Navigation – Plan du site

AccueilNuméros26“Corrupt as this dialect is, it w...

“Corrupt as this dialect is, it will not be totally useless”. Colonialism and an “anomalous” Indian language in Hadley’s Grammatical remarks

« Aussi corrompu que soit ce dialecte, il ne sera pas totalement inutile ». Colonialisme et un langage indien « anomal » dans les Grammatical remarks du capitaine G. Hadley
“Por corrupto que sea este dialecto, no será del todo inútil”. Colonialismo y una lengua india anómala en las Grammatical remarks del capitán G. Hadley
“Per quanto corrotto sia questo dialetto, non sarà del tutto inutile”. Colonialismo e un'anomala lingua indiana nelle Grammatical remarks del capitano G. Hadley
Louis Nagot

Résumés

Cette contribution, fondée sur les Grammatical remarks on the practical and vulgar dialect of the Indostan (1772) du capitaine George Hadley, vise à mettre en lumière la perception d’« anomalies » dans une langue indienne vernaculaire, et les efforts pour la « réduire à un système », selon l’expression d’Hadley, afin de l'adapter aux normes grammaticales européennes (c'est-à-dire essentiellement latines). Dans un contexte intellectuel admettant une hiérarchie des langues, selon leur degré supposé de « perfection », le projet de Hadley est de prendre en charge la « corruption » attribuée à la langue Moors et de la rectifier grâce à une organisation plus compréhensible de la langue. Ce faisant, cependant, il a puissamment contribué à rendre la langue plus « anomale » aux yeux du lecteur-apprenant, en insistant sur ses lacunes, ses défauts, ou en introduisant de nouvelles catégories grammaticales floues – renforçant les perceptions coloniales à propos de cette « langue barbare entre toutes ».

Haut de page

Texte intégral

  • 1 I am deeply indebted to Annie Montaut, whose linguistic expertise on Eastern Hindi dialects gave a (...)
  • 2 The full title being : Grammatical remarks on the practical and vulgar dialect of the Indostan lang (...)
  • 3 Varro, On the Latin Language, Volume II: Books 8-10. Fragments, ed. and trans. Roland G. Kent, Camb (...)
  • 4 Georges Canguilhem, The Normal and the Pathological, trans. Carolyn R. Fawcett & Robert S. Cohen, N (...)

1This contribution1 casts light on the way in which some British grammarians dealt with the so-called “anomalies” (we shall go through their own terminology in more detail) of Indian languages, and how, if we look closer, they introduced confusion leading to see anomaly in them. In that regard, the case of George Hadley’s Grammatical remarks on the practical and vulgar dialect of the Indostan language commonly called Moors is tremendously interesting.2 The notion of anomaly, along with its opposite, analogy, has been a fundamental part of the linguistic field: Varro, as early as the 1st century BCE, gave an account on the raging debates that were going on between grammarians supporting analogy (ratio) or anomaly (consuetudo), that is, regular and irregular forms.3 The point here is not to give an historical account of the notion of anomaly in linguistics, but rather to examine the role it played in Hadley’s legitimation and description of the language. For that, we may go back to the very concrete description of anomaly proposed by Canguilhem: “a crack, a relief or a gap in the linearity of things.” In other words, the opposite of the regular, even, smooth, uniform (omalos).4 I argue that Hadley claimed to explain, or to smooth out, what he perceived to be irregularities in Moors, but in fact achieved the opposite by creating more cracks in his representation of the language. In other words, his views on the language and his methodology were in themselves a powerful anomalizing factor.

2Hadley’s book is certainly fascinating in many aspects. For the present study, I will leave aside his Dialogues, captivating as they are, to focus on the grammatical account he gives on “Moors” (also indiscriminately called “Hindostanic” by Hadley), as well as to the dictionary he attached to it (New Vocabulary). Moreover, I will limit myself to the first three editions of his Grammatical remarks, spanning from 1772 to 1784, the later editions showing dramatic editorial changes that would need to be addressed separately. Before examining how Hadley dealt with Moors language, it is necessary to give a brief account on his life and the editorial process of his Grammatical remarks, since this author is practically unknown, little to no scholarship having been dedicated to his work to date.

“A philological Argonaut”

  • 5 William Carmichael Smyth, The Hindoostanee Interpreter, London, Richardson, 1824, p. iv.
  • 6 On this point, the Dialogues of the 3rd edition onwards are particularly instructive.

3Hadley as an author deserves a particular attention for three reasons. To start with, he is among the first European grammarians of Northern Indian languages (along with the Dutch Ketelaar, a century before), and broke new grounds in this regard. Hadley was indeed named a “philological Argonaut” by W.C. Smyth, author of an Hindoostanee Interpreter explicitly following Hadley’s method.5 Second, Hadley may be one of the authors whose colonial outlook and despise for local populations is expressed in the bluntest way.6 Third, due to the lack of competition in the editorial field, Hadley’s book was a “flattering success and rapid sale,” going through 7 editions in 37 years. This must have had major consequences: for decades, Hadley’s Grammatical remarks was virtually the only book available to any Westerner (whether English-speaking or not) to get a grasp of Moors.

  • 7 Richard Steadman-Jones, Colonialism and Grammatical Representation: John Gilchrist and the Analysis (...)
  • 8 Edward Dodwell & James Samuel Miles (eds.), Alphabetical list of the officers of the Indian army, 1 (...)
  • 9 The uncertainty on the date of his return to England makes it difficult to determine with any preci (...)
  • 10 Gordon Goodwin, “George Hadley,” Dictionary of National Biography (vol 23), London, Smith, Elder & (...)

4In spite of his monopoly and influence, Hadley’s life is hardly known. Steadman-Jones mentions that Hadley’s father worked with the East India Company, and that he spent some time and effort collecting books about China.7 Turning back to his son, the grammarian we’re interested in, I could find no date of birth. As far as we know, he was appointed in India in November 1762, promoted lieutenant in 1764, and then, rather quickly, captain in 1766, by Robert Clive himself.8 He resigned from service in December 1771; this may be due to the birth of an illegitimate child, Sarah, in October 1771. The date of his return to Europe is unknown, but since the first edition of the Grammatical remarks appeared in 1772 in London, we can suppose he was back in England by that time.9 By all means, we can safely assume that he was back in England in 1788, when the printer Thomas Briggs from Kingston-upon-Hull added his name and his credit to a compilation, A New and Complete History of the Town and County of the Town of Kingston-upon-Hull.10 Lastly, Hadley is credited with a translation from French to English of Joseph Labrosse’s Gazophylacium linguae persarum (1684), a thesaurus of the Persian language, and of diverse poetical pieces. Hadley died in London in 1798.

5Hadley asserts that his Grammatical remarks are a direct consequence of his responsibilities in India: the task of commanding Indian native soldiers (the sepoys), he confesses, being hardly achievable without basic notions of his men’s language:

  • 11 op. cit., p. i-ii.

I soon became sensible that it would be impossible to discharge my duty… without a knowledge of the language spoken by those whom I was to command, and experience soon showed me that the corrupt dialect would be more immediately useful [than English] in a military capacity.11

  • 12 op. cit., p. iii.
  • 13 I do not have the linguistic capacity to assess the value of this Persian grammar; however, it is o (...)

6Hadley would thus have composed a short memorandum of some 30 pages containing the basics of Moors “for [his] own private use only.” The rest of the story seems a fascinating editorial adventure. In 1770, a friend of his asks Hadley’s opinion about a short grammar published in Europe. The work in question happens to be plagiarized from Hadley’s initial notes (either found by the editor or sold to him by one of Hadley’s correspondents), to which was added a vocabulary, “erroneous in quality and deficient in quantity” according to Hadley.12 Hadley thus decides to republish his notes under his name, in 1772, which is now known as the first of seven editions. After the second edition of his Grammatical remarks (1774), Hadley publishes some Introductory grammatical remarks on the Persian language (1776).13

  • 14 John B. Gilchrist, A Grammar, of the Hindoostanee Language, or Part Third of Volume First, of a Sys (...)
  • 15 The 6th edition was not available at the British Library, where I consulted the others, nor online.

7In this section I highlight some of the most interesting features of key editions of Hadley’s book. In the 3rd edition (1784), Hadley includes bilingual dialogues, giving us an account of the daily life in British India, and its colonial context. The 4th edition (1796) is the last one that Hadley presided over; more importantly, Hadley added a new classification for the verbal system, probably “inspired” by his rival’s, Gilchrist, published earlier the same year14. After Hadley’s death in 1798, the 5th edition (1801) was presented as “corrected, improved and much enlarged by Mirza Mohammed Fitrut,” a native instructor. The title is almost misleading: the changes and corrections in the grammar were dramatic, integrating, it seems, the standardization of modern Hindustani as operated by Gilchrist. This 5th edition seems to be the final state of the grammar: I could not consult the 6th edition (1802),15 but the 7th and last edition (1809) is practically a facsimile of the 5th, suggesting that the 6th edition did not go through important changes.

The hierarchy of languages in the 18th century intellectual context

8In order to understand Hadley’s perspective in the Grammatical remarks, some elements of the intellectual context are useful, especially regarding grammarians and philosophers of the language. Thankfully, Varro’s concern was still up to date and the questioning surrounding the couple anomaly/analogy was quite vivid in Hadley's times. In 1762, Robert Lowth wrote A Short Introduction to English Grammar to demonstrate that the errors of spoken English were due not to an anomalous language, but to a defective use of it, that could be corrected and put back in order. This took place in a more general rationalist consensus among grammarians of the late 18th century, according to which language was based on logical, metaphysical categories that preexisted and transcended what may or may not be actually found in practice. For example, in the Encyclopeadia Britannica, William Smellie drew a distinction between grammar as an art (empirical description and observation of rules), and grammar as a science, theoretical and prescriptive. Grammar as a science, Smellie writes,

  • 16 William Smellie, “Grammar”, in William Smellie (ed.), Encyclopaedia Britannica, Edinburgh, Bell & M (...)

views language in itself, neglecting particular modifications..., distinguishes between those particulars those which are essential to language, and those accidental, and thus furnishes a certain standard by which different languages may be compared, and their excellencies or defects pointed out.16

  • 17 Sylvain Auroux, La révolution technologique de la grammatisation, Liège, Mardaga, 1994.

9We immediately see the consequence of this theoretical model: the science of grammar having established the general, rational principles of language, the actual languages could be evaluated, and ranked, according to their correspondence to the model. More precisely, the model elaborated by the Greek and Latin philosophical and rhetorical tradition. This phenomenon is described by Auroux as “grammatisation,” in other words, “a process that led to produce dictionaries and grammars of all languages around the world on the basis of the Greek-Latin tradition.”17

10Hadley was writing at a time when the belief in inequality between languages was ripe, and widely accepted by his peers. A hierarchy of languages organized superior languages at the top, and inferior languages at the bottom. Superior languages were identified by their omalos, or regularity, and their compliance with the general principles of classical European languages. By contrast, vernacular languages were treated as inferior since their suffered from consuetudo, or habit, irregularities or corruption. These languages did not fit the general grammatical theory and were deemed inferior as a result. Going back to British India, we find a good illustration of this perspective in the (too) famous praise of the Sanskrit language, given by the famous orientalist William Jones, who was also the founder of the Asiatic Society of Calcutta (1784). According to Jones,

  • 18 John Shore, Lord Teignmouth, The Works of Sir William Jones. With a Life of the Author, London, Sto (...)

the Sanskrit language, whatever be its antiquity, is of a wonderful structuremore perfect than the Greek, more copious than the Latin, and more exquisitely refined than either.18

  • 19 As shows Phiroze Vasunia in The classics and colonial India, Oxford, Oxford University Press, “Clas (...)
  • 20 Up to this day, the presentation of the catalogue of the Asiatic society reflects the hierarchy tha (...)

11Jones is delighted by the fact that Sanskrit not only complies with the regularity found in those European classical languages, but exceeds it. Regularity was the ground on which the dominant grammatical tradition was found, and the touchstone according to which the degree of perfection of a language was evaluated. Jones was therefore committing a bold statement in deeming Sanskrit’s structure as “more exquisitely refined” than Latin’s and Greek’s.19 This hierarchization of languages (classical vs. vernacular, more perfect vs. less perfect) is salient in the grammars of the time as well as in the organization of institutions that promoted the study of other languages, like the Asiatic Society.20

  • 21 Even if the purpose of this article is not to recreate the linguistic pedigree of Moors, we should (...)
  • 22 op. cit., p. vii.

12Moors’ structure, by contrast, was not considered exquisitely refined at all; in fact, it was not even viewed as a consistent language before Hadley.21 The very title of his grammar is eloquent: Grammatical remarks on the practical and current dialect of the jargon of Hindostan (my emphasis). Hadley’s objective was to reveal the grammatical coherence of what was until then only perceived as a “jargon” or a “barren dialect,”22 sitting at the bottom of the hierarchy (neither European, neither classical, neither a language). As Hadley explains in his Preface,

amongst the many difficulties stated by the Eastern literati, it was urged that the current Hindostanic [i.e. Moors] was too irregular to admit a grammar; that the Persian was so connected with it, being derived therefrom, that it was impossible to attain one without the other…: all of which is erroneous. It does admit a grammar, it has nothing to do with Persian… What is laid down here is more than sufficient to obviate the first objection.

13To correct this misconception (the alleged “irregularity” of Moors and closeness to Persian preventing from considering it a proper language, with a proper grammar), Hadley suggests that the merits of his works stems from its organization:

whatever merit can be assumed from the following remarks, it must be attributed intirely [sic] to the disposition of what is laid down, and not to any new lights thrown on the subject... Nothing of this sort has ever been attempted.

14Hadley wishes to submit a rational, concise and pragmatical presentation of the language: in other words, his works seeks to create an omalos, grammatical organization, out of the anomalos Moors jargon as it was seen before him. It is this claim that we will discuss.

  • 23 op. cit., p. 29.
  • 24 ibid., p. iii.
  • 25 Horace, Satires, Epistles and Ars Poetica, ed. H.K. Fairclough, Cambridge, Loeb Classical Library, (...)

15Before turning to the contents of Hadley’s grammar, I would like to underline one last keyword, being the heart of his project: “practical.” Moors, as the “practical and current jargon of Hindostan,” ought to be organized “for the common practice in Bengal” (my emphasis), as the subtitle of the grammar goes. Practice, and no more: Hadley himself, at the end of the grammar, states that his enterprise is “merely practical.”23 Is there something of the military man here? On the one hand, Hadley rejects the excessive pedantry of other “Eastern literati,”24 as we shall see with more detail below; on the other, while being aware of his modest scope, he does claim to be making the best of this 30-pages format. This is made clear in at least two instances. First, Hadley opens with Latin quote from Horace, to frame the spirit of the grammar: “if you found better than these [principles], share it genuinely, otherwise, follow them with me.”25 Second, Hadley comments on the feedback he received on the Grammatical remarks from the renowned William Jones. In a letter to Hadley, Jones wrote:

  • 26 Preface to the 4th edition of George Hadley, Grammatical Remarks (1796).

This book is small change of immediate use; mine is bank notes, with which in his pocket one may starve, and not be able to get what one wants. Where one buys mine, you will sell a hundred.26

16Hadley, as we can imagine, seemed very pleased with this straightforward acknowledgement of the practical essence of his work – moreover coming from such an authority as Jones was. He therefore quoted Jones’ comment in his Preface to the 4th edition (1796), and concluded: “a juster and more compendious estimation, so apposite to the intention of the author, could not have been given in so few words.”

From anomalia to omalia: the virtues of disposition

17We shall now consider the efforts of disposition Hadley displayed to give Moors proper credit. His first argument is to claim of the universality of an underlying grammatical structure, however “corrupted” the jargon may seem:

  • 27 op. cit., p. vii.

Every language has its idiom, that is, some peculiarity which often runs counter to most general maxims, bred by corruption and strengthened by custom; it prevails more or less in all tongues, and whilst keeping this in view we may safely attempt to reduce the most barbarous language to as regular a system as it will admit of since all grammar is composed from the language.27

  • 28 ibid., p. 18.
  • 29 ibid., p. 18.

18In other terms, the regularity of grammatical features, understood as “general maxims,” are blurred by their degradation in an idiom, and Hadley’s task, through his rational disposition, is to reorder it so as to be make it understandable. Here, note Hadley’s remark that the “corruption” phenomenon should happen “in all tongues”, so as to give a broader perspective on Moors’ irregularities, and smoothen the crack. Consider for instance the treatment of the manifest irregularities of the verb hona “to be” in Moors (Present: hum hoo,ah; Praeterperfect: Hum t,hau). Those irregularities are either looked over (“the auxiliary verb Hona, to be, is declined as follows, with three regular tenses,”28 my emphasis), either brought back to a comparable anomaly in European languages: “the Praeterperfect of this verb corresponds in its irregularity with the Latin, the French, English, &c., as it has not the least resemblance to any other tense.”29

  • 30 Sylvain Auroux, La révolution technologique de la grammatisation, Liège, Mardaga, 1994.
  • 31 op. cit., p. 2.
  • 32 Contrary to the Hindustani described by Gilchrist a few years later, dialect that will be the basis (...)

19But the best way to give credit to Moors, in terms of grammatical strategy, is for Hadley to explain the language according to the Latin model, or rather, to frame it within the “grammatisation” process described by Auroux.30 For this, Hadley turns to the “opaque loanword,” following Auroux’s terminology, that is, directly transposing the Latin terminology on Moors: “the cases are the Nominative, Genitive, Dative, Accusative, Vocative and Ablative.”31 This is particularly interesting, since Moors has no proper declension (see fig. 1),32 as Hadley himself concedes:

  • 33 op. cit., p. 3.

There is no variation through the cases in either substantive or adjective in the termination. They are only denoted by adding the article after the substantive.33

Fig. 1: Grammatical remarks, p. 3

Fig. 1: Grammatical remarks, p. 3
  • 34 It is true that the Sanskrit declension, being comparable to the Greek or the Latin model, has some (...)

20Thus, instead of a proper declension, as Hadley suggests it by borrowing from the Latin model, Moors displays a postpositional model: the function of the word in the sentence is not marked by a morphological modification of the substantive and adjective (such as, in Latin, Nom. magnus domus vs. Acc. magnum domum), but by adding a postposition (which Hadley calls “article”: kau, ko, sa) after the morphologically unchanged substantive and adjective (burrah ghurr). Hadley’s use of the Latin frame to account for Moors is therefore quite unsuccessful, since the notion of declension does not fit this language.34

  • 35 Marius Lavency, USUS. Grammaire latine. Description du latin classique en vue de la lecture des aut (...)

21Another notable feature of this “disposition” which for Hadley was the core of his project, lies in the organization of the verbal system: here again, the influence of Latin, or rather, the temptation to make it fit into the Latin model, is very clear. It is generally admitted that the Latin verbal system falls into 5 categories, according to their vocalic ending and their ensuing radical35: amāre (rad. amā), monēre (rad. monē), agĕre (rad. ag-), audīre (rad. audī), capĕre (rad. capĭ); four have a vocalic radical, and one, agĕre, a consonantic one (ag-). When Hadley comes to the verbal system of Moors he states, significantly:

  • 36 op. cit., p. 11.

The conjugations are five, the first having the imperative in a, the second in ee, the third in ou, the fourth in o, the fifth in a consonant.36

22That is, a model directly drawn from the Latin one. It becomes clear when considering the description and paradigm Hadley gives for each category of verb:

Table 1

1st, having the imperative ending in a, preterit in e,ah

2nd, having the imperative ending in ee, preterit in e,ah

3rd having the imperative ending in ou, preterit in i,ah

4th having the imperative ending in o, preterit in ah

5th having the imperative ending in a consonant, preterit in ah.

23If we assemble the first person of the five paradigms, we end up with the following table:

Table 2

Conjugation number

Imperative

First person

Translation

1st

Da

Hum data hy

I give

2nd

Pee

Hum peetah hy

I drink

3rd

Lou

Hum louta hy

I take

4th

Do

Hum dota hy

I wash

5th

Dur

Hum durta hy

I run

  • 37 The presence, or not, of a final -h (as the perception of the sound -h in general) is very fluctuan (...)

24In other words, the conjugation of the 1st person is strictly the same throughout the five categories: imperative-ta(h) hy.37 Thus, the neat distinction between five categories of radicals (a, ee, ou, o, consonant) falls short in practice, and has no proper relevance. However, even if this model is superfluous, it has the huge advantage, for Hadley and his reader, to allow Moors to be grammatized, according to Auroux’s terminology, and to be admitted as a proper language, including proper verbal tables.

25There is more. The choice of disposition in tables makes the Moors verbal system even more similar to the Latin model, in the way it is distributed in a complete 6-person conjugation… even though that disposition, again, is totally fruitless. Indeed, closer scrutiny reveals that the conjugation features identical verbal forms for the six persons, without distinction of gender or number; only the pronoun changes (see fig. 2).

Fig. 2, from Grammatical remarks, p. 19

Fig. 2, from Grammatical remarks, p. 19
  • 38 Hadley’s verbal system is all the more shaky that it is built on the (few) exceptions of that langu (...)

26This distribution in tables is reproduced for the six other tenses labeled by Hadley (praeterperfect, praeterimperfect, praeterpluperfect, future, future perfect), and for the four other conjugations throughout: the pronouns vary, but the verbal form remains the same all along the conjugation. What is remarkable is that the total amounts to 10 pages of tables, each of which could easily have been summarized in a line (i.e.: “hum / tum / oo,ah… data hy”). Hadley, in other instances, does sum up, as we will see later on with vector verbs. But instead, Hadley’s choice to fully develop the conjugation table makes it resemble a familiar Latin conjugation table. It is a way for Hadley to prove, visually, that he did “reduce the most barbarous language to as regular a system as it will admit of.” Those conjugation tables, however redundant in form, are the key that enables Hadley to declare Moors a proper language.38

27Faithful to the agenda exposed in his preface, Hadley made his best to “reduce” Moors to a grammatical system. This meant, according to the grammatical perspective of his times, converting it from the generalized anomaly of a jargon to the omalos of a language that could be compared to a European one, and understood in a similar way. Ironically, though, this ended up in generating unnecessary copious verbal tables. The imposition of the Latin model powerfully contributed to complexifying the language. In that sense, and very paradoxically, his classifying approach led him to “re-anomalize” the language, that is, to take it out of the reassuring regularity and categories he was trying to fit it in.

  • 39 op. cit., p. 12.

28This created anomaly takes several forms under Hadley’s analysis: the underlined randomness of grammatical phenomena, the labelling of the language as deficient, lacking proper categories, or the confusion arising from an exceedingly complex and undefined terminology, falling into the very “great abstruseness” Hadley himself seeks to avoid.39

Random idiom: the convenience of “sometimes”

  • 40 ibid., p. 1.

29This feeling of anomaly is undoubtedly what first strikes the reader upon opening the Grammatical remarks. They open with the statement: “the distinction between the substantive and adjective seems to be very vague.”40 Apart from being wrong technically speaking (there is nothing “vague” about the distinction between substantives and adjectives), this information is quite destabilizing as a first approach to the language, due to the uncertainty it conveys. Furthermore, Hadley himself “seems to be very vague” about it: “many substantives differ from the adjective by … the termination -ee… but this termination is more peculiar to qualities, arts, and science.” Although this statement is correct strictly speaking, it is misleading and unhelpful at this stage of learning. By pointing at this aspect of substantive derivation (gurm, hot, gurmee, heat), which Hadley perceives to be unclear, his analysis is imprecise from the very first line.

30This is a recurring feature of Hadley’s grammar, mostly visible with its use of “sometimes.” That word is used in several instances, and each time, creating more confusion than explanation. In fact, “sometimes” functions in Hadley as a way to take his “merely practical” description to the extreme, and to avoid adequate explanations for constructions he is not comfortable with:

  • 41 ibid., p. 1.

Beside the above tenses, by way of note observe the following, which occur sometimes; Marta hoga, will be beating, i.e. must be beating... But these are seldom used, and very vague, therefore not placed in the declensions above – which shall suffice.41

31From an alleged occasional use (“sometimes”), Hadley shifts to its scarcity (“seldom”), to conclude, again, by characterizing it as “very vague.” The most striking and disconcerting use of “sometimes.” for a European reader, certainly occurs when Hadley is dealing with gender:

  • 42 ibid., p. 29.

sometimes a distinction is made in the gender by the termination, as Dourta, he is running; Dourtee, she is running; but this little attended to.42

  • 43 See for instance Jean Clément, Parlons Bengali. Langue et culture, Paris, L’Harmattan, 1994, p. 68- (...)
  • 44 Earlier on, p. 2, Hadley notes that along this “barren and undeterminate mode of speech,” exists a (...)

32How can a grammatical gender marker be “little attended to”? This can be explained by the fact that Hadley in fact describes a Moors marked by several Eastern languages or dialects (Hindoustani, Bengali, Bhojpuri) functioning with several registers (Hindi, Urdu) which may or may not express gender marks.43 However, the fact that he should not state it, and deal with this peculiarity with the convenience of “sometimes,” leaves the reader with a seemingly arbitrary language.44

33Another expedient and cursory “sometimes” appears in Hadley’s account on the preterit:

  • 45 op. cit., p. 15.

sometimes they will preserve the imperative and add the preterits of Jaouna, to go, Kurna, to do, Dona, to give, &c. as Gul, melt; Gulgiah, gone melted45

34With that “sometimes,” and a couple of uncommented examples, Hadley deals with a critical aspect of the language, labeled since as “vector verbs.” These semi-auxiliaries are extremely frequent and idiomatical in Northern Hindi dialects, and used to convey a remarkable variety of meanings, as Annie Montaut explains:

  • 46 Annie Montaut, “Grammaticalization in standard Hindi/Urdu and Hindi dialects” in Walter Bisang & An (...)

the modern standard Hindi verb, jānā ‘go’, emphasizes, with transitive verbs, a sudden, uncontrolled action (pī gayā [drink went] ‘he drunk hastily, gulped’). A number of supplementary meanings are conveyed by the various verbs used as vectors … le ‘take’ and de ‘give’ direct the process towards… or away from the subject, whereas ḍāl ‘throw’ emphasizes a process performed with suddenness, and mār ‘strike’, a process performed in a brutal or rude way. The verb “sit” is added to transitive verbs for conveying a negative judgement on the process as extremely inappropriate or foolish: tum kyā samajh bait̩he? “what did you imagine, what did you crazily misunderstood ?”46

  • 47 Other examples of the blurring effect of “sometimes” in Hadley, concealing a widespread (if not sys (...)

35As we see, vector verbs cover a fascinating panel of use and meanings: it would surely be unfair to expect from Hadley, in his deliberately concise book, a full account on them. However, even by the “practical” standards he claims, the glimpse he proposes is clearly insufficient, as he only notices it appears “sometimes,” leaving open the question of its frequency, use and regularity.47

“Speaking with grammatical accuracy”

  • 48 op. cit., p. 29.

36This is all the more remarkable that Hadley, in spite of his assertion to be “merely practical,”48 sometimes abruptly dives into the very philological considerations he resents:

  • 49 Ibid., p. 12.

those very nice distinctions… which are better omitted than inserted, as they create perplexity, and by their great abstruseness discourage the learner from… what is absolutely necessary.49

37Hadley’s own unexpected “abstruseness” comes in two forms: engaging in philological discussions on the relevance of existing grammatical categories, and creating new subdivisions – often even more confusing.

38Hadley’s remarks are noticeable as they seem to spring from his very reflexion on the disposition and on the relevant terminology, for instance:

  • 50 op. cit. p. 13.

it is usual in the Grammar to place the praeterperfect after the present. Perhaps this tense was esteemed a variation from the present, in which case it should have been called a present perfect; but as the action is passed, it is certainly with greater propriety called the praeter imperfect; nor is there any thing to direct a conjecture why the tenses are placed in the order they are, for the true order of time is past, present, and to come, according to which the preterit should stand first.50

39Hadley’s concern to pay attention to the “true order” of things is significant, I believe, for two reasons. First, because it offers a nice glimpse on the intellectual context of grammatical works at the end of the 18th century, as mentioned beforehand, namely the belief that language, and the class of words that stand for it in a grammar, should reflect a pre-existing, metaphysical order. Second, it is a major lead to what I would call Hadley’s persona grammatica, that is to say, how he presents himself through the choices he makes in his work: here, and in the next philological digressions, he seems to be more concerned with reaching a supposed truth, an underlying universal reality (“the true order”) than with the pedagogical efficiency, that could be expected from a learning tool such as a grammar.

40The preceding reflexion does not, however, hamper on the traditional order of presentation: Hadley questions it but follows it all the same, and places past tense after present tense. But he does go a step further when commenting and dealing with the cases of the nominal declension. He is indeed particularly unhappy with the inadequacy he suspects between the name of the cases and the the semantic role they map in the sentence – revealing a rather literal interpretation on that matter:

  • 51 ibid. p. 5-6.

it is certainly absurd to call any thing Dative which does not imply a donation […] Here we may take notice (as before with the Dative) … that it is absurd to call any thing Ablative but what implies privation. And it may not be amiss to observe on the seeming misapplication of the words Dative and Ablative. Dative implies donation, Ablative privation, yet notwithstanding the words are used in a very different or even opposite sense, they still are made to retain that name. Accurrit mihi quidem, &c. How can mihi be here called Dative? In ore ejus, pro aris & focis, cum equis. How can these prepositions be said to govern an Ablative case? Should not the name of the case vary with the sense in which the word is used?51

41This is a fascinating discussion, but this dissertation on the accuracy of grammatical terminology takes us far from Moors and from the “merely practical” scope of Hadley’s. He does have a point in noticing the name of the cases do not match the span of their grammatical functions, but what is interesting is that, in an attempt to be clearer, Hadley is going to create arbitrary (and again, unexplained) subdivisions in some of the paradigms (see fig. 3).

Fig. 3: “Of pronouns substantive and adjective,” Grammatical remarks, p. 5

Fig. 3: “Of pronouns substantive and adjective,” Grammatical remarks, p. 5

42Discontent with the ambiguity of the “Dative” name, Hadley creates a “double dative”, namely, “donative” and “itinerant”. This is justified by a very laconic remark:

I have made two Datives, one of which I call Donative, the other Itinerant, because it still preserves the Dative termination of the Persian.

  • 52 Annie Montaut, relying on the examples given by Hadley, suggested me this “itinerant” dative would (...)
  • 53 op. cit., p. 14.

43If we can imagine that the “Donative Dative” fulfils the authentic (according to Hadley) function of the Dative, the reader is left to conjecture as far as the “Itinerant Dative” is concerned.52 This use of an unexplained technical terminology is repeated when Hadley mentions, in parenthesis, an enigmatic “responsive sense”53 for the verb, before shifting to another idea. Despite his claim to be straightforward, Hadley thus does engage in theoretical meta-discussions, the result being a noticeable complexification of his scheme of analysis.

“Barren and undeterminate”: a deficient language

  • 54 George Hadley, New Vocabulary, English and Hindostanic, London, Cadell, 1772, p. i.
  • 55 op. cit., p. 2.

44Hadley’s New Vocabulary, English and Hindostanic, attached to the Grammatical remarks, similarly opens with the mention of a lack in the language: “A or an: there is no such particle. See One.54 This is true, strictly speaking, but one would be equally true stating that the use of the numeral ek (one) works as a perfect substitute: Hadley, on the contrary, focus on the lack. Such remarks are interspersed in the grammar: “for a few, use dow teen, two or three, there being no such expression as a few.”55 This consideration is puzzling: if dow teen, literally “two-three”, means “a few” (which it does indeed), why should there be “no expression as a few”, as Hadley states? It appears that, in spite of his project to give Moors a proper grammatical apparatus, Hadley remains stuck to his vision of it as a “barren and undeterminate mode of speech,” and continually underlines its supposed shortcomings. In another telling instance, Hadley writes:

  • 56 op. cit., p. 10.

the degrees of comparison are very defective. The Persians have a termination to distinguish them, with a regular formation of the comparative from the positive, and the superlative from the comparative, but there is no such thing here [my emphasis]; the comparative and the superlative are made by the words more and very, as Burrah, great, Aur burrah, more great, Bhoat burrah, very great or Sub sa burrah, greater than all ; literally, great than all.56

  • 57 To be perfectly fair, Hadley does formulate a positive remark about Moors in the Preface to his New (...)

45Hadley sees the mode of comparison as “very defective,” but why? Moors is only “defective” in a certain grammatical perspective that expects a precise system, formally reproducing the distinction between positive, comparative and superlative, in other words, Persian, or more significantly, Latin and Greek. Could he have been embarrassed by his failure to transpose the Latin system, as he did elsewhere? If we indulged in grammar-fiction, we could imagine another Captain George Hadley arguing that the system of comparison in Moors is remarkably practical, since based on concise, invariable set of words – that are, besides, among the most used in the language: “more” and “very.”57

Conclusion

46In this brief presentation of Hadley’s Grammatical remarks, I argued that his claimed objective was to give a reasoned account of Moors, so as to render justice to what was looked down upon as a “jargon” full of anomalies. He conclusion is clear: “Moors does admit a grammar.” This “nomalizing project,” for Hadley, implied thinking and organizing the language through the categories of European languages, principally Latin, so as to legitimize it, and make it fit in the European grammatical framework. But those categories did not work well with Moors. Paradoxically, much of his work to smoothen the cracks, if I may say, resulted in an unnecessary complexifications and philological discussions leading to unintuitive neologisms and improvised grammatical categories. All this was bound to have an anomalizing effect from the learner’s point of view, as it emphasized the unpredictability, randomness, and significant lacunas of Moors. Canguilhem, as mentioned, defines anomaly as “a crack, a relief or a gap in the linearity of things, [that] could be a source of unquietness”. Hadley saw the gap and the unquietness in the language, but, in trying to fill up the cracks, he went over the edges, creating even more unevenness. I also pointed at cracks in Hadley’s persona grammatica, his “merely practical” approach, with a repeated wish to be practical and concise and refuse complex terminology, being at times threatened by over-complexification.

  • 58 op. cit., p. iv.

47Is that all? Canguilhem adds that the anomaly, as a “disturbance,” also reveals “the permanency of the normal state.” At the end of his military career at the head of native Indian troops, Hadley found himself emphasized the desirability of a normal and permanent (colonial) state, too. In his introductory dedication to Warren Hastings, governor-general of the presidency of Bengal, Hadley pays tribute to his “benevolence,” “generosity,” “honor,” “integrity.”58 That praise, expected as it is, contrasts with the immediately following blame of Indian language teachers and interprets (munshis):

  • 59 op. cit., p. v.

the scholar is obliged to rely on the wretched abilities of a servant who speaks English like a parrot, too ignorant to convey the true sense of a translation, too indolent to exert himself if he knows it, and too abject to correct a mistake in his master, if capable of doing it.59

48Is Moors a servant language, for servant people? When “reduc[ing] the most barbarous language” to rules and to rationality, does Hadley, metaphorically, speak of the language as of its speakers, to be reduced and tamed? It would be interesting, from this enslavement metaphor, to investigate further (especially in Hadley’s day-to-day Dialogues) into the implicit parallel Hadley drew between a wretched language and the wretched servant speakers, in need of being reduced and organized by a foreign British power.

Haut de page

Bibliographie

Primary sources

Antiquity

VARRO, On the Latin Language, Volume II: Books 8-10. Fragments, ed. and trans. Roland G. Kent, Cambridge, Loeb Classical Library, 1938.

HORACE, Satires, Epistles and Ars Poetica, ed. and trans. H.K. Fairclough, Cambridge, Loeb Classical Library, 1929.

XIXth century sources

DODWELL, Edward, MILES, James Samuel, Alphabetical list of the officers of the Indian army, 1760-1838, London, 1838.

HADLEY, George, Grammatical remarks on the practical and vulgar dialect of the Indostan, London, Cadell, 1772.

HADLEY, George, New Vocabulary, English and Hindostanic, London, Cadell, 1772.

GILCHRIST, John Bosworth, A grammar of the hindostanee language, Calcutta, Chronicle Press, 1796.

GOODWIN, Gordon, “George Hadley”, Dictionary of National Biography (vol 23), Smith, Elder & Co, 1900.

SMYTH, William Carmichael, The Hindoostanee Interpreter, London, 1824.

Scientific contemporary literature

AUROUX, Sylvain, La révolution technologique de la grammatisation, Liège, Mardaga, 1994.

BHATIA, Tej, K., A history of the Hindi grammatical tradition, New York, Brill, 1987.

CLÉMENT, Jean, Parlons Bengali. Langue et culture, Paris, L’Harmattan, 1994.

CANGUILHEM, Georges, The Normal and the Pathological, trans. Carolyn R. Fawcett & Robert S. Cohen, New York, Zone Books, 1991.

LAVENCY Marius, USUS. Grammaire latine. Description du latin classique en vue de la lecture des auteurs, Louvain-la-Neuve, Peeters, 1997.

MONTAUT, Annie, “Grammaticalization in standard Hindi/Urdu and Hindi dialects” in BISANG Walter & MALCHUKOV Andrej (eds.), Grammaticalization scenarios. Areal patterns and cross-linguistic variation, De Gruyter, 2020.

STEADMAN-JONES, Richard, Colonialism and grammatical representation: John Gilchrist and the analysis of the Hindustani Language, London, Wiley-Blackwell, 2007.

VASUNIA, Phiroze, The classics and colonial India, Oxford, Oxford University Press, “Classical Presences,” 2013.

Haut de page

Notes

1 I am deeply indebted to Annie Montaut, whose linguistic expertise on Eastern Hindi dialects gave a real twist to this article. I wish to thank warmly for her time, advice, suggestions and corrections.

2 The full title being : Grammatical remarks on the practical and vulgar dialect of the Indostan language commonly called Moors, with a vocabulary English and Moors, the spelling according to the Persian orthography, wherein are references between words resembling each other in sound and different in signification, with literal translations and explanations of the compounded words and circumlocutory expressions, for the more easy attaining the Idiom of the Language, the whole calculated for the common practice in Bengal, London, Cadell, 1772 [1st edition]. The following quotations, when unspecified, are from the 3rd (1801) edition of Hadley’s Grammatical remarks.

3 Varro, On the Latin Language, Volume II: Books 8-10. Fragments, ed. and trans. Roland G. Kent, Cambridge, Loeb Classical Library, 1938, § 23: “Greeks and Latins wrote quantity of books on the subject: some think that the rule of analogia should be followed (namely, same words being declined the same way); others think common use, or anomalia, should prevail.” (my translation)

4 Georges Canguilhem, The Normal and the Pathological, trans. Carolyn R. Fawcett & Robert S. Cohen, New York, Zone Books, 1991.

5 William Carmichael Smyth, The Hindoostanee Interpreter, London, Richardson, 1824, p. iv.

6 On this point, the Dialogues of the 3rd edition onwards are particularly instructive.

7 Richard Steadman-Jones, Colonialism and Grammatical Representation: John Gilchrist and the Analysis of the 'Hindustani' Language in the late Eighteenth and Early Nineteenth Centuries, London, Blackwell, 2007. The intellectual background given hereafter owes a lot to his fascinating book.

8 Edward Dodwell & James Samuel Miles (eds.), Alphabetical list of the officers of the Indian army, 1760-1838, London, Longman, 1838, p. 124.

9 The uncertainty on the date of his return to England makes it difficult to determine with any precision the extent to which Hadley was in contact with native speakers during the consecutive editions.

10 Gordon Goodwin, “George Hadley,” Dictionary of National Biography (vol 23), London, Smith, Elder & Co, 1900, p. 137.

11 op. cit., p. i-ii.

12 op. cit., p. iii.

13 I do not have the linguistic capacity to assess the value of this Persian grammar; however, it is obvious that the method used by Hadley was the same as in the Grammatical remarks: same pattern of analysis and notably brief format (26 pages).

14 John B. Gilchrist, A Grammar, of the Hindoostanee Language, or Part Third of Volume First, of a System of Hindoostanee Philology, Calcutta, Chronicle Press, 1796.

15 The 6th edition was not available at the British Library, where I consulted the others, nor online.

16 William Smellie, “Grammar”, in William Smellie (ed.), Encyclopaedia Britannica, Edinburgh, Bell & Macfarquhar, 1771, p. 716 (my emphasis), quoted in Richard Steadman-Jones, op. cit. (2007).

17 Sylvain Auroux, La révolution technologique de la grammatisation, Liège, Mardaga, 1994.

18 John Shore, Lord Teignmouth, The Works of Sir William Jones. With a Life of the Author, London, Stockdale and Walker, 1807, p. 24-46. The speech was given on the 2nd of February 1786 ; my emphasis.

19 As shows Phiroze Vasunia in The classics and colonial India, Oxford, Oxford University Press, “Classical Presences”, 2013, p. 33-34.

20 Up to this day, the presentation of the catalogue of the Asiatic society reflects the hierarchy that prevailed in Jones’ times (ancient and modern European languages being mixed in the same top category, followed by ancient Indian languages, then vernacular Indian languages, then peripherical Asian languages):

1. English, French, German, Greek, Latin, Portuguese, Spanish and other European Languages;

2. Pali, Prakit, Sanskrit & other ancient Indian languages;

3. Hindi, Gujrati, Bengali, (…) & other modern Indian Languages;

4. Perso-Arabic and Urdu Languages;

5. Chinese, Japanese, Burmese, (…) & other Asian languages.

21 Even if the purpose of this article is not to recreate the linguistic pedigree of Moors, we should at least mention that what Hadley describes could be qualified an Eastern pre-standardised Hindi-Urdu, with clear influences from Bengali and Eastern dialects such as Bhojpuri, both in pronunciation and grammatical features, as will appear in examples.

22 op. cit., p. vii.

23 op. cit., p. 29.

24 ibid., p. iii.

25 Horace, Satires, Epistles and Ars Poetica, ed. H.K. Fairclough, Cambridge, Loeb Classical Library, 1929; book 1, epistle 6, v. 67-68 : “Si quid nouisti rectius istis, / Candidus imperti ; si non his utere mecum.” (my translation)

26 Preface to the 4th edition of George Hadley, Grammatical Remarks (1796).

27 op. cit., p. vii.

28 ibid., p. 18.

29 ibid., p. 18.

30 Sylvain Auroux, La révolution technologique de la grammatisation, Liège, Mardaga, 1994.

31 op. cit., p. 2.

32 Contrary to the Hindustani described by Gilchrist a few years later, dialect that will be the basis for the standardization of the language, and will pave the way to standard Hindi.

33 op. cit., p. 3.

34 It is true that the Sanskrit declension, being comparable to the Greek or the Latin model, has sometimes been used to describe modern Indian languages such as Hindustani (even Hindi up to the 20th century), but this is more due to the massive Sanskrit grammatical tradition than to its relevance in Hindustani. Hadley, furthermore, is explicitly at odds with this erudite tradition in his “merely practical” endeavor: as far as I am aware of, he does not mention Sanskrit, or the Sanskrit grammatical model, a single time in the whole of his Grammatical remarks.

35 Marius Lavency, USUS. Grammaire latine. Description du latin classique en vue de la lecture des auteurs, Louvain-la-Neuve, Peeters, 1997.

36 op. cit., p. 11.

37 The presence, or not, of a final -h (as the perception of the sound -h in general) is very fluctuant in Hadley’s work, and the variation between -a / -ah should not be considered significant. Also note that what Hadley calls “imperative,” though it does also works as an imperative, should be more properly called “radical” in this case, being the basis on which verbal terminations are added in the whole verbal conjugation. “Hy,” interpreted (mistakenly ?) as an “adverb possessive” by Hadley p. 16, seems to be no other than the auxiliary (from hona, “to be”) accompanying the main verb.

38 Hadley’s verbal system is all the more shaky that it is built on the (few) exceptions of that language: the irregular verbs dana (to give), peena (to drink), louna (to take). In other words, Hadley interpreted their irregularity as the mark a verbal category in itself. The issue is that we end up with 3 categories containing only one (irregular) verb: apart from those three irregular verbs, mentioned above, all verbs follow the same pattern. Interestingly, Hadley realized his mistake when consulting Gilchrist’s grammar in 1796, and fully adopted his own much more practical verbal analysis. On that point, see Steadman-Jones, op. cit., 2007.

39 op. cit., p. 12.

40 ibid., p. 1.

41 ibid., p. 1.

42 ibid., p. 29.

43 See for instance Jean Clément, Parlons Bengali. Langue et culture, Paris, L’Harmattan, 1994, p. 68-69.

44 Earlier on, p. 2, Hadley notes that along this “barren and undeterminate mode of speech,” exists a “politer [dialect]… having terminations to distinguish the number, though such distinctions savour strong of Persian”: he is most probably thinking of Urdu.

45 op. cit., p. 15.

46 Annie Montaut, “Grammaticalization in standard Hindi/Urdu and Hindi dialects” in Walter Bisang & Andrej Malchukov (eds.), Grammaticalization scenarios. Areal patterns and cross-linguistic variation, De Gruyter, 2020.

47 Other examples of the blurring effect of “sometimes” in Hadley, concealing a widespread (if not systematic) and unexplained feature of the language, can be found in the Preface to his New Vocabulary, English and Hindostanic, London, Cadell, 1772, p. iv, vi, 91.

48 op. cit., p. 29.

49 Ibid., p. 12.

50 op. cit. p. 13.

51 ibid. p. 5-6.

52 Annie Montaut, relying on the examples given by Hadley, suggested me this “itinerant” dative would be the equivalent of an allative.

53 op. cit., p. 14.

54 George Hadley, New Vocabulary, English and Hindostanic, London, Cadell, 1772, p. i.

55 op. cit., p. 2.

56 op. cit., p. 10.

57 To be perfectly fair, Hadley does formulate a positive remark about Moors in the Preface to his New Vocabulary: “the compounded words are very strong and satisfactory: Lana is to take, Ouna is to come, Laouna is to bring, i.e. to take and come” (other examples follow for about a page).

58 op. cit., p. iv.

59 op. cit., p. v.

Haut de page

Table des illustrations

Titre Fig. 1: Grammatical remarks, p. 3
URL http://journals.openedition.org/trans/docannexe/image/4494/img-1.png
Fichier image/png, 27k
Titre Fig. 2, from Grammatical remarks, p. 19
URL http://journals.openedition.org/trans/docannexe/image/4494/img-2.png
Fichier image/png, 101k
Titre Fig. 3: “Of pronouns substantive and adjective,” Grammatical remarks, p. 5
URL http://journals.openedition.org/trans/docannexe/image/4494/img-3.png
Fichier image/png, 20k
Haut de page

Pour citer cet article

Référence électronique

Louis Nagot, « “Corrupt as this dialect is, it will not be totally useless”. Colonialism and an “anomalous” Indian language in Hadley’s Grammatical remarks »TRANS- [En ligne], 26 | 2021, mis en ligne le 26 février 2021, consulté le 14 mai 2021. URL : http://journals.openedition.org/trans/4494 ; DOI : https://doi.org/10.4000/trans.4494

Haut de page

Auteur

Louis Nagot

Agrégé de Lettres classiques, ancien élève de l'École Normale Supérieure-Ulm, Louis Nagot est depuis 2017 en thèse de Littérature comparée à l’université Sorbonne Nouvelle – Paris-3. Sa recherche, codirigée par Claudine Le Blanc et Charles Delattre, porte sur la traduction et l’appropriation des langues en contexte de domination, et sur la négociation d’espaces de résistance subalternes, dans la Grèce romaine et l’Inde britannique.

Haut de page

Droits d’auteur

Tous droits réservés

Haut de page
  • Logo Presses Sorbonne Nouvelle
  • OpenEdition Journals
Rechercher dans OpenEdition Search

Vous allez être redirigé vers OpenEdition Search