A History of English Negation

Gabriella Mazzon

• Publisher: Routledge
• Year: 2016
• Print ISBN: 9780582381858
• eBook ISBN: 9781317877721
• Topic: Languages & Linguistics
• Subtopic: Historical & Comparative Linguistics
• Index: Languages & Linguistics

Negation is one of the main functions in human communication.A History of English Negation is the first book to analyse English negation over the whole of its documented history, using a wide database and accessible terminology.

After an introductory chapter, the book analyses evidence from the whole sample of Old English documents available, and from several Middle English and Renaissance documents, showing that the range of forms used at any single stage is wider, and the pace of their change considerably faster, than previously commonly assumed.

The book moves on to review current formalised accounts of the situation in Modern English, tracing the changes in rules for expressing negation that have intervened since the earliest documented history of the language. Since the standard is only one variety of a language, it also surveys the means of negation used in some non-standard and dialectal varieties of English. The book concludes with a look at relatively recently born languages such as Pidgins and Creoles, to investigate the degree of naturalness of the principles that rule the expression of English negation.

Chapter 1
General and Typological Issues

The history of English negation is made up of a succession of important changes and a handful of basic continuities. The aim of this chapter is to look at this history from a distance, so to speak, that is, from a wider perspective, before examining each aspect in detail in the following chapters. This first exploration will also grant us the opportunity to compare and contrast English negation with the systems of other languages, which will enable us to assess the extent to which English stays with the mainstream of other languages in its family and /or its area in this particular respect, both synchronically and diachronically. In this first, introductory chapter I will briefly introduce the forms of English negation and their main syntactic features; I will also look at some facts from a typological perspective, and highlight some problems that standard accounts do not seem to explain satisfactorily. Before turning to this, let us review some relevant theoretical points.

Negation is a basic phenomenon of human language that is, however, capable of taking a variety of different surface forms. Since negation is a psychological, cognitive universal, linguists have been interested in the acquisition and realization of the language forms that convey it, looking for language universals or, at least, strong tendencies showing that the cognitive similarity of negation phenomena all over the world can override language-specific constraints. For a long time, there has been a tendency to reduce linguistic negation to logical negation, i.e. the kind of negation which is employed in philosophy and mathematics, and which has some properties, and is subject to some constraints, very different from those of negation in natural languages. In any case, the most widespread view was that negation is 'derived' from affirmation, in the sense that a negative sentence seems to presuppose an affirmative one, while an affirmative sentence does not carry a negative presupposition. This means that a negative sentence is often considered 'marked', but at the same time less informative than an affirmative one, mainly because negatives seem to refer to already known referents, rather than to introduce new topics or entities (Givón 1979:102–3; for a contrary opinion, see Taglicht 1985: 99–100; see also 4.3. below).

The notion of presupposition belongs to pragmatics, and the pragmatic perspective undoubtedly seems to have a lot to say about negation, since the latter is often used to convey an attitude on the part of the speaker. This seems confirmed by evidence from the acquisition of negation (see 6.1.); the negative element often appears initially and has a single form that can have various functions. For instance, an utterance like (1) can have an existential value,

• (1) No candy

or it can be a negative imperative or a simple expression of a personal attitude on the part of a child (i.e. it can be paraphrased as 'there is no candy', 'don't take away my candy', 'I don't want any candy', etc.).

The pragmatic value is reflected also in the fact that there is a frequent tendency, in several languages, to multiply the number of 'negative words' in a sentence to 'reinforce' its negativeness or to emphasize it; see example (2) from It.

• (2) Non devo nienle a nessuno
  literally: 'I not owe nothing to nobody'

Here the emphasis by means of multiple negation offers a sharp contrast with the value of negation in logic, where given a proposition p, its contrary is ~p, and ∼∼p = p (non-non p = p, or 'two negatives = positive'). By contrast with the behaviour of logical negation, several languages besides Italian use multiple negation to make negation stronger, not weaker. This is not the only aspect that the systems of negation have in common across several languages: especially within a relatively compact area like Europe, for instance, it is possible to notice that the rules of negation are sometimes shared, or that the differences are not so deep. Similarly, several European languages show resemblances in the way their systems of negation have evolved over time, and seem to present analogous tendencies as regards change within these systems. Some of these tendencies will be mentioned below; let us turn now to some distinctions between various types of negation.

A first distinction that has enjoyed some popularity is that between sentence negation and constituent negation, sometimes also indicated as nexal vs. special negation. This is a fundamental distinction in some ways, because it indicates which portion of an utterance is negated, it indicates, that is, what portion of the sentence is under the scope of the negation (i.e. within its logical domain). The latter type of negation is often used with contrastive value; this applies also to sentence negation, but in this case the whole predication, and not just one phrase or one word, will be under the scope of negation. This is an important conceptual distinction, but it has consequences on the syntactic level, especially in a language like Modern English, where sentence negation is almost always attached to an auxiliary verb. Cf. the difference between (3a, 3b) vs. (3c, 3d), where negative forms are capitalized when pronounced with contrastive stress:

• (3a) I didN'T go to Rome.
• (3b) I didN'T go to Rome, but to Vienna.
• (3c) I went to Vienna, NOT to Rome.
• (3d) I went NOT to Rome, but to Vienna.

In these examples, the structural differences between the two types of negation is quite apparent, and they stand out even more when the negative elements that signal the two types of negation are not formally identical, as it was in Old English:

• (4) Heo ne beon Godæs.
  'They are not Gods.'
• (5) heo ne beon na þreo Godæs, ac is an Almihtig God ...¹
  'There are not three Gods, but one Almighty God ...'

Quite often, contrastive negation can be a special realization of a particular type of negation, i.e. the so-called metalinguistic negation. This is to be distinguished from ordinary negation because it refers not to the quality of a proposition of being true/false, but to the possibility of an utterance of being acceptable in a context; it acts as a sort of contradiction or correction of the form or content of an utterance, rather than a comment or opinion on a state of affairs (Horn 1989: 363–402, 421–5). This kind of negation is normally accompanied by special intonation patterns or other devices to signal that it is a non-literal negation, and that the normal presuppositions that hold in normal cases of negation are suspended here. Let us look at some examples (6–9, again with capitals signalling heavy stress on the relevant words):

• (6) I don't like him.
• (7) (A: -You seem to like Jack a lot...)
  B: -Jack? I don't LIKE him, I LOVE him!
• (8) He didn't call the [pólis], he called the [polís] (uttered by a teacher of English).
• (9) He hasn't got three children.

Example (8) illustrates that it is the form and not the meaning of a previous utterance that is 'corrected' by this kind of negation, while (6–7) show the reversal of presupposition: 'normal' negation tends to be interpreted as 'less than', while metalinguistic negation can imply 'more than': for example, (9) does literally mean that the number of children is not exactly three (it may be two, or four), but is normally interpreted as 'less than three'. In metalinguistic negation, this kind of presupposition is contradicted, and the point on any scalar value that is intended may well be higher than what is literally indicated, rather than lower.

It is interesting to notice that metalinguistic negation seems to possess a number of special properties, among them the constraint whereby the elements juxtaposed have to belong to the same semantic range; keeping in mind example (7), note the unacceptability of (10) if the negation in the first clause is interpreted as metalinguistic:

• (10) *I don't LIKE him, I'm going to KILL him.

Note also that the latter kind of negation cannot be conveyed by affixal negation (11) and cannot co-occur with Negative Polarity Items (NPIs), as shown in (12):²

• (11a) The Queen of England is NOT happy /UNhappy.
• (11b) The Queen of England is NOT HAPPY, she's ecstatic.
• (11c) *The Queen of England is UNHAPPY, she's ecstatic.
• (12a) I don't give a damn if you live or die.
• (12b) *I don't GIVE A DAMN IF YOU LIVE OR DIE, I care a lot about you.

The case of metalinguistc negation represents a case in which the negative element is bound to a specific interpretation. There are cases which look like the opposite, since there appears a negator which is semantically empty, and must not be interpreted literally, unless the meaning is to be reversed. This phenomenon is called expletive or paratactic negation, and is often treated as an extreme form of Negative Concord (which is in turn examined below), since it extends over a clause boundary. This kind of negation occurs only in environments subject to some constraints: in counterfactual sentences; with verbs in the main clause expressing fear, prohibition, caution; in comparative constructions and in clauses expressing 'before' or 'after'; the dependent clause is often in the subjunctive mood. As for its communicative value, expletive negation is totally rhetorical, as can be seen from the examples below:

• (13) Timeo ne veniat.
• (14) Je crains qu'il ne vient.

Latin and French are both languages which have expletive negation; in both examples, the sentence means 'I'm afraid he will come'. Thus, the negator is totally redundant, and can be misleading if taken literally, giving as a result 'I'm afraid he won't come', i.e. the opposite of the meaning intended. In English, it seems that this rhetorical device was never very popular (despite the influence on writing styles of both Latin and French), but it is possible to find sporadic examples in older documents, and even today there are some occurrences in non-standard varieties:

• (15) I wouldn't be surprised if it didn't turn to rain. [= ... if it rains]

One of the main typological differences between European languages as regards negation concerns the position of the main negator in the sentence. This element extends its scope over the whole predication, but it is significant that its position can vary, in fact tends to vary, over time. There are general trends in language which are connected to communicative needs, and some of these have a bearing on the position of negation. These trends include phenomena like the Neg-First Principle. This was formulated by Horn (1989: 292–3), but its existence had already been noted by Jespersen (1917, 1924), who claimed that negation tends to be attracted leftwards and generally to precede the words over which it has scope. This tendency would be for reasons of communicative efficiency, i.e. 'to put the negative word or element as early as possible, so as to leave no doubt in the mind of the hearer as to the purport of what is said' (Jespersen 1924: 297), and it could be responsible for phenomena such as the predominance of negative prefixes over suffixes.

Another of these general trends is the so-called End-Weight Principle, i.e. the tendency to concentrate communicatively significant elements towards the second part of a clause: Given (already known) elements tend to come first, while New information comes later. As regards its consequences on the placement of the negative element within a clause, this latter principle is in competition with the former, and this competitive dynamics was held responsible by Jespersen for the diachronic process undergone by several languages and often called the negative cycle.

The first formulation of the 'cycle' goes back to Jespersen's (1917) detailed study of negation in several languages. The cycle consists of successive phases of weakening and reinforcing of the formal means of expressing main sentential negation. The initial phase of the cycle is typically represented by a single pre-verbal negator (Neg + V = Stage NEG1), which comes later to be optionally reinforced, to convey emphasis or for other communicative reasons, by other expressions placed post-verbally (possibly due to temporary predominance of the End-Weight Principle). These new negators may well not be inherently negative originally, but acquire a negative meaning along with the new reinforcing function. The next stage is attained when the reinforcement is no longer optional but becomes obligatory (Neg + V + Neg = Stage NEG2), and the following stage is the dropping of the original pre-verbal negator, after progressive phonological and pragmatic weakening, so that negation comes to be expressed by a single post-verbal element (V + Neg = Stage NEG3). At this point, the Neg-First Principle intervenes and the single remaining negator is again attracted to pre-verbal position, which brings the cycle to completion. One of the clearest examples of the negative cycle among European languages is constituted by French, with its ne > ne ... pas > (ne) ... pas pattern; English is also an often-quoted example, but we will see that the outcome is not so clearcut in its case.

The 'cycle' involves various phenomena which are not easy to explain, for instance the second stage (Neg + V + Neg) presents what formal grammar calls a discontinuous constituent, i.e. two elements belonging to the same slot in the underlying or formal structure of the sentence are separated by another element, the verb, which is a different constituent altogether. The fact that two elements which are not contiguous in the linear structure nevertheless form part of the same constituent may not be difficult to grasp, but may pose severe problems to certain types of formal grammars, e.g. those trying to represent the structure of language strings by means of 'trees'. Also relevant is the fact that the post-verbal forms that acquire negative meaning do so by way of grammaticalization, i.e. they often are originally semantically full elements that indicate small quantities or entities (NPIs like Lat. passum, guttam 'a step, a drop' giving Fr. pas, goutte) and that are used as reinforcers of negation, and then lose their lexical status to become purely grammatical markers of negation. Old English had similar expressions, but it is interesting to notice that only inherently negative words came to be used as reinforcers and thus became negators proper. These matters will be taken up again in a more formalized perspective in Chapter 6.

The historical development of English negation, as will be seen in the chapters that follow, shows several of the phenomena introduced so far, but also presents peculiarities. In OE, simple sentential negation was often expressed by means of pre-verbal ne, though this was not the exclusive or totally predominant expression of this type of negation, as some standard accounts seem to claim (see Chapter 2 for further qualification of this statement). In accordance with the 'rules' of the cycle, ne is found more and more often reinforced by other negative words, mostly indefinites and adverbs, and notably by nawiht ('nothing'), later variously contracted in noght, nawt and finally not. This reinforcement became obligatory and, at the same time, pre-verbal ne was lost (a stage that French is reaching only nowadays), partly, as often maintained, because of its low phonetic salience, which encouraged gradual weakening, but partly also because of a progressive loss of communicative salience once not was established firmly in the structure, and in a position that increases communicative dynamism (according to the already mentioned End-Weight Principle).

At this point, we are left with a single post-verbal negator; simple clauses now follow the pattern that, at ...

Table of contents

• Cover
• Half Title
• Title
• Copyright
• Contents
• List of Tables and Figures
• List of Abbreviations and Conventions
• Preface
• Acknowledgements
• Chapter 1. General and Typological Issues
• Chapter 2. Negation in Old English and Early Middle English
• Chapter 3. The Middle Ages and Early Modern English
• Chapter 4. Present-Day English Negation
• Chapter 5. Varieties of English
• Chapter 6. Further Theoretical Considerations and Conclusions
• Appendix I: Sources for Text Analysis
• Appendix II: Negative Forms
• References
• Index