11.2 Semantic change Meaning change is everywhere, and no words are immune from it. A strik- ing example of this is the English conjunction and. At face value, it seems that this is such a simple and basic word that we would be safe in assum- ing that its meaning has been the same throughout the history of English. But this isn’t at all the case. In premodern English, and was polysemous with ‘if’, as exemplifi ed in (1): (1) And I had but one penny in the world, thou shouldst have it to buy ginger- bread. (Shakespeare, Love’s Labours Lost, Vi 71–2) The only possible reading of this sentence is ‘If I had just one penny in the world. . .’. This polysemy has been lost in modern English. But it shows that even elements of the vocabulary that one would think are conceptu- ally the most basic, and hence the least likely to shift, can change their meaning. QUESTION Before we begin exploring semantic change, ask yourself what semantic change could consist in. How might we know when a change has occurred? What evidence could we draw on? Like many other branches of linguistics, the modern study of semantics began with a largely diachronic focus, investigating meaning change. Knowledge of the history of Indo-European languages had sensitizedscholars to the extreme fl uidity of words’ meanings through time. Traditional scholarly study of ancient languages in the nineteenth cen- tury (philology) meant that the details of meaning change in European languages were well known. The availability of a long written tradition, going back in the case of Greek to well before the sixth century BC, sup- plied an enormous quantity of texts through which changes in words’ meaning could be traced. Because of this history, Indo-European lan- guages have had an overwhelming importance in the study of semantic change. The rich textual tradition of European languages and its associ- ated history of scholarship mean that studies of semantic change have traditionally relied on Indo-European evidence much more than have other domains of modern linguistics. In contrast, we are largely in the dark about long-term sense developments in the languages of oral societies, which lack the written evidence on which historical study needs to be based. Unlike sound change, which seems to be governed by regular laws of great generality which were open to ‘scientifi c’ study, meaning change has often struck investigators as chaotic and particularistic. Since changes in words’ meaning are often determined by socio-cultural factors, much meaning change is not even linguistically motivated. For instance, since the advent of modern air transport, the verb fl y can refer to travelling as a passenger in an aeroplane. This is a meaning that was obviously unavail- able before the twentieth century. But it does not necessarily correspond to any change in the sense of fl y itself: this is still arguably ‘travel through the air’. What has caused the change of meaning is arguably not anything to do with language, but simply a change in the word’s denotation. An important characteristic of semantic change is that it crucially involves polysemy (see 5.3). A word does not suddenly change from meaning A to meaning B in a single move; instead, the change happens via an intermedi- ate stage in which the word has both A and B among its meanings. Consider the French noun glace ‘ice’. In the course of the seventeenth century, it acquired the additional sense ‘ice cream/iced drink’, but this did not replace the original sense. Instead, glace had simply acquired an extra polysemous sense in addition to its original one. This is the usual case in semantic change. Meaning change most often takes the form of an addition of polyse- mous senses. The loss of the original sense is less common. In all of the changes discussed in this chapter, we will assume the presence of an inter- mediate polysemous stage, though we won’t always mention it specifi cally.