EDP Sciences logo
Search
Menu
Home All issues Volume 41 / No 2 (2019) Histoire Epistémologie Langage, 41 2 (2019) 67-81 Full HTML
Free Access
Issue
Histoire Epistémologie Langage
Volume 41, Number 2, 2019
Prescriptions en langue
Page(s) 67 - 81
DOI https://doi.org/10.1051/hel/2019017
Published online 28 January 2020

© SHESL, 2020

For roughly the first 2,000 years of grammatical thought, prescription was most often the explicit purpose of grammatical writing: how to write and speak well, in the consecrated phrase.1 More recently, particularly in the last 200 years, prescription is what linguists do not do, and what linguistics is not interested in. Linguists observe, they do not prescribe.2 Linguistics is a science that studies language as a part of nature; prescription is considered not natural. In this article we examine the nature of prescription and its place in a science of linguistics.

We seek neither to defend, nor to reject, prescriptive linguistic behavior, but rather to understand it. To do this, we make the broadest definition of prescription: Prescription is any intervention in how another person speaks. This may be an individual correction, as a mother to a child, or a societal one, as a religious taboo on uttering a specific word, or an official institutional one, as a government to its citizens. In this breadth of coverage, we consider a much broader range of phenomena than, for example, Thomas (1991) in Linguistic purism, who limits himself to attitudes and practices regarding foreign loanwords. With Thomas’ focus, prescription is considered primarily a by-product of nationalism, as nation-states try to define their uniqueness in terms of language, creating a sense of solidarity among citizens, a sense of distinctness from neighboring countries, and a sense of pride in the prestige of the national language. For example, Wexler describes the attack on borrowings from Russian and Polish in Belorussian, and the lament of one writer that:

We have forgotten our language, its old strength and originality, its smooth, melodious form […] [the language] gradually borrowed […] and became spoiled […] In many places now Belorussians speak a terrible mixture of Belorussian-Polish-Russian. (Wexler 1974, p. 211)

Defining a nation is accomplished by defining what is and is not part of a language, delimiting the linguistic borders just as political borders are reshaped. The equation of the nation and the state inserts a racial element to the linguistic one: in the value judgments of prescriptivism, hybridity is viewed as a step towards putrefaction.

There are many other types of solidarity besides nationalism, and each form of solidarity has a linguistic component expressed through prescriptive behavior. Milroy and Milroy (1985) analyze prescription primarily in terms of class divisions within a country, and thus are less focused on borrowings than on other types of prescription. They are interested in how sociolinguistics can help alleviate the ways in which prescription plays a role in social class discrimination within a given society. Prescription is seen as a way of rejecting the language of working-class men, and thus perpetuating social inequalities.

Joseph (2006, p. 32‑36) has taken a very different approach. He notes that linguists stress the constant evolution of language, and find prescriptivism reactionary in its desire to maintain a status quo, a project they consider doomed to failure. That failure is not certain or absolute, however; prescriptivism can slow the pace of change, or initiate change (as we shall see below). More importantly, he traces the power of prescriptivism to the connection between language and knowledge, a connection stressed by the Christian tradition. A single language is a more certain vehicle for the accurate transmission of God’s word. This single-language approach ultimately gets transferred from the religious domain to the nationalist domain, and the idea of a single language for each nation-state.

All the scholars cited connect prescription to power – prescriptors, after all, are attempting to impose their will – but the power wielded through prescription is not necessarily that of the state, or of the ethnic group, or of the dominant social class. Power and prestige depend upon particular circumstances, and may reverse expected dominance in a given situation. According to Leeman-Bouix (1994), prescriptivism is a kind of linguistic egotism in which the prescriptor destabilizes the targets of the prescription, robbing them of their linguistic security. Linguistic insecurity can be individual or a product of the educational system,3 which acts to denigrate the native speech patterns of the child in favor of a “more correct” version. As Marckwardt (1958) noted, “few Americans, even among the well-educated, are confident and assured of the essential aptness and correctness of their speech” (cited in Drake 1977, p. 3).4

How the power is exercised, or received, depends upon an understanding on both sides of the implications of the prescription. Understanding the intent of the prescriptor, or the perception of the intent by the receptor, is a subjective judgment, which might be admitted, and even then with caution, when the prescriptor explicitly expresses an intent, or with even greater caution, when an observer feels s/he has sufficient contextual information to interpret the situation.

The assumption of power is justified by the prescriptor for one of three reasons, according to Berrendonner (1982, p. 55‑62): (1) The value of speaking well, whether moral, political, esthetic or sensual. A moral value would attribute incorrect usage to laziness, a political one to disloyalty, the esthetic one to discord, and the sensual one to physical failing (for example, a pronunciation “mouillée” or “pâteuse”). (2) Social values, whether based on class or geographic dialect. (3) Utilitarian values, based on the functionality of language as a means of communication.

The power relationship is expressed in comments about exceptions or apparent contradictions in a given language: irregular forms, uncommon words or inflections, inconsistencies in the relationship between spelling and pronunciation. We will examine these features more closely below, but it is important to recognize the accuracy of Goffman’s observation concerning the criteria of social valuation: “evidence of social worth and of mutual evaluations will be conveyed by very minor things, and these things will be witnessed, as will the fact that they have been witnessed” (cited in Joseph 2004, p. 67).

There is certainly an element of truth to these analyses. How much truth is the question we wish to explore. The way to determine this is not by extracting the most outrageous examples of foolishness that one can find in prescriptive manuals, but by considering their entire contents as a means to comprehending their logic and method. To do this we need to consider:

  • the types of sources, their authors and their target audience;

  • the social context of the language, placing the usage guidance within a sociolinguistic and a sociopolitical situation;

  • the linguistic categories of the errors;

  • the types of justification provided for preferring one form over another;

  • the ways in which the prescription is phrased;

  • the relationship between demonstrated usage and the usage prescribed;

  • the demonstrated effect of the prescription.

Let us now look at each of these features, the parameters within them, and what the data concerning them can tell us about the nature of prescription. Many examples will be from the collection we gathered at the National Center for Supercomputer Applications at the University of Illinois. These treat specifically the French tradition since 1800. We collected print copies of several dozen works and digitized a few of them to aid in the analysis. This was a joint project between Institute for Computing in the Humanities, Arts and Social Sciences (I-CHASS) at the University of Illinois and the CNRS laboratory Histoire des théories linguistiques in Paris.

Types of sources. Modern linguists’ interest in prescriptivism has typically been limited to grammars, dictionaries and collections of remarks about usage. Grammars have claimed purely descriptive status in recent years, but the failure to acknowledge sociolinguistic factors in grammatical description often leads to a type of prescriptivism linguists claim to be avoiding. If a descriptive linguist states, for example, that there are 11 oral and 4 nasal vowels and no diphthongs in French (e.g. Tranel 1987, p. 3‑4), the many French speakers whose vowel inventory is different are implicitly categorized as errant. As Walter (1994) has pointed out, French speakers in France might have between 3 and 6 nasal vowels, and many varieties of French are highly diphthongized (e.g. Québécois). Minus the references to royalty, the language described in descriptive grammars is often not much different from Vaugelas’ (1647, a i verso) overtly prescriptive formula that there are two sorts of usage, one good and the other bad: “[le bon usage est] la façon de parler de la plus saine partie de la cour, conformément à la façon d’escrire de la plus saine partie des Autheurs du temps.”5 There has been a reaction in recent years to this criticism, with more recognition on the part of sociolinguistics of the source material for grammars (e.g. Quirk et al 1985, p. 15‑33).

Prescriptive remarks are frequently found in dictionaries, particularly in the form of usage markers. Usage markers can be quite diverse, are not necessarily used with consistent meaning, and are not all prescriptive in nature. Wooldridge (1977) created a five-part typology of such words: temporal, dialectal (spatial), dialectal (social or professional), stylistic, quantitative.6 Typical examples from the French tradition would be familier “familiar”, bas “lower class”, vieilli “old-fashioned”, “dated”, provincialisme “regional”, rare “rare”. Issue 9 of Lexique is devoted to the topic of “les marques d’usage dans les dictionnaires (XVIIe‑XVIIIe siècles)”.

In 1961 the Webster’s Third new international dictionary removed most usage markers, declaring itself descriptive not prescriptive. In reaction, the American heritage dictionary created a usage panel that would decide which labels to put on which words. This attempt at transparency and linguistic democracy is somewhat tempered by the fact that the panel is heavily weighted towards highly educated writers and other intellectuals (see Wolk 1972), but at least we know in each case what the vote was and can decide if educated speakers’ opinions are split evenly or tilted towards one alternative or another.

A third type of source is the usage manual, itself often organized in an alphabetical dictionary format. These works are overtly and uniquely prescriptive, although the authors may claim to be merely descriptive. Vaugelas (1647), for example, claims to be simply recording usage at the court, but the purpose of his remarks is to elaborate on Malherbe’s restrictions on proper usage. This tradition in France continues to the present day with works such as Laygues (2003) and Aristide (1989).

A fourth source is official governmental interventions concerning language use. The commissions terminologiques in France, which were started in 1970, and the ministerial decrees that they have produced are typical examples. These have resulted in the publication of numerous ministerial brochures, as well as the compilation Dictionnaire des termes officiels de la langue française (DGLF 1994). Here we consider only “corpus planning” (prescribing the use of particular words or forms), though this is frequently part of “status planning” (giving one language or dialect functional preference over others).

A fifth source is columns and letters to the editor concerning usage found in periodicals. Some of these periodicals are devoted entirely to questions of proper usage, such as the Journal de la langue française in the first half of the 19th century, the Courrier de Vaugelas which ran for a decade in the second half of that century, and Défense de la langue française, started in the late 1950s and continuing to this day. Others are daily or weekly newspaper columns in the mainstream press. Quemada (1972) led a team investigating “Chroniques de langage” published in newspapers during the period 1950‑1970. More recent columns of this sort are William Safire’s “On language” in the New York Times Magazine, and the “Langue sauce piquante” blog in Le Monde.7 These journalists are the “language mavens” attacked by Pinker (1994).

These sources are professional or semiprofessional commentaries on usage, composed by people who are trained linguists (grammarians, lexicographers), teachers, or other educated writers. They are aimed at a broad audience. Harder to capture, but no less interesting, are the countless instances of informal correction. We can identify some examples from corpora such as the following found in the ARTFL database:

Parlez nettement ! Articulez ! Vous ne dites pas « papa », vous dites « vava », par mollesse. Et c’est honteux. (Georges Duhamel, Le jardin des bêtes sauvages, 1934, p. 53; from the ARTFL database).

From these literary examples we can sometimes see the reaction of individuals to correction, which is generally lacking in the other types of sources, as illustrated in this exchange from a 19th-century novel:

On me l’a montré un jour dans le phaéton qu’il conduit lui-même à quatre chevaux.

– Tu devrais dire qu’il conduisait...

– Pourquoi ?

– Parce qu’il est mort.

– Bah !

(Ponson du Terrail, Le club des valets-de-cœur. 2e partie. Turquoise la pécheresse, 1859, p. 91; from the ARTFL database).

As more corpora become available and our search mechanisms become more precise, our access to a broader variety of prescriptive statements will become easier, thus enriching our understanding of prescriptive phenomena.

The social context of the language. Prescriptive efforts are inextricably bound with the prescriptors’ perceptions of the status of their language. While the prescription itself is corpus planning, the motivation is frequently one of status planning. At a time when endangered languages are a worldwide concern, it is not uncommon for prescriptivists to view the use of non-standard forms or borrowed words as a threat to the very life of the language, as we noted in the Belorussian example cited earlier. However, it is important that we understand for each prescriptivist, for each period, and for each language tradition, how the social and political context is expressed, and what each prescriptivist text can tell us about changing attitudes. Are borrowings from one language more acceptable than borrowings from another? Is that preference based on historical connections, contemporary politics or some other factor? In France, borrowings and the expansion of synonyms were widely accepted in the first half of 16th century, but rejected in the second half of the century (for example, in Henri Estienne’s Deux dialogues du nouveau langage françois italianizé et autrement desguizé, principalement entre les courtisans de ce temps, 1578). In England, the Saxonist movement that rejected borrowings from Latin and French was a by-product of hostility towards France and towards Catholicism, but by the end of the 17th century had shifted towards a less insular perspective (Kidd 1999, p. 222; Jones 1953, p. 214‑271).

The political situation of the francophone population of Québec has led to different prescriptive recommendations than those of the French government. For example, the French equivalent of “podcasting” in Québec is baladodiffusion, and in France diffusion pour baladeur. The provincial government explicitly rejected the French neologism, calling it too descriptive, incapable of producing derivative words, and already superseded by the term created in Québec.8 In general, the terminologists in Québec have been less accepting of calques than their French counterparts, at least in part because of historical concerns about the domination of English. What both countries agree on is the need to avoid the pure Anglicism “podcasting” or the simple addition of a French ending to the English root, podcastage.

Thus a broad and complete study of prescriptive behavior would help us to understand the interplay between identity-related social forces and prescriptivism. These may be nationalist, as Thomas (1991) emphasized in his book, but they can have other motivations as well. With a broad-based approach that covers multiple language traditions, over an extended history, from a variety of types of sources, our conclusions in this regard can be scientifically validated, rather than impressionistic.

The linguistic categories of the errors. In order to compare the focus of prescriptivists through the ages we9 have created a classification of types of errors. The broad categories of orthography, pronunciation, morphology, syntax, lexical/semantics, and sociolinguistics are further elaborated, so that we can see, within syntax for example, how many prescriptive remarks are devoted to subject-verb agreement, how many to word order, and so forth. Within sociolinguistic comments temporal, spatial and social (both by profession and by class) variation can be identified and compared.

Types of justification. Purism, defined as the desire to keep foreign elements out of a language, is one reason to prescribe one form over another, but many other reasons are expressed in the texts – and many linguistic decrees are not accompanied by any justification. The purist strain is sometimes presented in the form of an etymological argument: one should use a certain form because it has the longest history in the language. Etymology can also be invoked in morphological and orthographic debates, for instance insisting that a noun be of a specific gender because it had that gender in the source language. Selecting the gender of French nouns derived from Greek nouns ending in –e in their French form is a frequent source of uncertainty for French speakers; the prescriptivists cite the etymon for guidance, though there are cases where the etymological gender has clearly been superseded by usage.10

Prescription can also be justified on the basis of usage, both general – a kind of democratic principle more often invoked than followed – and specific to a class, or group of linguistic models, literary or social. In this regard a list of authors and other authorities cited is an important contribution to understanding the nature of prescription, along with an indication of whether the authors are considered positive or negative evidence. Biscarrat (1835, p. 94) cites two authors concerning the correct auxiliary verb to form compound past tenses of courir “to run”:

Courir se construit avec l’auxiliaire avoir. Racine a dit cependant :

Il en était sorti lorsque j’y suis couru.

Je doute fort, dit l’abbé d’Olivet, qu’il en soit du simple courir comme de son composé accourir. On dit indifféremment, j’ai accouru, je suis accouru. Mais je suis couru me paraît une de ces distractions dont les meilleurs écrivains ne sont pas toujours exempts.

Racine is the source of the negative example, while d’Olivet, author of Remarques sur la langue françoise (1767), is another grammarian citing unnamed literary transgressors.

“Reason” is another justification, which can mean analogy to another construction, or the logic of a particular construction. In opting for a silent <s> in the word “mœurs” (and condemning the practice of some actors of his period), Blondin (1823, p. 57) declares this pronunciation “suivant la raison et suivant l’Académie, de l’autorité de laquelle les comédiens français ressortissent plus que personne”. Reason as in logic is the source of Dupleix’s distinction between furie and fureur. Vaugelas (1647, p. 449) argues that reading good authors will be enough to help readers determine which term to use, but Dupleix (1651, p. 279) argues for a specific difference, concluding “Cete distinction prise de la Logique plustost que de la Grammaire est souvent necessaire en pareilles rencontres”.

The phrasing of prescription. While some prescriptions are absolute and unbending, even a cursory look at a few of the usage manuals makes it clear that many degrees of certainty are expressed in acts of prescription. In French we can point to a continuum ranging from the firmness of an imperative (ne dites pas…) or the modal falloir (il faut/ne faut pas), to less confrontational constructions, such as on dit, or on devrait dire. Much of the battle between Dupleix and Vaugelas can be captured in the effect of the adverb indifféremment, with Dupleix stating on dit indifféremment (meaning that two options were equally valid), while Vaugelas would insist, by the omission of the adverb (stating simply on dit) that only one was acceptable in court society. At the strong end there is complete interdiction of a particular form, softening to an obligation (more or less insistent), then to a recommendation, a reasoned preference, and finally, at the other end, complete tolerance.

Many prescriptivist manuals condemn the use of malgré que as a synonym of quoique, but the nature of the prescription varies from text to text, as expressed by the phrasing of the prescription. At the strong end we find:

Cette locution [malgré que] n’est pas françoise (Molard 1810, p. 170)

Ne dites pas… (Michel 1807, p. 40)

On ne peut pas dire… (Laveaux 18222 [1818], II, p. 173)

Dans aucun cas, malgré ne peut être suivi d’un que conjonctif (Blondin 1823, p. 54)

The opposition weakens a little in some other formulations, as it is admitted that one does see or hear the expression, but it is considered archaic or regional or simply lower class:

Malgré que est une expression dont ne se sont servis nos bons écrivains. (Sauger-Préneuf 18433 [1825], p. 51)

Malgré que, qui était français dans le 17e siècle, ne l’est plus ; il a été remplacé en France par quoique et bien que ; mais nous le conservons en Suisse. (Guilbert 1858, p. 144)

Ne dites pas : « Malgré que » je sois malade, il fait du bruit. « Quoique » je sois malade... Beaucoup de personnes cherchent aujourd’hui à remettre en usage l’ancienne tournure malgré que, tombée en désuétude depuis le XVIIe siècle, et remplacée par quoique, bien que, encore que, lors même que, etc. Nous ne voyons aucun avantage à cette tentative de régression : les courants du langage ne se remontent pas. (Joran 19152 [1911], p. 83‑84)

Malgré que au sens de quoique est considéré par certaines gens comme élégant et archaïque. Là gît l’erreur. Il appartient au langage populaire et l’on n’en trouve pas d’exemples avant certains auteurs lâchés du XVIIIe siècle. Casanova en est farci. Jusqu’à nouvel ordre, malgré que est de la langue concierge, bien qu’on puisse expliquer par une analogie toute simple l’abus de cet emploi.

M. Clédat a voulu l’autoriser au nom de la science. M. Gide au nom du naturel ; M. Paul Souday, M. Jacques Boulenger ont défendu les positions de l’art de l’usage littéraire. Je ne saurais retracer en détail leurs arguments excellents. (Thérive 1929, p. 140‑141)

More recently prescription has changed to tolerance, albeit with some recognition of others’ condemnation:

L’académicien Lancelot,11qui fut un des pépères de la Grammaire illustre des illustres de la Coupole, appelait tout spécialement l’attention de Xavier sur malgré que (p. 261) et lui apprenait avec majesté que « c’est une faute de prendre malgré que pour quoique. »

Oui, mais l’Académie, Littré et Lancelot proposent, les Académiciens disposent. Veuillez les ouïr un moment : « Malgré qu’elle eût la bouche bonne et le corps souple... » (H. de RÉGNIER, de l’Ac. fr., Le Mariage de Minuit, p. 31). « Malgré qu’il fût bien comique parfois, son amoureux... Malgré qu’il fût ponctuel, personne n’avait fait de réflexions sur son absence... » (G. LECOMTE, de l’Ac. fr., Les cartons verts, p. 10 et 487, Fasq., 1901).

Il est vrai que Lancelot pourra vous dire à ce propos ce qu’il disait au sujet de vis-à-vis de employé au sens de envers, à l’égard de : « Cette façon de parler est hélas ! usuelle et les Académiciens mêmes l’emploient, mais l’Académie en corps le condamne, c’est l’essentiel. » (Xavier, p. 244).

Et des gens simples répondront tout bêtement : A quoi qu’ça sert de prendre en groupe des décisions savantes, si chaque membre du groupe a le droit de s’en moquer ?... Les députés peuvent-ils, comme citoyens, se gausser des lois qu’ils votèrent comme parlementaires ? Et le Préfet de police, qui se balade en auto, n’est-il pas soumis au code de la route ? (Le Gal 1932, p. 70‑71)

[…] la première de ces fautes est tout simplement l’emploi de la locution conjonctive malgré que, qui, en bonne règle, ne devrait s’employer qu’avec le verbe avoir (malgré, c’est-à-dire mauvais gré, que j’en aie). Faute vénielle, en vérité, et qui est passée dans l’usage. Gide lui-même, dans une lettre à Paul Souday (Incidences), a déclaré que si cette « expression était fautive hier, elle a cessé de l’être. Elle ne se confond pas avec bien que, qui n’indique qu’une résistance passive ; elle indique une opposition ». Il faut reconnaître que cela est fort bien vu. (Aristide 1989, p. 278)

Although these excerpts make it appear that there is a linear progression towards tolerance − and this is certainly a general tendency in the French tradition −, there are in fact very recent examples of total interdiction. It is only through a thorough study of a large corpus of prescriptive works that we can find patterns and tendencies, and make the study of prescriptivism a matter of scientific inquiry.

Usage and prescription. The construction of digitized corpora offers us a remarkable opportunity to compare prescriptions to usage. Frantext/ARTFL is a collection of thousands of published books, mostly literary, throughout the history of the French language. The search capacity allows scholars to compare usage of disputed forms in specific time frames, so that we can determine how the prescription relates to the usage of the time it is issued. There are obvious limitations: the texts digitized are mostly literary or written by other intellectuals of the period. The database is thus a reflection of a particular usage, though such authors might occasionally try to imitate more popular forms of the language. The corpus is particularly useful for judgments about the currency of a particular word: if a word is described by a prescriptivist as archaic, the historical record of its usage can help us understand the context of the remark, and its accuracy. Jullien (1853, p. 46) claims that débours is outmoded, replaced by déboursés. ARTFL shows us that during the 19th century déboursés is used 14 times in the 897 works included from that period, distributed fairly evenly across the century, and by the most renowned authors (Stendhal, Balzac, Hugo, Flaubert, Zola, among others). Débours is used 10 times, predominantly by Balzac in the 1840s (4 of the 10 examples). Jullien (ibid., p. 94) also tells us that moyennant que is not acceptable (suggesting the alternatives pourvu que or à condition que). ARTFL shows that the condemned expression is used a half-dozen times during the 19th century, mainly by Charles Nodier in the 1830s. Pourvu que is used 776 times in the same period, and à condition que 102 times. Clearly Jullien has captured the sentiment of the time.

Caron (2002) has exploited this resource to demonstrate a critical period for the transformation of the norm in the period 1610‑1620, developing a concept that he terms a chronolecte. He studies a number of features showing variation in the 16th century, in which the variation virtually disappears after 1620. Many of these are thus faits accomplis by the time of Vaugelas’ Remarques in 1647, though still challenged by representatives of the older generation, such as Dupleix.

The effectiveness of prescription. The comparison of usage and prescriptions provides indirect evidence of the effectiveness of prescriptivism. Does language behavior change after the publication and dissemination of prescriptive rules? Alternatively, do prescriptive rules simply confirm usage that has already been established? Orthodoxy in linguistics would tell us that prescriptivism does not work, or only in rare exceptions. The efforts of the terminological commissions and the language academies are widely mocked, particularly by linguists in countries that lack these institutions. However, prescriptivism clearly does work in some instances. The widespread use of alternatives to the “masculine generic” in English is one example of success, as are the successful implantation of words such as ordinateur and logiciel in French. Scientific evaluation of success has been scarce, but not totally lacking. Langer (2001) studies the effect of grammatical writing on the use of the German auxiliary tun in Early New High German. He picked this feature because certain uses of the auxiliary remain widespread in colloquial or regional usage, but have disappeared in standard usage. Comments about the construction began to appear, first stigmatizing it in poetry early in the 17th century, then, in all genres in the late 17th century, with the label “regional” attached to the condemnation in the 18th century and then “lower-class” by mid-century. The subsequent disappearance of the structure in published work is indicative, then, of the success of prescriptivism.

While Langer concentrated on syntactic questions internal to German, another example of research on the effectiveness of prescriptivism is found in the collective work directed by Loïc Depecker, and commissioned by the Délégation générale à la langue françoise (DGLF). This focused on the success of efforts to supplant specific vocabulary items borrowed from English by French words, the terms proposed by the terminological commissions between their creation in 1970 and the beginning of the study in 1991. The terminological commissions work primarily in technical vocabulary, as reflected by the subject areas studied: computer science, broadcasting and advertising, aerospace, health and medicine, and genetic engineering. Some of these terms have extended into popular usage; in these instances, the English term has often been unshakable (e.g. “prompter”, for which the official suggestion is télésouffleur). For those terms confined to specialists working in a specific scientific domain, the chances of success are greater, though not assured (for example, few appear to use éveinage instead of “stripping” [varicose veins]). The techniques used in this study include analyses of dictionaries, technical documents, and questionnaires addressed to specialists in the field.

More general studies, including all the varieties of prescriptive comments (morphological, phonological, semantic, etc.), in addition to the syntactic and lexical studies discussed above, would be a logical outcome of the creation of an extensive database of prescriptive materials. Combining that database with the linguistic corpora appropriate for each linguistic tradition will allow us to evaluate the role that prescriptivism plays in language change. Thus we can replace the assumption that prescription is unnatural with a new conception of language change that incorporates all the influences upon that process, without a priori exclusions.

Conclusions

Our proposal then is to build a database of prescriptive materials that allows us to study prescription in a scientific manner. Combining searches based on the features we have described above in a relational database will allow us to see correlations and tendencies that previous work on prescriptivism has not uncovered. Such a project will allow us to distinguish national attitudes and traditions concerning prescriptive behavior in language, as well as changes over time of such attitudes. Thus we can replace stereotypical assumptions that mark discussions of prescriptivism both in professional linguistic literature, and in more popular venues.

Prescriptive works

  • Aristide [Maurice Chapelan], 1989. La langue française dans tous ses débats, Paris, F. Bourin. [Google Scholar]
  • Biscarrat, Félix, 1835. Nouveau manuel de la pureté du langage. Revu et augmenté par A. Boniface, Paris, Librairie encyclopédique de Roret. [Google Scholar]
  • Blondin, Jean-Noël, 1823. Manuel de la pureté du langage, Paris, The author. [Google Scholar]
  • DGLF, 1994. Dictionnaire des termes officiels de la langue française, Paris, Journal officiel de la République française. [Google Scholar]
  • Dupleix, Scipion, 1651. Liberté de la langue françoise dans sa pureté, Paris, Bechet. [Google Scholar]
  • Guilbert, Alphonse, 1858. Glossaire neuchâtelois, ou, Fautes de langage corrigées, Neuchatel, C. Gerster. [Google Scholar]
  • Joran, Théodore, 19152 [1911]. Le péril de la syntaxe et la crise de l’orthographe : recueil de locutions vicieuses, dressé par ordre alphabétique, Paris, Savaète. [Google Scholar]
  • Jullien, Bernard, 1853. Le langage vicieux corrigé. Liste alphabétique des fautes les plus ordinaires dans la prononciation, l’écriture et la construction des phrases, Paris, Hachette. [Google Scholar]
  • Laveaux, Jean-Charles, 18222 [1818]. Dictionnaire raisonné des difficultés grammaticales et littéraires de la langue française, Paris, Ledentu. [Google Scholar]
  • Laygues, Bernard, 2003. Évitez de dire… Dites plutôt…, Paris, Albin Michel. [Google Scholar]
  • Le Gal, Étienne, 1932. Cent manières d’accommoder le français. Un divorce attristant entre les académiciens, l’Académie, M. Toulemonde et la syntaxe, Paris, Nouvelle librairie française. [Google Scholar]
  • Michel, Jean-François, 1807. Dictionnaire des expressions vicieuses usitées dans un grand nombre de départemens, Paris/Nancy, Le Normand/Colas/Vigneulle/Bontoux. [Google Scholar]
  • Molard, Étienne, 1810. Le mauvais langage corrigé, Lyon, Yvernault & Cabin. [Google Scholar]
  • Olivet, Pierre-Joseph, 1767. Remarques sur la langue françoise, Paris, Barbou. [Google Scholar]
  • OQLF 2013: Office québécois de la langue françoise, 2013. “Baladodiffusion” [https://www.oqlf.gouv.qc.ca/ressources/bibliotheque/dictionnaires/terminologie_blogue/baladodiffusion.html, consulted on Aug. 22, 2019]. [Google Scholar]
  • Sauger-Préneuf, François, 18433 [1825]. Dictionnaire des locutions vicieuses usitées dans le midi de la France, Limoges, Ardillier. [Google Scholar]
  • Thérive, André, 1929. Querelles de langage, Paris, Librairie Stock. [Google Scholar]
  • Vaugelas, Claude Favre de, 1647. Remarques sur la langue françoise utiles à ceux qui veulent bien parler et bien escrire, Paris, Veuve Jean Camusat et Pierre le Petit. [Google Scholar]
  • Barnbrook, Geoff, 2005. “Usage Notes in Johnson’s Dictionary” , International journal of lexicography 18, 189‑201. [CrossRef] [Google Scholar]
  • Berrendonner, Alain, 1982. L’éternel grammairien. Étude du discours normatif, Berne, Peter Lang. [Google Scholar]
  • Bopp, Franz, 1816. Über das Conjugationssystem der Sanskritsprache in Vergleichung mit jenem der griechischen, lateinischen, persischen und germanischen Sprache, Frankfurt, Andreäischen Buchhandlung. [Google Scholar]
  • Calvet, Louis-Jean, 1998. “L’insécurité linguistique et les situations africaines”, in Calvet, Louis-Jean and Moreau, Marie-Louise (ed.), Une ou des normes ? Insécurité linguistique et normes endogènes en Afrique francophone, Paris, Cirelfa, 7‑38. [Google Scholar]
  • Caron, Philippe, 2002. “Vers la notion de chronolecte ? Quelques jalons à propos du français préclassique”, in Sampson, Rodney and Ayres-Bennett, Wendy (eds.), Interpreting the history of French. A Festschrift for Peter Rickard on the occasion of his eightieth birthday, Amsterdam, Rodopi, 329‑352. [Google Scholar]
  • Drake, Glendon F., 1977. The Role of prescriptivism in American linguistics, 1820‑1970, Amsterdam, John Benjamins. [CrossRef] [Google Scholar]
  • Francard, Michel, 1993. L’insécurité linguistique en Communauté française de Belgique, Brussels, Service de la langue française. [Google Scholar]
  • Jones, Richard F., 1953. The Triumph of English. A survey of opinions concerning the vernacular from the introduction of printing to the Restoration, Stanford, Stanford University Press. [Google Scholar]
  • Joseph, John E., 2004. Language and identity, Basingstoke, Palgrave Macmillan. [CrossRef] [Google Scholar]
  • — 2006. Language and politics, Edinburgh, University of Edinburgh Press. [Google Scholar]
  • Kidd, Colin, 1999. British identities before nationalism. Ethnicity and nationalism in the Atlantic world, 1600‑1800, Cambridge, Cambridge University Press. [CrossRef] [Google Scholar]
  • Landau, Sidney, 20012 [1984]. Dictionaries: The art and craft of lexicography, Cambridge, Cambridge University Press. [Google Scholar]
  • Langer, Nils, 2001. Linguistic purism in action. How auxiliary tun was stigmatized in Early New High German, Berlin/New York, Walter de Gruyter. [CrossRef] [Google Scholar]
  • Ledegen, Gudrun, 2000. Le bon français. Les étudiants et la norme linguistique, Paris, L’Harmattan. [Google Scholar]
  • Leeman-Bouix, Danielle, 1994. Les fautes de français existent-elles ?, Paris, Seuil. [Google Scholar]
  • Malmkjaer, Kirsten, 1991. Linguistics encyclopedia, London, Routledge. [Google Scholar]
  • Marckwardt, Albert H., 1958. American English, New York, Oxford University Press. [Google Scholar]
  • Milroy, James and Milroy, Leslie, 1985. Authority in language. Investigating standard English, London/New York, Routledge. [Google Scholar]
  • Pinker, Stephen, 1994. The language instinct, New York, W. Morrow. [Google Scholar]
  • Quemada, Bernard, 1970. Bibliographie des chroniques de langage publiées dans la presse française 1 : 1950‑1965, Paris, Didier. [Google Scholar]
  • — 1972. Bibliographie des chroniques de langage publiées dans la presse françoise 2 : 1966‑1970, Paris, Didier. [Google Scholar]
  • Quirk, Randolph, Greenbaum, Sidney, Leech, Geoffrey and Svartvik, Jan, 1985. A comprehensive grammar of the English language, London/New York, Longman. [Google Scholar]
  • Remysen, Wim, 2004a. “La variation linguistique et l’insécurité linguistique : le cas du français québécois”, in Bouchard, Pierre (ed.), La variation dans la langue standard, Actes du colloque tenu les 13 et 14 mai 2002 à l’université Laval dans le cadre du 70e Congrès de l’Acfas, Québec, Office québécois de la langue française, 23‑36. [Google Scholar]
  • — 2004b. “L’insécurité linguistique des francophones ontariens et néo-brunswickois. Contribution à l’étude de la francophonie canadienne”, in Langlois, Simon and Létourneau, Jocelyn (eds.), Aspects de la nouvelle francophonie canadienne, Québec, Presses de l’université Laval, 95‑116. [Google Scholar]
  • Saussure, Ferdinand de, 1972 [1916]. Cours de linguistique générale (ed. T. de Mauro), Paris, Payot. [Google Scholar]
  • Thomas, George, 1991. Linguistic purism, London/New York, Longman. [Google Scholar]
  • Tranel, Bernard, 1987. The sounds of French. An introduction, Cambridge, Cambridge University Press. [CrossRef] [Google Scholar]
  • Walter, Henriette, 1994. “Variétés actuelles des voyelles nasales du français” , Communication and cognition 27, 223‑235. [Google Scholar]
  • Wexler, Paul N., 1974. Purism and language: A study of modern Ukrainian and Belorussian nationalism (1840‑1967), Bloomington, Indiana University Press. [Google Scholar]
  • Wolk, Anthony, 1972. “Linguistic and social bias in the American Heritage Dictionary” , College English 33, 8, 930‑935. [CrossRef] [Google Scholar]
  • Wooldridge, T. Russon, 1977. Les débuts de la lexicographie française : Estienne, Nicot et le Thresor de la langue françoyse (1606), Toronto, University of Toronto Press. [Google Scholar]

Secondary literature

  • Aristide [Maurice Chapelan], 1989. La langue française dans tous ses débats, Paris, F. Bourin. [Google Scholar]
  • Biscarrat, Félix, 1835. Nouveau manuel de la pureté du langage. Revu et augmenté par A. Boniface, Paris, Librairie encyclopédique de Roret. [Google Scholar]
  • Blondin, Jean-Noël, 1823. Manuel de la pureté du langage, Paris, The author. [Google Scholar]
  • DGLF, 1994. Dictionnaire des termes officiels de la langue française, Paris, Journal officiel de la République française. [Google Scholar]
  • Dupleix, Scipion, 1651. Liberté de la langue françoise dans sa pureté, Paris, Bechet. [Google Scholar]
  • Guilbert, Alphonse, 1858. Glossaire neuchâtelois, ou, Fautes de langage corrigées, Neuchatel, C. Gerster. [Google Scholar]
  • Joran, Théodore, 19152 [1911]. Le péril de la syntaxe et la crise de l’orthographe : recueil de locutions vicieuses, dressé par ordre alphabétique, Paris, Savaète. [Google Scholar]
  • Jullien, Bernard, 1853. Le langage vicieux corrigé. Liste alphabétique des fautes les plus ordinaires dans la prononciation, l’écriture et la construction des phrases, Paris, Hachette. [Google Scholar]
  • Laveaux, Jean-Charles, 18222 [1818]. Dictionnaire raisonné des difficultés grammaticales et littéraires de la langue française, Paris, Ledentu. [Google Scholar]
  • Laygues, Bernard, 2003. Évitez de dire… Dites plutôt…, Paris, Albin Michel. [Google Scholar]
  • Le Gal, Étienne, 1932. Cent manières d’accommoder le français. Un divorce attristant entre les académiciens, l’Académie, M. Toulemonde et la syntaxe, Paris, Nouvelle librairie française. [Google Scholar]
  • Michel, Jean-François, 1807. Dictionnaire des expressions vicieuses usitées dans un grand nombre de départemens, Paris/Nancy, Le Normand/Colas/Vigneulle/Bontoux. [Google Scholar]
  • Molard, Étienne, 1810. Le mauvais langage corrigé, Lyon, Yvernault & Cabin. [Google Scholar]
  • Olivet, Pierre-Joseph, 1767. Remarques sur la langue françoise, Paris, Barbou. [Google Scholar]
  • OQLF 2013: Office québécois de la langue françoise, 2013. “Baladodiffusion” [https://www.oqlf.gouv.qc.ca/ressources/bibliotheque/dictionnaires/terminologie_blogue/baladodiffusion.html, consulted on Aug. 22, 2019]. [Google Scholar]
  • Sauger-Préneuf, François, 18433 [1825]. Dictionnaire des locutions vicieuses usitées dans le midi de la France, Limoges, Ardillier. [Google Scholar]
  • Thérive, André, 1929. Querelles de langage, Paris, Librairie Stock. [Google Scholar]
  • Vaugelas, Claude Favre de, 1647. Remarques sur la langue françoise utiles à ceux qui veulent bien parler et bien escrire, Paris, Veuve Jean Camusat et Pierre le Petit. [Google Scholar]
  • Barnbrook, Geoff, 2005. “Usage Notes in Johnson’s Dictionary” , International journal of lexicography 18, 189‑201. [CrossRef] [Google Scholar]
  • Berrendonner, Alain, 1982. L’éternel grammairien. Étude du discours normatif, Berne, Peter Lang. [Google Scholar]
  • Bopp, Franz, 1816. Über das Conjugationssystem der Sanskritsprache in Vergleichung mit jenem der griechischen, lateinischen, persischen und germanischen Sprache, Frankfurt, Andreäischen Buchhandlung. [Google Scholar]
  • Calvet, Louis-Jean, 1998. “L’insécurité linguistique et les situations africaines”, in Calvet, Louis-Jean and Moreau, Marie-Louise (ed.), Une ou des normes ? Insécurité linguistique et normes endogènes en Afrique francophone, Paris, Cirelfa, 7‑38. [Google Scholar]
  • Caron, Philippe, 2002. “Vers la notion de chronolecte ? Quelques jalons à propos du français préclassique”, in Sampson, Rodney and Ayres-Bennett, Wendy (eds.), Interpreting the history of French. A Festschrift for Peter Rickard on the occasion of his eightieth birthday, Amsterdam, Rodopi, 329‑352. [Google Scholar]
  • Drake, Glendon F., 1977. The Role of prescriptivism in American linguistics, 1820‑1970, Amsterdam, John Benjamins. [CrossRef] [Google Scholar]
  • Francard, Michel, 1993. L’insécurité linguistique en Communauté française de Belgique, Brussels, Service de la langue française. [Google Scholar]
  • Jones, Richard F., 1953. The Triumph of English. A survey of opinions concerning the vernacular from the introduction of printing to the Restoration, Stanford, Stanford University Press. [Google Scholar]
  • Joseph, John E., 2004. Language and identity, Basingstoke, Palgrave Macmillan. [CrossRef] [Google Scholar]
  • — 2006. Language and politics, Edinburgh, University of Edinburgh Press. [Google Scholar]
  • Kidd, Colin, 1999. British identities before nationalism. Ethnicity and nationalism in the Atlantic world, 1600‑1800, Cambridge, Cambridge University Press. [CrossRef] [Google Scholar]
  • Landau, Sidney, 20012 [1984]. Dictionaries: The art and craft of lexicography, Cambridge, Cambridge University Press. [Google Scholar]
  • Langer, Nils, 2001. Linguistic purism in action. How auxiliary tun was stigmatized in Early New High German, Berlin/New York, Walter de Gruyter. [CrossRef] [Google Scholar]
  • Ledegen, Gudrun, 2000. Le bon français. Les étudiants et la norme linguistique, Paris, L’Harmattan. [Google Scholar]
  • Leeman-Bouix, Danielle, 1994. Les fautes de français existent-elles ?, Paris, Seuil. [Google Scholar]
  • Malmkjaer, Kirsten, 1991. Linguistics encyclopedia, London, Routledge. [Google Scholar]
  • Marckwardt, Albert H., 1958. American English, New York, Oxford University Press. [Google Scholar]
  • Milroy, James and Milroy, Leslie, 1985. Authority in language. Investigating standard English, London/New York, Routledge. [Google Scholar]
  • Pinker, Stephen, 1994. The language instinct, New York, W. Morrow. [Google Scholar]
  • Quemada, Bernard, 1970. Bibliographie des chroniques de langage publiées dans la presse française 1 : 1950‑1965, Paris, Didier. [Google Scholar]
  • — 1972. Bibliographie des chroniques de langage publiées dans la presse françoise 2 : 1966‑1970, Paris, Didier. [Google Scholar]
  • Quirk, Randolph, Greenbaum, Sidney, Leech, Geoffrey and Svartvik, Jan, 1985. A comprehensive grammar of the English language, London/New York, Longman. [Google Scholar]
  • Remysen, Wim, 2004a. “La variation linguistique et l’insécurité linguistique : le cas du français québécois”, in Bouchard, Pierre (ed.), La variation dans la langue standard, Actes du colloque tenu les 13 et 14 mai 2002 à l’université Laval dans le cadre du 70e Congrès de l’Acfas, Québec, Office québécois de la langue française, 23‑36. [Google Scholar]
  • — 2004b. “L’insécurité linguistique des francophones ontariens et néo-brunswickois. Contribution à l’étude de la francophonie canadienne”, in Langlois, Simon and Létourneau, Jocelyn (eds.), Aspects de la nouvelle francophonie canadienne, Québec, Presses de l’université Laval, 95‑116. [Google Scholar]
  • Saussure, Ferdinand de, 1972 [1916]. Cours de linguistique générale (ed. T. de Mauro), Paris, Payot. [Google Scholar]
  • Thomas, George, 1991. Linguistic purism, London/New York, Longman. [Google Scholar]
  • Tranel, Bernard, 1987. The sounds of French. An introduction, Cambridge, Cambridge University Press. [CrossRef] [Google Scholar]
  • Walter, Henriette, 1994. “Variétés actuelles des voyelles nasales du français” , Communication and cognition 27, 223‑235. [Google Scholar]
  • Wexler, Paul N., 1974. Purism and language: A study of modern Ukrainian and Belorussian nationalism (1840‑1967), Bloomington, Indiana University Press. [Google Scholar]
  • Wolk, Anthony, 1972. “Linguistic and social bias in the American Heritage Dictionary” , College English 33, 8, 930‑935. [CrossRef] [Google Scholar]
  • Wooldridge, T. Russon, 1977. Les débuts de la lexicographie française : Estienne, Nicot et le Thresor de la langue françoyse (1606), Toronto, University of Toronto Press. [Google Scholar]
1

“La science qui s’est constituée autour des faits de langue a passé par trois phases successives avant de reconnaître quel est son véritable et unique objet. On a commencé par faire ce qu’on appelait de la « grammaire ». Cette étude, inaugurée par les Grecs, continuée principalement par les Français, est fondée sur la logique et dépourvue de toute vue scientifique et désintéressée sur la langue elle-même ; elle vise uniquement à donner des règles pour distinguer les formes correctes des formes incorrectes ; c’est une discipline normative, fort éloignée de la pure observation et dont le point de vue est forcément étroit” (Saussure 1972 [1916], p. 14). Saussure goes on to describe the second phase as philology, the study of language for the interpretation of texts, with the third being that of comparative grammar, beginning with Bopp’s (1816) study of Sanskrit conjugations.

2

While descriptive linguistics eschews prescriptivism, prescriptive elements in descriptive linguistics are pervasive. For example, the notion of “hypercorrection” reflects the attitude of linguists towards prescriptive grammar. “As an occasional consequence of prescriptive pressures, some speakers have mistakenly extended particular prescriptive rules in an attempt to avoid mistakes” (Quirk et al 1985, p. 14). The label “hypercorrection” assumes (a) that the speaker has been influenced by a prescriptive rule, and (b) that s/he has applied the rule beyond the originally targeted usage. However, these claims are rarely, if ever, investigated and proven, and allow diachronic elements and assumptions about speakers’ intentions to substitute for the descriptive, synchronic principles of modern linguistics. The Quirk grammar, for example, lists a variety of relative pronoun options, without categorizing any as incorrect − except for the so-called “hypercorrection” of “*The Ambassador, whom we hope will arrive at 10 a.m., …” (ibid., p. 368). In one instance, the so-called hypercorrection “has been institutionalized… in fact there is no alternative relative pronoun: Here is Captain Morse than whom there is no braver soldier” (ibid.). If a form has been institutionalized and there is no other option, then by labeling that form a hypercorrection the authors insert a historical element into a synchronic description of the language, and a subjective judgment upon such historical developments.

3

For a study of prescriptivism in the French educational system, see for instance Ledegen (2000).

4

Linguistic insecurity is itself a vast topic. Calvet (1998) describes the phenomenon in the Francophone African context, Remysen (2004a and 2004b) in Francophone Canada, Francard (1993) in Francophone Belgium.

5

Berrendonner (1982, p. 14) remarks that “derrière les professions de foi anti-normatives que ces textes [grammaires descriptives] exposent bruyamment se dissimulent souvent sous couvert de « descriptivisme » des procédés pragmatiques, et même des affirmations, tout aussi « normatifs » que ceux des grammairiens traditionnels”.

6

Barnbrook (2005, p. 190‑191) cites other classificational schemes, by Landau (20012 [1984]) and Malmkjaer (1991). Wooldridge’s “temporal” corresponds to Barnbrook’s “currency”; similarly, “spatial” = “regional” or “diatopic”; “social and professional” = “technical” or “diatechnical”; “stylistic” = “diaintegrative” or “diastratic”. Wooldridge’s “quantitative” has no equivalent in their typology, but they add “slang”, “taboo”, and “insult” as distinct categories. Such categories are not clearly distinguished, and words could easily belong to several categories, being, for instance, both regional and taboo (e.g. “shite”).

7

See the blog’s webpage [http://correcteurs.blog.lemonde.fr].

8

“En raison de sa forme trop descriptive, plus difficilement implantable, de son inaptitude à produire des dérivés adéquats et d’une concurrence inutile avec le terme baladodiffusion, déjà utilisé par un grand nombre d’usagers du Québec et de la francophonie, le terme diffusion pour baladeur n’a pas été retenu pour désigner le présent concept.” (OQLF 2013).

9

E-Jung Choi and Caron have been instrumental in elaborating this classification scheme. Another is proposed by Quemada (1970) in his collection of remarks on language found in the press.

10

Énigme, neuter in Greek and Latin (aenigma), was frequently masculine in the 16th century, but was generally considered feminine by the end of the 17th century; 19th‑century prescriptivists all call for feminine, e.g. Biscarrat (1835, p. 161): “ÉNIGME, s.f. Quelques personnes font à tort ce substantif masculin, et disent; UN ÉNIGME. Dites UNE ÉNIGME”.

11

Lancelot was the pen name of Abel Hermant, a writer elected to the Académie française, and expelled from the Académie in 1945 for his collaboration with the Nazi occupation. He was the author of the cited Xavier ou les entretiens sur la grammaire française (1923).