- Social Sciences
- African Studies
- American Studies
- Asian Studies
- Communication Sciences
- Ethnic Studies
- European Studies
- Gender Studies
- Physical Sciences
- Life Sciences
- Animal Communications
- Cell Biology
- Evolutionary Biology
- Food Science and Technology
- Human Anatomy
Popular Derivation and Linguistic Inquiry: Les Javanais
by Barbara E. Bullock
Popular Derivation and Linguistic Inquiry: Les Javanais
Barbara E. Bullock
The French Review
Updated: January 9th, 2013
Povular Derivation an& Linguistic Inquiry: Les Javanais
by Barbara E. Bullock
THE TERM ARGOT, like other words which define linguistic varieties (lan- guage, patois, dialect, jargon) has no precise referent. Therefore, there is often some debate about what forms of linguistic behavior actually con- stitute argot. There are some researchers who follow Sainean in inter- preting it only in its strictest sense as referring to l'ancien argot, the secret languages of the Coquillards and other bands of criminals who roamed France from the thirteenth through the nineteenth century. The present day use of the term is more liberal as evidenced by the Bibliographie des argots fran~ais published by Le Centre DrArgotologie of the Sorbome in which works devoted to borrowings, specialized or technical vocabular- ies, and "street French" are found along with the more conventionally accepted forms of argot. Despite the lack of consensus concerning its denotation, most linguists recognize two general forms of argot. There is one type in which the meaning of a word is masked either because the word is coined for a specific purpose or because its original meaning has been replaced; this is the lexical form. The words pic, a coinage, and poulet, a shift in meaning, both which refer to an "agent of the police", are examples of lexical argot.' The second type masks the form of a word through affixation or through the displacement of sounds and syllables within a
This article follows Plenat in using Delvau's (1865) application of the term javanais to these kinds of secret languages in order to avoid the am- biguity and controversy of the term argot (Plenat "Presentation" 5). Ver- lan is one contemporary form of javanais in which, for example, the slang term poulet becomes lkpou (i.e., pou.let > le(t).pou >lkpou) through an inversion of syllables. The other types of structure changing argot consid- ered in this paper are infixing javanais, where the sound sequence [-av] is inserted within a syllable (poulet > pavoulavet), and largonji des louchkbtms a combination of affixation and inversion (poulet > louletptm). These forms of argot have recently attracted the attention of phonologists and this article addresses the issue of what such popular forms of deriva- tion lend to our understanding of language structure.
A great deal of research in the late nineteenth and early twentieth cen- tury was devoted to characterizing argot, particularly lexical argot, in a scientific manner.3 Most linguists point to the publication of Vidocq's Les Voleurs in 1837 as the main source of inspiration for this interest in argot. Vidocq, a picaresque character with a criminal past who eventually be- came chef de la Sure^ti, had himself been a convict and his book contains a lexicon of around 1,500 argot terms in use in the prisons of contemporary nineteenth-century France, a vocabulary that, considering the author's personal experiences, was considered to be authentic. Victor Hugo, Eugene Sue, and Balzac all borrowed liberally from Vidocq's glossary and Balzac's character of Vautrin (La DerniPre Incarnation de Vautrin) is surely modelled, in part, on Vidocq (Citron 16).
While Vidocq's lexicon provided an esoteric resource that appealed to the authors of the romantic period and allowed them to capture the spirit of l'homme criminel (Guiraud 116-117), it also supplied a corpus that seemed designed for the then-current historical and taxonomic approach to linguistics since, among his original entries, Vidocq included words of ancien argot for which he claimed to know the meaning (Sainkan 97). This attracted the interest of linguists who were preoccupied with etymology and with defining the characteristics of argot as precisely (i.e., as "scien- tifically") as possible. Sainean, in particular, scrutinized Vidocq's corpus and discarded anything that he felt did not properly constitute the argot of the time. For Sainean and others, true argot had to be secret, parasitic (in that it lacked a syntax and morphology), and artifi~ial.~
These were the criteria that distinguished argot from either patois, which has its own grammar, or the bas-langage of Paris which was neither secret nor artifi- cially derived.
It was the requirement of artificiality, however, with which later argotologues took issue. Various documents, such as the transcript of the trial of the Coquillards of 1455 and the Jargon de l'argot rgormi (1628), attest to the existence of individuals within criminal communities who were respon- sible for inventing and/or teaching argot to candidate bandits. Sainean recounts that the trial of the Chauffeurs d'orgtres in 1800 revealed the existence of:
un Instituteur des Mioches, qui etudiait avec soin le caractere de ces futurs bandits, pour les 6duquer et les rendre propres au genre de vie qu'ils devaient mener. I1 s'appliquait surtout a leur apprendre a parler l'argot, a jouer adroitement du biiton, a frapper un hornme de maniere 2 le tuer sur-le-champ, etc .(86)
While Sainean subscribed to the idea that such brigand "Acade'mies" were the source for the creation and spread of argot, Esnault (199-201), Dauzat (19-20) and others believed the notion to be entirely fictional and reflective only of the seventeenth century pre'cieuse belief that anything linguistic must be properly sanctioned. In their view, words in argot are derived in exactly the same ways as they are in French. As summed up by Guiraud the dividing line between the two is then not so much linguistic as it is social: "En un mot, l'argot-l'argot en tant que langage secret-n'est pas artificiel dans ses modes de creation lexicale, mais il l'est dans son emploi" (26).
It is this relationship between popular derivation and word formation in the standard language that has placed argot, in general, and javanais, in particular once again within the scope of inquiry of contemporary lin- guistics. As long as argot was held to be artificially derived and lacking in its own internal rules of organization, argotologues were content to debate its origins and its purity, and to catalogue its attested forms. Javanais, on the other hand, often escaped the notice of lexicographers and linguists precisely because it is a process of derivation, rather than one of lexical- ization. Javanais can be created spontaneously by native speakers as long as they are aware of the rules of its formation. Thus, aside from a few javanais terms that are so frequently used that they become lexicalized, this type of argot rarely figures in specialized di~tionaries.~
The spontaneity of javanais is precisely what interests linguists today. The fact that native speakers can internalize the rules necessary to pro- duce words in verlan, javanais, or largonji on the spot is, in itself, an example of linguistic creativity in the Chomskian (i.e., generative) sense. Although Chomsky relegates the linguistic behavior of native speakers to the performance aspect of language and focuses instead on underlying linguistic competence, it has become more common lately in the sub- fields of morphology and phonology to view language games and secret languages as an important source of corpus-external evidence.
Corpus-external evidence from such sources as javanais, language acquistion data, poetry, and aphasia studies focuses on the linguistic behavior of native speakers. Arguments based on corpus-external evi- dence are not restricted to an examination of the basic structural proper- ties of a transcribed utterance rather they appeal to a native speaker's knowledge of language and assume that this knowledge is constructed in a way that mirrors the internal structure of a grammar. Corpus-internal evidence consists only of a body of linguistic data from a particular speech community and eliminates any factors that are outside of the sys- tem that linguists are attempting to de~cribe.~
This type of evidence refers only to the phonological behavior of morphemes with respect to phonetic information contained within the representation of a lexeme. The value of corpus-external evidence is that it can bolster claims made about such representations by providing researchers with a window on what a native speaker's knowledge of language (i.e., competence) must be like. Javanais, in particular, is interesting to linguists because it generally involves the insertion or rearrangement of sound segments and syllables. These mutations can lead to structures that violate the constraints of the standard language and these violations are often repaired in ways that indicate a speaker's awareness of the underlying rules of well-formed- ness in the grammar.
However, until recently, most analyses in phonological theory have relied heavily, if not exclusively, on internal evidence (Kenstowicz and Kisseberth 139). But, as the theoretical tools of phonology have been refined, the domain of its application has been extended to include mor- phological processes as well as phonological ones (Kenstowicz ch. 11). Sincejavanais deconstructs and reconstructs the morphology of French in systematic ways, it is a good "proving-ground" for various proposals concerning the organization of the language (Plenat "Pr6sentation" 5). Chief among current researchers of the internal structure of various javanais are Natalie Lefkowitz, and Vivienne Mela on verlan, Marc Plenat on infixing javanais, Franqoise Mandelbaum-Reiner, Franqoise Robert- L'Argenton, and Marc Plenat on largonji des louche'b2rns. Their work on secret language is largely dedicated to discerning its organizing princi- ples through the application of contemporary phonological analysis.
The internal structure of javanais continues to be an object of inquiry because the ways in which the base form of a word is masked provides insights about the phonological component of French. This motivation harkens back to Jakobson's oft-cited observation that the metrics of a lan- guage's poetry tend to be constructed from the distinctive prosodic or phonological categories of that language. Thus, for instance, prosodic tonality fails to play a role in French verse because tones are not a distinc- tive property of French as they are in Chinese, a language with con- trastive tones and tonal poetry. By extension, we should expect that the processes through which the various javanaisare created are the same as those which derive words in language from which the slang derives. This means that they should obey the constraints imposed on lexical creation by the morphophonological component of the grammar. The question is, do they?
Bauche (ch. 3) and Guiraud (28; ch. 3) explicitly state that the deforma- tion of words common in javanais is indeed identical to that found in the standard language. On this issue, Bauche declares:
Elle n'a rien de mysterieux, ni m@me d'artificiel comrne on le croit encore generalement: tous les procedes de deformation argotique, sauf une ou deux exceptions contemporaines, se rattachent, en derniere analyse, a des phenomenes normaux du langage que l'argot n'a fait que develop- per et hypertrophier . . . il n'est rien d'inconnu aux langues genkrales, pas m&me la metathese dont le point de depart est un lapsus linguae. Seules l'anagramme et les deformations visuelles des mots, d'un empoi recent et tres restreint, sont vraiment artificielles. (90)
The close and detailed scrutiny of javanais data by current argotologues reveals that this issue is not so straightforward as Bauche would have us believe. Writing on verlan, Mela asks a similar question to that posed above and she gives a very tentative response:
Est-ce que le verlan possede ses structures phonologiques et grammati- cales propres aussi bien que son lexique? On est tent6 de repondre bien s~rque non; pourtant on trouve, a tous les niveaux, des differences importantes qui conferent au verlan son etrangete et augmentent son attrait. ("Le verlan" 85)
According to her corpus of data, Mela finds that verlan does not always conform in expected ways to the phonological and morphological con- straints of French. If verlan is to be considered as corpus-external evidence on the underlying structure of French phonology (Lefkowitz "Verlan") then Mela's musing on the "strangeness" of its organization is not trivial but becomes, instead, an issue of theoretical importance. Essentially, lin- guists at the end of the twentieth century are debating the same issue as those at the beginning of the century-the artificiality of argot.
In order to better understand the essence of the problem, we need to examine various types of javanais and consider how they conform to and deviate from the morphophonological structure of French. Each secret language discussed here uses the syllable as the primary phonological unit of organization. A basic syllable structure representation for French (Mela 75) consists of an onset, generally a single consonant or a conso- nant plus a liquid, and a rhyme that obligatorily contains a nuclear vowel. Aside from the nucleus, the rhyme may contain a coda to close the syllable:
Thus, there is a primary division between onset and rhyme and a rhyme internal division between nucleus and coda. The javanais forms of the word mec are given below and the rules for these languages are dis- cussed in the following sections.
Verlan: keum [kcem]
Infixing Javanais: mavec [mavek]
Largonji des louch&b&ms: lecme' [lekme]
Among the javanais to be considered, verlan is the most familiar today. Certain verlanized terms have become so familiar that they are now
entered into standard and bilingual dictionaries: beur 'arabe', meuf 'femme', ripou 'pourri'. The task of the speaker of verlan is to syllabify a word and then to invert the order of the syllables. This, of course, works best with words of two syllables but it can be applied to polysyllabic words as well with sometimes competing results (e.g.,encule' + [le%ky],
Monosyllables are treated in different ways depending on whether they are open or closed syllables (Mkla "Le verlan" 77-79). A closed mono- syllable is treated as if it actually terminates in an e-muet and is, thus, underlying bisyllabic: femme = [fama] +[fa.ma] +[ma.fa]. The inverted form is then subject to final vowel deletion and the schwa is adjusted to [ce] since schwa is never found in a closed syllable: [mcef]. With open monosyllabic words, speakers detach the onset consonant of the syllable and displace it to the coda (post-vowel) position: pue = [py]] + [yp].' Occasionally, if a verlanized form becomes too wide spread and recogruz- able, it undergoes the process once again; in the words of Lefkowitz ("Le Verlan" 319) and Mela ("Le verlan" 84), it is "reverlanized": beur = [bce.ra] +[rer.ber].
Verlan, then, seems to respect the syllable structure of French and its internal constituents, the onset, the nucleus, and the coda. We can pre- sume from this evidence that speakers are implicitly aware of prosodic structure since they appear to manipulate it systematically in this game. On the other hand, Mela ("Le verlan" 75-76) demonstrates that the syl- labification strategies employed by verlanistes is not always that of French. Specifically, in polysyllabic forms, the only consonants that can fall into the coda of a nonfinal syllable of the base form are the liquids, [r] and [l]. Thus, while a phonological analysis of the common language would syllabify bifton as bif.ton, a verlan user must syllabify the internal cluster as an onset: bitton +ftonbi.
The resulting verlan form, ftonbi, violates the phonological constraints of French in that syllables and words do not begin with clusters of a frica- tive plus an obstruent (with the exception of word initial clusters begin- ning with [s]). Similarly, it is rare to find such sounds and segments as
[A] and [ks] in word initial position outside of verlan. Thus, with respect to the natural language, the "rules" of the game are artificially contrived. Another area in which the facts of verlan and those of the common lan- guage collide concerns the syllabication of glides.
In phonological analyses of French, glides sometimes act as part of the onset and are, thus, considered to have consonantal properties. For in- stance, a glide can pattern like a consonant with respect to a rule of pro- gressive voicing assimilation in French in which a voiceless consonant /p/, /t/, /k/, provokes the devoicing of the following vowel or liquid: clef [k,le], toi [@a]. Such devoicing does not impact a glide that clearly falls into the nucleus: trois [t~wa].~
Considered in this way, we can say that there is a single value for voicing across the onset. In verlan, how- ever, glides always pattern as part of the nucleus; thus, the inverted form of toi is [wat].
Aside from inconsistency with respect to the sequencing of sound seg- ments, verlan also deviates from the standard phonology in that it in- creases the frequency of the mid round vowels, [ce] and [a] due to the number of verlanized monosyllables: pPre [rep], cher [reg], juif [fcei],flic [kcef]. Finally, as demonstrated by (Mela "Le verlan") phonological sandhi rules (those that occur across word boundaries) like vowel dele- tion or liaison can be suspended, leading to an increase in vowel hiatus or vowels separated by glottal stops: les seins [lees], le cul [layk], ca pue
Infixing javanais employs a parasitic sequence [av] that is inserted between the onset and the rhyme of the syllable. Some forms of this game, like the one that appears in Queneau's Exercices de style do not have a fixed parasitic morpheme rather they copy, or reduplicate, the vowel of the base. Thus, in Queneau, midi becomes mividinrather than mavidavi (123).
Although infixing secret languages are not as popular as they were in the later nineteenth century, they are still current and Plenat was able to interview an expert practitioner of the [av] form ("Le javanais"). His study reveals, among other things, a great deal of inconsistencies in his participant's use of the language game such as the occasional addition or deletion of syllables, the repetition of certains sounds, and simple lapses in the corpus. This is the risk of performance data. By and large, how- ever, the infix is simply inserted after an onset. Interestingly enough, it is with respect to the patterning of glides that this rule becomes fuzzy.9 A representative sample from Plenat ("Le javanais" 98-101) is given below.
Notice that the same diphthongs is treated in two different ways. That is, in fois  and point  the diphthong [wa] stays together as a nucleus while in poignet [Ill andpointe , the [w] portion of the diphthong is treated as part of the onset. Also, the [j] in derridre  and in variable  syllabifies as part of the onset, but in atelier and in bien , the diph- thong is vocalic. On the subject of these diphthongs, Plenat remarks:
I1 faut donc admettre ou bien qu'un m6me groupe peut etre analyse de deux facons differentes, ou bien que la place de l'infixe n'est pas entiere- ment determinee. La premiere solution n'est pas absolument inenvisage- able, mais, faute d'autres arguments la confirmant, on retiendra ici la seconde. (106)
Essentially, Plenat takes the position that is better to put the rules of the language game in doubt than it is to question the organization of phono- logical structure on the basis of such data. In other words, it is not entirely clear what infixing javanais can tell us about underlying phonog- ical structures. Thus, its use as corpus-external evidence is limited.
Largonji des Louche'bdms
Largonji des Iouche'bdms (jargon des bouchers) is a secret language of butchers that operates by a combination of inversion and parasitic affixa- tion. It is a fairly celebrated form of javanais among argotologues ever since Larchey published an appendix of largonji in his revised dictionary (1899). The terms linspre' 'prince' (Vidocq "Les voleurs") and lorcefe' 'La Force, prison' (Vidocq "Memoires") predate Larchey's dictionary in publi- cation and indicate that a form of this language game was in use for a good part of the nineteenth century.'" Although it is often said to have disappeared, Mandelbaum-Reiner ("Secrets") conducted an interview on largonji in 1987 with a 34 year old butcher and she found that it is still in use. The familiar term loufoque you' is a lexicalised form of largonji.
The simplest way of explaining the system of coding in largonji is to say that it is a processes of suffixation and substitution. There is a variety of parasitic suffixes from which to choose: -lem, Ids, -lope, lik, le', luche. How one choses a particular suffix is unclear since the function of the suffix is only to disguise and to indicate the slang nature of a word; in this respect, the suffixes are semantically indistinct.
The [l] onset of the suffix is detached before affixation and exchanged with the initial onset of the base, unless that onset happens to be [I] as well.
|suffix||[l]6+[llec + [m]e= lecmC|
Onset clusters behave exactly like single consonant onsets: tronche + lonchetre'. Plenat ("Morphologie" 82) examines Larchey's corpus and
demonstrates that for vowel initial forms in the nineteenth-century form of the game, the vowel is skipped over and the [l] is interchanged with the initial consonantal onset encountered: ardeur +arleurdk. This example is particularly pertinent since the [r] coda consonant of the base word (ar.deur) is skipped over, indicating that it is not merely the first conso- nant that is replaced, it is crucially the first onset.
Mandelbaum-Reiner (3940), on the other hand, finds that her speakers tend to retain the initial [I] in the output as an indicator of largonji: huit + luitk, kcouter +leloutkgu2rn. Nonetheless, we retain the insight that it is the first onset of the word that is replaced. Glides tend to be treated as part of the nucleus rather than the onset just as they are in verlan: voir + lwarvtm, viande +liandvt.
Largonji, while highly morphologically based, seems to be particularly sensitive to morphology in another way. That is, in Larchey's corpus, prefixes generally fall outside of the domain of the rules of the language: entresol +entrelolsoque, refaire +relairefem (Pl6nat "La morphologie" Bl)." Actually, even sequences that are homophonous with prefixes can be treated in this way: redingote + relingoted2m. Interestingly enough, this dovetails with corpus-internal analyses of the phonology of prefixation in French. That is, prefixes and initial sequences that are simply homo- phones of prefixes are generally treated as if they are separate phonologi- cal words unlike suffixes that integrate fully with the stem (Hannahs).
What, then, can these secret languages actually lend to our knowledge of the structure of French? From this overview of the rules of three types of javanais, we have somewhat conflicting information. First, the rules internal to verlan sometimes create output sequences that are prohibited in the phonology of French. In particular, sequences are syllabified as onsets that are not normally allowed and sound segments occur word initially in violation of the phonotactics of the language. While these fac- tors apparently contradict the constraints of French phonology proper, they could well be significant in that it is the coda portion of the syllable that is eliminated and promoted to onset position; it is not usually the case that the onset is demoted to coda position except in open syllable verlanization.
In a broad way, all these languages do confirm that there is a primary division between the onset of the syllable and the syllable rhyme. In verlan, open monosyllables simply invert the order of these constituents, in infixingjavanais, the -av morpheme tends to separate these constituents, and in largonji, the onset clearly has a special status. The evidence con- cerning the status of diphthongs in these secret languages, however, is not so clear. While the glide portion of the diphthong generally falls into the nucleus inverlan, it appears to be treated somewhat sporadically as part of the onset or part of the nucleus in infixing javanais. Thus, taken together, these games do not tell us very much about the phonological status of diphthongs in French.
One factor so far left out of this discussion is the playful nature of these javanais. Perhaps the criterion of secrecy so important to the criminals of previous centuries is not as critical today since, clearly, verlan is not a secret these days. It may be that it is less important for javanais users to disguise language than it is for them to show a certain linguistic virtuos- ity. In this respect, we could never expect the rules of javanais to be homogenous nor to follow the constraints of French since one object of this linguistic behavior must surely be to sound conspicuously different from the standard language.
The argot of today is certainly not so artificial that it requires an "insti- tuteur des Mioches" to create and disseminate words in a secret lan- guage. Its creation is built on the basic structures of French prosody. But, as part of its function is to play with and to transform that architecture, we should assume that its formation is sometimes contrived and that its patterns of construction are often in flux. Nevertheless, it continues to provide a rich external corpus for the testing of linguistic principles and it continues, as well, to intrigue a new generation of argotologues.
'An anonymous reviewer correctly points out to me that Guiraud (Structures ety- mologiques du lexique franqais) proposes that the popular usage of poulet in this context is derivable from its meaning of "horse", not "chicken".
2Such language games are not restricted to French, of course. Other examples include jeringosa in Puerto Rican Spanish (Muller 1993) in which the syllable -chi-is appended to each syllable onset (e.g., dedo +chidechido "finger"), jagon, an infixing process in Haitian Creole, and Pig-Latin in English.
?A bibliography of works on argot from this time period would be extensive. Yves-Plessis covers works up to the twentieth century. Early twentieth-century works include Bruant and de Bergy, Dauzat, Delvau, France, Larchey, Niceforo, and Sain6an.
'This attitude of purism concerning argot continues today as discussed by Goudaillier and Mandelbaum-Reiner.
5A few examples of lexicalized javanais include loufoque, the largongi form of fou, ripou, the verlan form for pourri. It is perhaps at this stage that the interaction between the "natural" rules of the standard language and the rules of secret languages intersect since a word like loufoque, for instance, is subject to the same processes of derivation as any other noun in French (e.g., loufoquerie). This is true, as well, of verlan.
6Corpus-internal evidence can be culled from an abstract speech community such as in the vastly conceived and often idealized franfais standard or the similarly broad and also idealized langue populaire.
71n my opinion, the open monosyllables can be treated in the same fashion as the closed in that the onset consonant of the base does not necessarily become a coda, rather it can become the onset of a "headless" syllable (one that has no nucleus): pue = [py]+ [y.p(a)]. In this way, we respect the notion that onsets always remain onsets in permutations.
'The constraints on syllabification in French only allow two consonants in the onset; thus, anything following a cluster in a syllable, including glides, must fall into the nucleus. 'The numbers refer to the lines of P16natfs corpus in which the items appear. This is not
an exhaustive list of the occurrence of diphthongs in the data but it is, I believe, a represen- tative one. I0Guiraud (68) claims that the first attestation of largonji comes from a document concern- ing the slang of the Brest prison dated 1821-lombem 'bon'. "In these forms, I have standardized the orthography to be consistent with preceding examples.
Bauche, Henri. Le Langage populaire: grammaire, syntaxe et dictionnaire dufranfais tel qu'on le parle duns le peuple de Paris. Paris: Payot, 1920. Bibliographie des argotsfranfais. vol.1 Paris: Le Centre d'argotologie de I'UER de Linguistique, Paris V-Sorbonne, 1989. Bruant, Aristide et Leon de Bergy. L'Argot au XXe siecle,dictionnairefranfais-argot. Paris,
1901. Chomsky, Noam. Syntactic Structures. The Hague: Mouton, 1957. Citron, Pierre. Preface. Le Pere Goriot. By Honor6 de Balzac. Paris: Gamier-Flammarion,
Dauzat, Albert. "Notes argotique". Romania (1922): 403-18. . Les Argots et les mitiersfranco-provenfaux. Paris: Champion, 1917. . Les Argots. Caract?res, holution, influences. Paris: Delagrave, 1929.
Delvau, A. Dictionnaire de la langue verte. Argots parisiens comparis. Paris: E. Dentu, 1867.
Esnault, Gaston. Dictionnaire historique des argotsfranfais. Paris: Larousse, 1965.
France, Henri. Dictionnaire de la langue verte. Paris: Librairie du ProgrPs, 1898-1910.
Goudaillier, Jean-Pierre. "Argotolbtrie et argotophobie." Languefranfaise 90 (1991): 10-12.
Guiraud, Pierre. L'Argot. Paris: PUF [Que sais-je? 7001,1956.
. Structures itymologiques du lexiquefranfais. Paris: Larousse, 1967. Hannahs, Steven J. Prosodic Structure and French Morphophonology. Diss. University of Delaware, 1991. Larchey, Loredan. Dictionnaire historique d'n~got et des excentricitis du langague, huiti2me idi- tion augmentie d'un suppliment, Paris: E 3entu, 1881. . Nouveau suppliment du dictionnaire d'argot avec le vocabulaire des chauffeurs de l'an
VIII et le ripertoire du largongi, Paris: E. Dentu, 1889. Kenstowicz, Michael. Phonology in Generative Grammar. New York: Basil Blackwell, 1994. Kenstowicz, Michael, and Charles Kisseberth. Generative Phonology. New York: Academic
Press, 1979. Lefkowitz, Natalie J. Talking Backwards and Looking Forwards: The French Language Game Verlan. Tiibingen: Gunter Narr, 1991. . "Verlan: Talking Backwards in French." French Review 63.2 (1989): 312-22. Mandelbaum-Reiner, Franqoise. "Secrets de bouchers et largonji actuel des 1ouchkbPm." Langage et Sociiti 56 (1991): 2149.
Mandelbaum-Reiner, Franqoise, and Franqoise Robert-L'Argenton. "Larlepem largomuche du louchebem (premier rapport de recherche)." Documents de travail du Centre d'argotolo- gie 7. Paris: Universite Rene Descartes, 1988: 48-66.
. "Larlepem largomuche du louchkbem (deuxiPme rapport de recherche)". Documents de travail du Centre d'argotologie 8. Paris: Universite Renk Descartes, Paris, 1988: 54-72.
Mela, V. "Parler verlan: rPgles et usages". Langage et sociiti45 (1988): 47-72.
. "Le verlan ou le langage du miroir". Langages 101 (1991): 73-94. Muller, Jennifer. Syllable structure in Spanish. unpub. ms., Penn State University, 1993. Niceforo, A.La Ginie de l'argot. Paris: Mercure de France, 1912. Plenat, M. "Morphologie d'un langage secret: le javanis de Queneau". Cahiers de Grammaire
6 (1983): 152-94.
. "Lotulenoque sure la Lorphologiemique du loucherbem de Layrondmuche Leneauqui.~."Cahiers de Grammaire 8 (1984): 173-202. . "Morphologie du Largonji des loucherbems". Langages 78, (1985): 73-122. . "Presentation des javanais", Langages 101, (1991): 5-10. . "Le javanais: concurrence et haplologie", Languages 101, (1991): 95-117. Queneau, Raymond. Exercices de style, Paris: Gallimard, 1947. SainCan, Lazare. Les Sources de l'argot ancien. 2 vol., Paris: Champion, 1912. . Le Langage parisien du XIF si2cIe. Paris: de Boccard, 1912. Vidocq, Eugi.ne Francois. Me'moires de Vidocq: Chefde la police de Surtte' et jusqu'en 2827 aujourd'hui propriitaire etfabricant de papiers li Saint Mande'. Paris: Tenon, 182s29. . Les Voleurs. Paris: Chez l'auteur, 1837 [?I. Yves-Plessis, R. Bibliographic raisonne'e de l'argot et de la langue verte en France du XVe au XXe si2cle. Paris: H. Daragon, 1901.