linguistics in modern Turkey as a prolific translator, teacher and, not least, an outspoken advocate and practitioner of language reform.

This essay examines the influence of ideas found in Ferdinand de Saussure's Course in General Linguistics (CGL hereafter) on Turkish language planning activists. Despite its relatively late publication in Turkey in the 1970s 1 , CGL's pioneering theses have been widely used to refute conservative objections to language reform even before its translation was achieved. Indeed, CGL was published by the Turkish Language Society [Türk Dil Kurumu (TDK hereafter)], a pivotal institution in the conception and implementation of a wide-scale language reform that started in the late 1920s. The translator, Berke Vardar (1934-1989, played a major role in the institutionalisation of linguistics in modern Turkey as a prolific translator, teacher and, not least, an outspoken advocate and practitioner of language reform. Following the proclamation of the republic in 1923, Turkish society passed through a series of cultural reforms. These included adoption of the Swiss civil code, compulsory and free primary education, dismantling of Islamic courts and schools, adoption of a new dress code (notably the interdiction of religious attire), and a comprehensive language reform. The language reform started with the official adoption of the modified Latin alphabet in 1928 as a replacement for the old Arabic-based script in all public communications. The decree was followed by nation-wide meetings where Mustafa Kemal Atatürk in person introduced the new letters to the public. The shift to the Latin script had a great symbolic value in the building of a modern nation: though the writing of Turkish with the old alphabet was impractical, the main reason behind the script reform and the subsequent lexical reform was to cut off the new Turkey's ties with its Ottoman past, in a move towards westernisation.
In 1932 the Turkish Language Analysis Society [Türk Dili Tetkik Cemiyeti (to become TDK four years later)] was established. In 1939 the Society started its lexicographic work, and published its first dictionary in 1944. TDK pursued a pioneering activity of creating neologisms as part of a wide-scale attempt at "purifying" the Turkish language of its foreign influences, returning to popular and old expressions as a resource to dig from in search of alternatives to Arabic, Persian and French loanwords. In this context, it was predictable for the language reform to encounter conservative resistance, and neologisms by TDK have not infrequently been accused of extreme artificiality, and its policy has been judged unnecessarily radical.
With the ending of the single-party regime in 1950, a change in the government resulted in TDK being deprived of its funding, until the language reform resumed following the coup of 1960. During the 1960s and 1970s an abundant literature on the language reform flourished, with important contributions from Tahsin Yücel , a novelist, semiotician, literary critic and translator who, with Vardar, introduced structuralism and functional linguistics to Turkey. In these years Vardar translated classics of modern linguistics and created a linguistic terminology for modern Turkish. Translator and life-long friend of André Martinet, Vardar introduced Antoine Meillet, Joseph Vendryes and Roland Barthes to the Turkish public. In his tribute to Vardar, Martinet 2 mentions his importance for the promotion of functional linguistics as well as his defence of Saussure against the criticism of Louis-Jean Calvet. Considering Vardar's double activity as a Saussurean linguist and a language planner, I suggest reading his translation of CGL with the Turkish language reform in the background, all the more so because the theses of CGL have been frequently used by reformers close to TDK in their polemics with language conservatives.
That being said, Saussurean linguistics presents some challenges for the theory of language planning, as demonstrated by its largely unfavourable reception by other language planning theorists (such as Charles Kay Ogden and Ivor Armstrong Richards, Otto Jespersen or Valter Tauli). Considering Saussure's explicit stance against the feasibility of language construction, his warm reception by language planning theorists in Turkey may come as a surprise. In this essay I present a mildly reformist reading of Saussurean theses as performed by Turkish linguists. Its plausibility and limits are discussed within the specific context of the Turkish language reform.

THE ARBITRARINESS OF THE LINGUISTIC SIGN AND THE RESULTING (IM)MUTABILITY OF LA LANGUE
The potential appeal of Saussure's main thesis -the arbitrary nature of the linguistic sign -from the perspective of language planning is obvious. The essentially unmotivated nature of the linguistic sign liquidates all scientific argument against language change, for there is no rational reason to prefer the old forms over the new ones. Reformists make clear that resistance to new words is truly a matter of attachment to old habits, since conservationists are unable to provide a rational criterion for what actually is a personal preference based on prejudice or tastes.
Agop Dilaçar 3 argues that difficulties of expression, use of extralinguistic gestures, and literary innovations all prove that language is a never-finished tool, always bound to be in the making. Furthermore, he stresses that the continual divergence between etymological origin and contemporary use excludes any possible natural relation between reality and language. Referring to Martinet's Elements of General Linguistics, Yücel 4 states that the relative stability of the written language lies at the origin of the erroneous belief in the stability of the way the language is spoken through generations 5 . He warns against a linguistic fetishism consisting in the belief in an essential relation between word and concept. The observed dynamism of language is used by language reform advocates in favour of free creation, to contest established use as the only possible norm for correct language. In CGL, to which Yücel refers as the founding text of modern linguistics, Saussure says that " [t]here is no reason for preferring soeur to sister, Ochs to boeuf, etc.", and that "[n]o one disputes the principle of the arbitrary nature of the sign" 6 . Only secondary motivatedness can exist in language, through consistency across word formations, such as using the same morpheme in order to convey the same semantic relationship. If within a given language we have synonyms such as Turkish ak and beyaz, both corresponding to English 'white', we can legitimately pick one of them, or even "make up" one ourselves. "Made-up" is a negative term often employed by language conservationists to attack new words, suggesting that these are false or unauthentic. In their turn, reformers reclaim it by giving it a more positive sense, stressing the creative and inventive power implied in the construction of new words.
Nevertheless, the Saussurean opposition of individual and community, one closely linked to the thesis of arbitrariness, leads to a conservative view of language and a negative attitude towards language planning. Absence of a natural link between signifier and signified, while making language vulnerable to change, ensures at the same time a contractual value to the social conventions ruling the use of the sign ("Because the sign is arbitrary, it follows no other rule than that of tradition, and because it is based on tradition, it is arbitrary" 7 ). Unlike other social institutions, as pervasive as they may be, this constitutive arbitrariness deprives society of any rational ground for discussion, rendering all planned change in language baseless and unscientific. Indeed, Jespersen 8 and Tauli 9 reacted to the radical irrationality implied by Saussurean arbitrariness by stressing that the prejudicial view of unquestioned equivalence of all language forms is a big hindrance to language planning, in that it dismisses all attempts at language change from the start by denying that it might constitute an improvement at all. Furthermore, Jespersen criticised Saussure's langue/parole distinction as a dichotomisation of community and individual. In this distinction Jespersen sees a misleading substantiation of collective phenomena, as in theories of "folk-mind". He prompts us to guard against metaphysical illusions, arguing that society is a set of individuals, not an autonomous entity itself 10 . Jespersen ranges Saussure's theory along with the naturalist view of language as an organism, though Saussure is precisely credited by other language planning advocates 11 for the overcoming of this theory. As a response, Jespersen replaces the substance model of language (language as opposed to speech) with an empirical one: language as a collection of individuals' speech, vocabulary as a sum, other aspects such as grammar and punctuation as an average. Insisting on the neglected transformative power of ordinary individuals, he draws a model of linguistic innovation in agreement with the natural dynamics of language, based on repetition and imitation 12 .
Faced with Saussure's explicit view on the uncontrollability of language, Vardar, as a language reform partisan with a deep veneration for Saussure, adds this footnote at the beginning of the chapter about the immutability of the sign: "Some of the opinions expressed by Saussure in this chapter have been refuted by studies made or disseminated later and a series of newly discovered facts. We especially want to mention innovation efforts on the word level and language planning attempts inspired by applied linguistics and sociolinguistics. It should be considered normal that Saussure who, deprived of convincing examples in his time, could not completely overcome the conception dominating linguistics in the 19th century despite his groundbreaking aspects, was mistaken on some points. Moreover, it should not be forgotten that Saussure did conceive the interactive relation of society and time; therefore, he cannot be used for arguing against attempts at changing and innovating language" 13 . Although Saussure attributed both mutability and immutability of the sign to its arbitrariness, Turkish reformers retained more of the characteristic of mutability and relatively neglected his insistence on the speakers' helplessness concerning language planning. For instance, Yücel specifies the limits of language's resistance to change. This lack of impressionability of language is only theoretical, he says, for Saussure himself states that linguistic changes have their roots in individuals 14 . Nevertheless, it should be re-10 "A mystically assumed 'common mind' (Steinthal, Wundt) really explains nothing whatever in any department of life, any more than the assumption of a mystical 'common-stomach' would serve to explain how it is that people react to foods and poisons" (Jespersen 1926, p. 15). 11 Martinet 1946Yücel 1968. 12 For Jespersen, all linguistic phenomena appear first on the level of speech (as admitted by Saussure himself). By repetition, innovative individual action becomes custom (agreement between actions of the same individual), then, by repeated imitation, the custom becomes a collective habit (agreement between customs of different individuals). Imitation, which is at the basis of all social life including language, explains the integration of an individual verbal creation into the collective texture of language. Its origin in the individual action is what makes linguistic innovation possible, in the theory that Jespersen advances against Saussure's (Jespersen 1926). 13 Vardar, in Saussure 1916Saussure [1998. 14 Yücel 1982, pp. 148-149. membered that Saussure heavily limited the possibility of language planning through both mutability and immutability of language: not only is the speakers' control over language limited by its extra-individual nature and the arbitrariness of the linguistic sign, but even a successfully achieved constructed language is doomed from the start because of the unavoidable mutability of language 15 . Saussure puts a double stress on the uncontrollability of language -as opposed to speech -owing simultaneously to mutability and immutability, both resulting from the ultimately arbitrary link between the signifier and the signified. It is therefore worth noting that, despite Saussure's open scepticism concerning language planning -notably through the example of Esperanto 16 -his theory has been a major inspiration for Turkish reformers.

ANTI-NATURALISM
The anti-naturalist conception of language was another major aspect of Saussurean linguistics adopted by Turkish language reformers. Reform advocates insisted that language as an inherited social institution exists through its use in speech by its speakers. As Martinet 17 points out, the social nature of language is incompatible with its organicist view, prevalent in some anti-reformists, as an autonomous entity that doesn't accept external modification by user intervention 18 . Though Jespersen pointed to the similarity of Saussure's view of language (as separated from speech) with this organicist conception, Turkish reformers insist on the social nature of language as defined by Saussure and reject the organicist metaphor as pre-Saussurean -therefore, pre-scientific. 15 "Whoever creates a language controls it only so long as it is not in circulation; from the moment when it fulfils its mission and becomes the property of everyone, control is lost. Take Esperanto as an example; if it succeeds, will it escape the inexorable law? Once launched, it is quite likely that Esperanto will enter upon a fully semiological life; it will be transmitted according to laws which have nothing in common with those of its logical creation, and there will be no turning backwards. A man proposing a fixed language that posterity would have to accept for what it is would be like a hen hatching a duck's egg: the language created by him would be borne along, willy-nilly, by the current that engulfs all languages" (Saussure 1916(Saussure [1959). 16 Saussure's brother René was a fervent Esperantist who defended the language against reformation proposals that eventually led to a competitor, Ido, designed by Louis de Beaufront and Louis Couturat. John Joseph (Joseph 2012) retraces René's influence on Ferdinand's view on the linguistic status of Esperanto. In the first lectures of 1907, Saussure follows his Esperantist French student Théophile Cart's view that Esperanto would undergo an evolutionary dynamic comparable to natural languages'. In the third course (1910)(1911), however, he seems to have adopted René's thesis of Esperanto's singularity which states it apart from natural languages, namely, that since Esperanto is not the native language of any of its speakers, it is relatively immune from linguistic change. 17 Martinet 1946. 18 K. Brugmann and A. Leskien's criticism of artificial languages is a major example of this organicist view in the service of language conservatism (Brugmann, Leskien 1907) (see Moret 2004 for details).
Describing language as a living being, sacred in its natural development and untouchable without compromising its authentic value, is misguided, says Yücel. Following Vardar 19 , he declares that treating language in such a way is ill-founded and unproductive. Language is not "a natural creature" but "a human creation" 20 : "If one can talk about a life for language, this life can be the life of a social institution at best. And institutions, as deep as their roots may reach, are human products" 21 . Yücel attacks the above-mentioned obsolete view of language as a divine system emerging out of the common national genius, above and beyond all persons or classes. Needless to say, for him, this romanticist view merely amounts to an excess of meaningless poetic metaphors used for propaganda.
Still other reformists criticised the naturalist conception of language. Hikmet Dizdaroğlu 22 blames the organicist conception, dating from the 19th century, for its normative ideas of growth and decay, assimilating planned language change to unnatural "deviation". Emin Özdemir 23 makes the connection between naturalism in theoretical linguistics and conservatism in applied linguistics. Indeed, a very common unscientific argument against language reform is that language, as an independent living creature, does not accept external modification. Nevertheless, Özdemir reminds us, alongside the obvious cases of successful language reforms in Israel, Hungary, Germany and Norway, the established reality of agriculture and pedagogy suffices to refute the deeper plausibility of this view; for, even though soil and children are natural organisms, they do receive human intervention. On the contrary, neglect of planned intervention causes degeneration in language through invasion of alien words, just as harmful herbs need to be eliminated from the soil for improving its quality. Likewise, Yücel places language planning in the wider framework of transformative human activity. Doing so, he breaks the opposition of natural to artificial. He describes linguistic creation as one of the countless, "natural" ways by which humans intervene in their surroundings, a praiseworthy attempt to turn "what is" into "what ought to be" 24 . A similar attitude is found in interlinguists such as Couturat 25 and Jespersen 26 , who refer to everyday examples of such creations that start out as novelty and get eventually accepted as a natural part of the ordinary life (numerous technologi- 19 Vardar 1967, quoted in Yücel 1982, p. 75. 20 Yücel 1982 Ibid. 22 Dizdaroğlu 197122 Dizdaroğlu . 23 Özdemir 1969 Yücel 1982, p. 14. 25 "C'est justement le privilège de l'homme de diriger et de corriger la nature, de la perfectionner au besoin et de la discipliner. Dans toutes les institutions et dans toutes les productions humaines, le progrès consiste à remplacer l'action spontanée par l'action réfléchie, l'instinct par la raison. Il ne faut donc pas s'en laisser imposer par le respect superstitieux de la nature, de l'évolution ou de la vie: ce n'est au fond qu'un sophisme paresseux" (Couturat 1906, pp. 25-26). 26 Jespersen 1933a. cal creations such as fire, writing, clothing etc.). As such, language planners contest the conservative idea of nature by highlighting humans' transformative activity affecting their surroundings.
Furthermore, reformists used the accusations of artificiality, directed at their neologisms, against their holders. Restricted to a tiny circle of highly educated elites, Ottoman Turkish didn't find a large field of use, therefore being condemned to stay an "artificial" language from the perspective of the language as a social tool meant to accomplish the function of effective communication. Reformers advance new criteria of naturalness, based on usability and accessibility. As opposed to the pejorative connotation that "making up" ("uydurma" in Turkish) has in the context of a hoax or a falsification, the making up of new words is redefined as a creative act that enriches the language 27 . Like Yücel and Nurullah Ataç 28 , Peyami Safa 29 insists that all words are "made up". As for legal coercion through law and schooling, advanced as a proof of unnaturalness by their opponents, reformers recall that this was equally the case with Ottoman.

VALUE-NEUTRALITY OF LINGUISTICS AS A SCIENCE
The organicist conception of language contested by language reformers has been responsible for the rejection of artificial languages as well. Martinet 30 notes that the conception of language as an autonomous natural organism was widely discredited in the 20th century, without affecting the general reception of constructed languages positively -even though the first linguists who gave serious consideration to constructed languages are the ones who highlighted the conventional nature of language as a social fact 31 . Faced with this anachronistic attitude, international auxiliary language advocates put forward the pragmatic success of Esperanto as a fact -a strategy that helps legitimating language construction in the field of linguistics 32 . The empirical evidence for the use of constructed languages is used by Martinet to justify their inclusion in the field of linguistics as a legitimate object of study, despite a "répugnance instinctive" 33 commonly found against them 34 . 27 Onat 1967. 28 Ataç, Dizdaroğlu, Özerdim (eds.) 1962, p. 240, quoted in Yücel 1982, p. 95. 29 Safa 1951. 30 Martinet 1946 "Il est intéressant de constater à cet égard que les premiers linguistes qui aient marqué leur intérêt pour la question des langues construites, sont ceux qui ont vu dans le langage un fait social, une convention imposée, et qui ont su ainsi libérer leur pensée de la conception de la langue comme un organisme vivant" (ibid., p. 38). 32 Moret 2004. 33 Martinet 1946 "Aujourd'hui l'humanité civilisée s'est plus ou moins accoutumée à la recréation dans le domaine des sciences de la nature. Mais dans tout ce qui touche à l'homme, à son corps, à son esprit surtout, si l'analyse s'est généralement imposée, la simple pensée d'une synthèse remplit encore certains esprit d'un effroi vague. On en nie tout d'abord la possibilité; puis, In the same vein, for Turkish reformists, excluding evaluation from linguistics is a means of dismissing criticism directed at language reforma widespread social movement to be impartially observed by linguists. This is a somewhat defensive position taking the language reform outside the realm of scientific activity. Indeed, unlike Tauli 35 who calls for an applied science of language planning, Yücel stresses that language reform is not a scientific activity in itself, limiting the latter to the task of external observation only -a conception criticised by Jespersen for conservatism. Following the opposition langue/parole, Yücel assigns language planning to the latter field, associated with practical concerns and free action. It is in this sense that he employs the Saussurean concept of speech as a space for linguistic innovation. For Yücel, language reformers act totally within the confines of the speech: they can, at best, use new words in their writing and hope to be followed by the wider society. Only in this way, through speech, can any innovation enter the communal space of language, at which point it emerges as a phenomenon to be described by linguists without judgment: "Yes, the institution of language may well not be individual, but speech is, and purificationists' efforts to 'create' only take place on the level of speech. Once the fruits of these efforts have appeared as a fact on the level of language, the only thing one can do as a linguist is to detect and describe them" 36 .
Yücel criticises the elitist criteria of some conservatives who set an unjustified local standard of correctness, such as language spoken in Istanbul, or other value judgments about fineness or beauty associated with sounds. For Yücel, these considerations lacking in scientific value shouldn't have their place in a serious discussion about language. Another misguided accusation of unscientificity directed at language planning is due to a faulty conception of linguistics, confused with etymology. Awareness of the Saussurean distinction between synchrony and diachrony spares one from such misguided claims of correctness in virtue of conformity to etymology or past use. Unlike the normative aspect of grammar, to which the science of language is mistakenly reduced by conservatives, linguistics relies on factual observation free from aesthetic or moral considerations. Vardar stresses that linguistics excludes all value judgments, including truth, falsehood, naturalness and artificialness 37 . In this descriptivist framework, he adopts the argument of widespread use to defend, for instance, the newly created and much contested suffix -sal/-sel 38 . lorsque les faits sont là, on les ignore, grâce à l'admirable pouvoir qu'ont les hommes d'éluder ce qui les heurte ou qui les peine" (ibid., p. 37). 35 Tauli 1968. 36 Yücel 1982, p. 95. 37 Vardar 1967, quoted in Yücel 1982 The previously nonexistent -sal/-sel suffix (making adjective from noun) was introduced during the language reform as an alternative to the Arabic -î. Words derived using -sal/-sel were judged by conservatists to be improper constructions that violate the established Turkish grammar. Following Vardar (Valdar 1967), Yücel (Yücel 1982 affirms that the widespread Responding to prescriptivist arguments about the impoverishing effect of purism (elimination of old words, referred as "liquidation" ["tasfiye" in Turkish] by its opponents), Yücel reiterates that those arguments fall short of the standards of scientificity, affirming that value judgments such as "civilised", "primitive", "rich", "poor" don't belong to linguistics 39 . Mentioning Claude Lévi-Strauss' findings about the actual civilisation of socalled primitives, he dismisses any discussion of civilisation with regard to language. On this subject, Yücel also rejects as a baseless prejudice the correlation between richness of thoughts, or civilisation, and the number of words. For the same reason, he states that subjective criteria such as a language's musicality, sound beauty, vulgarity or fineness can't be measured scientifically. For Yücel, these lie outside the boundaries of rationality, and may only be of interest for poets. Thus, contesting language reform on such grounds is defending unconscious habits, and habit is "not a value of reason" 40 . Defending old words on the basis of such criteria is erecting socially conditioned, accidental habits into absolute values. Here, Yücel joins Jespersen who exposed the lack of objective rationale behind this type of standards for correctness, though Jespersen advanced an utilitarian alternative to traditional prescriptivism, which partly corresponds to the anti-elitist stance of Turkish reformists detailed below.

ANTI-ELITISM
The claim of linguistic neutrality presents an interesting contrast with Jespersen's and Tauli's open advocacy of evaluation in language considering its man-made nature, and their unapologetic anthropocentrism inferred from it. For them, modern linguistics' dogmatic reluctance toward evaluation (the view that no language form is superior to another) has a paralysing effect on language planning, depriving it of a rational ground that could form the basis of a constructive effort. Against Saussure, they draw the need for language planning from their anti-organicist stance, in their view of language as a tool for communicative needs that can and should be improved accordingly 41 .
The ground needed to negotiate changes in language will be provided in the Turkish context by an appeal to democratisation, a planned evolution toward public accessibility and the formation of a new national identity purified from Eastern-Islamic influences. Though not non-existent, pureuse of these words in the contemporary practice invalidates all such arguments. In arguing this, Yücel refers to the distinction between grammar (normative) and linguistics (descriptive). 39 Yücel 1968, p. 73;1982, p. 86. 40 Yücel 1982 "Against the anachronistic views of purists who will 'save' languages from 'corruption' and those of some modern linguists to whom one expression is as good as another, one must spread the knowledge that it is efficiency that is relevant in language and that man is free to alter and improve language according to his will. Sticking to tradition and desire for stability is a natural human trait, but so is desire to have a better tool" (Tauli 1974, p. 61). ly linguistic reasons advanced in favour of a language reform occupy a minor place overall. These concern advocacy for more transparent word formation, generalised on consistent principles, such as neologisms made by adding suffixes to existing words. For Saussure, such derivations are essential to introduce some degree of motivation in language, despite the ultimately arbitrary character of the lexical units. "In fact, the whole system of language is based on the irrational principle of the arbitrariness of the sign, which would lead to the worst sort of complication if applied without restriction. But the mind contrives to introduce a principle of order and regularity into certain parts of the mass of signs, and this is the role of relative motivation" 42 . Thus, unlike pomme and cerise, pommier and cerisier are not totally arbitrary (Saussure calls this "secondary motivation"). Increasing the secondary motivation was a somewhat important case for language reform from a democratic perspective. In virtue of this rationality, it was argued that uçak 'plane' was a better replacement for the old word tayyare due to its clear derivation from uçmak 'to fly'. Another example used to highlight the pedagogical benefits of language reform is the new word for 'triangle', üçgen, coined by Atatürk in the geometry booklet he wrote in an attempt to offer modernised alternatives to the old terminology. Derived from the number üç 'three' with the new suffix -gen used in similar constructions for geometrical terms (beşgen 'pentagon', altıgen 'hexagon', and so on), the new term helps the student to grasp elementary notions of geometry without departing from their everyday language. In contrast, the old word müselles will not be understood as easily, and needs to be learned separately. In such cases, secondary motivation plays a democratising role by making the meaning transparent, accessible to the lay public -an especially important task considering the high level of illiteracy among the general population at the time. Following the philosopher Macit Gökberk, Yücel presents the language reform as an Enlightenment project in continuity with the emancipation of national languages in 18th century Europe, replacing Latin as the default language of science. It is important to understand that this social motivation was at the root of Turkish language reform. In contrast, purely linguistic arguments were more actively used for the advocacy of the script reform, on the grounds that some Ottoman letters didn't have corresponding sounds in Turkish. The new Latin alphabet was designed to be a phonetic one, eliminating ambiguities in the Ottoman script arising from the dissonance between the Arabic and Turkish phonologies 43 .
As mentioned above, Yücel does not stop at denying the artificiality of new words by recalling that all words are man-made. He extends this characterisation of artificiality to the old language itself, claiming that Ottoman was much more artificial than its modern day counterpart. Indeed, even though speech is free (unlike language), too big a shift between established use and individual creation leads to incomprehension. To return to the objection against a made-up language, Yücel points at unnatural formations dominating the Ottoman language itself, such as badly translated loanwords from Arabic or Persian, or pseudo-Arabic constructions that violate the structure of both languages. To reverse this situation, language reformers explored popular language spoken outside the court and confined within the limits of oral transmission. The first two decades of the TDK were mainly dedicated to this lexicographic task of collecting and recovering popular expressions through extensive fieldwork. Some words already existing in the spoken language of peasants have been adopted into the official language, and some of this lexical material provided the basis of new creations, alongside older Turkic vocabulary that has been revived as an alternative to Arabic and Persian loanwords.
Thus, after rejecting such values altogether within a linguistic inquiry, Yücel redefines naturalness in language pragmatically, connecting it to popular use and clarity. He discredits Ottoman precisely for its extreme disconnection from people's spoken language, calling it an artificial language, or even a non-language. Furthermore, Yücel denounces the elitism of Ottoman as a "means of exploitation", a "mark of privilege" cherished by those who care for their class privilege. Starting from the 19th century modernisation movement Tanzimat, Ottoman appears as an obstruction to democratic understanding of decrees and laws, and the new journalism aspires at reaching a wider audience through democratisation of language. Attempts at purifying the written language with a view to popularising literary creation beyond a small circle of elites were made by late 19th and early 20th century avant-garde writers 44 . After the pre-republican Yeni Lisan ("new language") movement starting in 1911, the avant-garde poetry circle Garip ("strange") continued this line with their effort to bring the poetic language closer to everyday language, rebelling against aristocratic standards and old-fashioned formalism 45 . These poets were promoted by Ataç (mentioned above), an essayist and fervent spokesperson for language reform famous for his numerous neologisms well ahead of his time.

CONCLUSION
An examination of the most commonly advanced arguments against language conservationists suggests that Saussurean theses ensured a defence against reactionary attitudes regarding language reform in Turkey. They were mostly used as negative arguments to dismiss criticism of language reform by referring to the modern science of linguistics. Through these epi-stemological statements, language planning advocates contest the authority of old established forms in language. Fetishism of old forms is criticised notably through the universal academic recognition of arbitrariness as the major characteristic of the linguistic sign. Moreover, in its Turkish reading, the Saussurean definition of language as a realm separate from speech provides a legitimate ground for free creation without violating standards of scientificity. In this respect, Turkish language reform demarcates itself from the anti-Saussureanism of interlinguists, who integrated their planning activity in their philosophy of linguistics. In contrast, Turkish language planners put forward practical arguments for modernisation and nation-building to advance their cause, reserving Saussure's theses mainly to refute conservative opposition to their activity. Nevertheless, some positive statements by Yücel and others about the artificiality and inappropriateness of Ottoman may be read as a defence of linguistic democracy for scientific reasons, on the grounds that language is a man-made institution to serve in interpersonal communication, so that increased intelligibility means increased efficiency. Therefore, where the arbitrariness of the sign, anti-naturalism and anti-normativity of Saussurean linguistics fail at justifying language reform other than as polemical elements, the concept of secondary motivation serves positively the cause of language planning. In the light of these observations, Saussure appears as a central authority figure of modern linguistics against opponents of language reform, although his theses potentially fit both conservative and reformist readings, and were indeed contested by other renowned language planners. © Başak Aray