It's 'not' history

What would we do without the word ‘not’? Language depends on negation: ‘the defendant is not guilty’, ‘it’s not fair’, ‘it’s not you, it’s me’. Not can impart a subtlety of meaning to speech, as in the chastising ‘are you not home yet?’ compared with the query ‘are you home yet?’ But little do many of us realise that the word has a fascinating linguistic history, involving a pattern of change that has been echoed again and again in languages across the globe.

Now, research at the University of Cambridge has traced how and why the words used to express negation have changed in the languages of Europe and the Mediterranean over the past millennium. Led by Dr David Willis at the Department of Theoretical and Applied Linguistics, the study is the first comprehensive attempt to look for patterns across such a breadth of languages and over such a timescale. It has also involved the first systematic analysis of the history of negation in languages such as Arabic, Berber, Breton and Welsh.

Not only does the new research on the history of 'not' provide a better insight into how languages evolve, and indeed provides a marker of the stage at which a language has reached in the evolution of negation, but it also sheds light on how emphatic words such as 'literally' creep into common parlance.

Jespersen’s cycle

It was at the turn of the 20th century that a linguist named Otto Jespersen discovered a cyclical pattern in how languages express negation. Put simply, the cycle moves from a single word meaning not, placed before the verb (stage I), through a stage in which two words are used either side of the verb (stage II) and then back to a single word after the verb (stage III).

Jespersen’s idea was that at some stage the original negative word is found to be insufficient to express negation and becomes strengthened by a second, emphatic word. In due course, the emphatic word is used so frequently that it becomes the only word needed to express the negative view and the first word drops away.

Many different languages have independently gone through this cyclical evolution, and have done so at vastly different times in history. Take French, for example. To say 'I don’t know' in Old French, ne was used before the verb (as in 'Je ne sais') until about 800 years ago, when it became strengthened to the ne…pas version ('Je ne sais pas') that we recognise today. And this is largely how it has remained: “Standard French is effectively at stage II of the cycle,” explained Willis. “In everyday spoken French, people are losing the ne, but children are taught to use the full two-word form of the negator. With this conservative influence, it will be interesting to see whether the language will complete the cycle or stay indefinitely at stage II.”

English, on the other hand, had finished its cycle by the mid-14th century, moving from the Old English ne before the verb, to the strengthened ne…nawit (nothing) in Middle English, to the simplified, post-verbial not of Early Modern English. And some languages completed the cycle even earlier, such as Scandinavian, which had completed the cycle before written records began, as shown through the reconstructed history of the language. Other languages such as Spanish or Russian have yet to begin the cycle.

“Cycles of usage such as that seen in negation appear to be a normal part of language development,” continued Willis who carried out the research with Dr Anne Breitbarth and Dr Chris Lucas. “We are interested in what these cyclical developments are and what they tell us about why languages change. The results have helped us draw out a new understanding of how language contact, language acquisition, and psychological and social factors might influence the evolution of language in the past and in the future.”

Patterns and pathways

To complete the study, which was funded by the Arts and Humanities Research Council and is resulting in a series of publications including a two-volume book in 2012, vast amounts of written texts were analysed for examples of stage I, II or III expressions. Up to a thousand years of linguistic history were interrogated to observe Jespersen cycles in some languages. Fieldwork interviews were used to resolve how negation is used in spoken Arabic, where the written form is distinct and conservative.

“One of our greatest difficulties was working out when a change in the expression of negation was meant emphatically or not,” said Willis. “This can tell us at what stage the language is at. At the transition between stage I and II, a second word starts being used as a means to emphasise the negative. When it is used more frequently, the language progresses to a stage when the first word is dropped, and stage III is complete.”

For Welsh, the researchers identified the emphatic use of the negative by comparing Welsh and English versions of the Bible. “In Welsh, ni was the first negator, with dim or ddim coming into use sporadically for emphasis from the end of the 13th century. In the 16th and 17th centuries, usage went up enormously. Welsh then remained in stage II for about 200 years and then moved into stage III, dropping the use of ni around 1820.”

The team’s discovery that Welsh only began Jespersen’s cycle at the time that English had completed it helps to answer why different languages show different cycle lengths. In some cases, the researchers can point towards one language influencing the development of a nearby language. For example, the initiation of Jespersen’s cycle in North African Arabic dialects was triggered by contact with Coptic in Egypt at the end of the first millennium. But just as often, as in the case of Welsh and English, contact seems to have no effect and Jespersen’s cycle progresses independently. “We believe that stage II is generally unstable and that prescriptive pressure, as seen in French, can sometimes retard progression from stage II to stage III,” said Willis.

The influence of emphasis

One question the study has considered is how a word that is used emphatically to strengthen certain phrases becomes mainstream. The researchers believe that language acquisition by children has a role to play in this type of language change, as Willis explained: “We suggest that adults initiated using the second negator emphatically for certain verbs, usually as part of a measure phrase, as in 'It didn’t help a bit', but that children misinterpreted this, and extended use of the emphatic negator to all phrases.” The effect is called bleaching: essentially where the meaning of a word is reduced or broadened.

An example of bleaching that many will be aware of today is the linguistic misuse of the word literally (as in 'I literally ran all the way'). “Although it strictly means the literal use of a word, for some it now means that they have a strong emotional commitment to what they are saying,” explained Willis. “The word has caught the eye of prescriptivists, who dispute its overuse as an emphatic, and it has also become quite generally overtly stigmatised. This may well impede a natural linguistic development of just the same kind that we have seen time and time again for the development of negation in different languages across the whole of the last millennium.”