Some linguists believe that Homo erectus, some 1½ million years ago, already had a language that was rather similar to what I have called the ‘me Tarzan’ stage.
==========
So even if the brain of Homo erectus had the capacity for something resembling human language, there is no compelling reason to assume that the capacity was ever realized. The arguments for an early date are therefore fairly shaky.
==========
According to some researchers, hominids prior to Homo sapiens could not, for instance, produce the vowel i {ee}. But ultimately, this does not say very much, since by all accounts, et es perfectle pesseble to have a thoroughle respectable language wethout the vowel i.
==========
Small tribes with stone-age technology speak languages with structures that sometimes make Latin and Greek seem like child’s play.
==========
‘When it comes to linguistic form, Plato walks with the Macedonian swineherd, Confucius with the head-hunting savage of Assam,’ as the American linguist Edward Sapir once declared. (Later on, I shall even argue that some aspects of language tend to be more complex in simpler societies.)
==========
One of the most original theories was surely that of Frenchman Jean-Pierre Brisset, who in 1900 demonstrated how human language (that is to say, French) developed directly from the croaking of frogs. One day, as Brisset was observing frogs in a pond, one of them looked him straight in the eye and croaked ‘coac’. After some deliberation, Brisset realized that what the frog was saying was simply an abbreviated version of the question ‘quoi que tu dis?’ He thus proceeded to derive the whole of language from permutations and combinations of ‘coac coac’.
==========
Linguists of the ‘innatist’ school believe that some of the fundamental rules of grammar are biologically pre-wired, and that babies’ brains are already equipped with a specific tool-kit for handling complex grammatical structures, so that they do not need to learn these structures when they acquire their mother-tongue.
==========
These apes grow up in families, sometimes side by side with human babies of the same age, but the orang-utans never end up learning Indonesian. And despite popular myth, not even chimpanzees can learn a human language, although some chimpanzees in captivity have developed remarkable communicative skills.
==========
In the early 1980s a pygmy chimpanzee (or bonobo) called Kanzi made history by becoming the first ape to learn to communicate with humans without formal training. The baby Kanzi, born at the Language Research Center of the Georgia State University, Atlanta, used to play by his mother’s side during her training sessions, when researchers tried (rather unsuccessfully) to teach her to communicate by pointing at picture-symbols. The trainers ignored the baby because they thought he was still too young to learn, but unbeknownst to them, Kanzi was taking in more than his mother ever did, and as he grew up he went on to develop cognitive and communicative skills far surpassing any other ape before. As an adult, he is reported to be able to use over 200 different symbols, and to understand as many as 500 spoken words and even some very simple sentences. Yet although this Einstein of the chimp world has shown that apes can communicate far more intelligently than had ever been thought possible, and thus forced us to concede something of our splendid cognitive isolation, even Kanzi cannot string symbols together in anything resembling the complexity of a human language.
==========
Chomsky and other linguists have argued that children manage to acquire language from scanty and insufficient evidence (in other words, from ‘poor stimulus’). After all, most children are not taught their mother-tongue systematically, and even more significantly, they are not exposed to ‘negative evidence’: their attention is rarely drawn to incorrect or ungrammatical sentences. And yet, not only do children manage to acquire the rules of their language, but there is a variety of errors that they don’t seem to make to start with.
==========
The debate is known in linguistic circles as the ‘poverty of stimulus’ argument, and revolves around a perennial miracle: the speech that comes out of the mouth of babes and sucklings. How is it that children manage to acquire language with apparently so little difficulty?
==========
Chomsky claimed that since children could never have worked out all the correct rules purely from the evidence they were exposed to, the only plausible explanation for their remarkable success is that some rules of grammar were already hard-wired in their brain, and so they never had to learn them in the first place.
==========
it seems implausible that specific features in the structure of language are pre-wired in the brain if they could have developed only ‘recently’ (say within the last 100,000 years), and if their existence can be accredited to the natural forces of change that are steering languages even today. In other words, the details of language’s structure which can be put down to cultural evolution need not be coded in the genes (although the ability to learn and handle them must of course be innate). It thus seems implausible to me that the specifics of anything more sophisticated than the ‘me Tarzan’ stage, to which we’ll return in Chapter 7, need to be pre-wired.
==========
… the definition of ‘foreign aid’ as the transfer of money ‘from poor people in rich countries to rich people in poor countries
==========
And as if to prove the point, Chaucer’s (and Wycliffe’s) English – from just over half a millennium ago – already looks ‘wonder nyce and straunge’. But go back a full ‘thousand yeer’, and Ælfric’s English is not merely strange – it sounds like double Dutch. Within a span of only about thirty generations, ‘English’ has undergone such a thorough overhaul that what is supposed to be one and the same language is barely recognizable
==========
Perhaps the most surprising feature of Ælfric’s English is that, like Latin, it had a complex case and gender system,
==========
Here is a short excerpt from the Book of Genesis, which relates the story of the Flood: English around 2000 The Lord regretted having made humankind on the earth … So the Lord said: ‘I will wipe the human beings I have created off the face of the earth, people together with animals and reptiles and birds of the air, because I regret having made them’… And God said to Noah ‘… Make yourself an ark of gopher wood … and cover it inside and out with pitch. For my part, I am going to bring a flood of waters on the earth, to destroy all flesh in which there is the breath of life.’ From modern, albeit literary English, let’s now jump four centuries back in time, to the year 1604, when King James I, newly installed on the throne of England, and desiring to soothe the religious strife that had plagued the realm for more than a century, commissioned the best scholars in the land to produce a translation of the Bible into the English of the day. Forty-seven scholars laboured on the text for the suitably biblical period of seven years, until finally, in 1611, what has come to be known as the King James Version was published: English around 1600 (King James Version) It repented the Lord that he had made man on the earth … And the Lord said: ‘I will destroy man whom I haue created from the face of the earth, both man, and beast, and the creeping thing, and the foules of the aire, for it repenteth me that I haue made them.’ And God said vnto Noah: ‘Make thee an arke of gopher wood … and [thou] shalt pitch it within and without with pitch. And behold, I, euen I, doe bring a flood of waters vpon the earth, to destroy all flesh wherein is the breath of life.’
==========
English around 1400 (Wycliffe Bible) It forthou3t* him that he had made man in erthe. ‘I shal do awey,’ he seith, ‘man, whom I made of nou3t, fro the face of the erthe, fro man vnto thingis hauynge soule, fro crepynge beest vnto fowles of heuen; forsothe it othenkith me to haue maad hem.’ He seide to Noe: ‘Make to thee an ark of planed trees; and with ynne and with oute thow shal di3ten it with glew. Se, I shal lede to watres of a flood vpon the erthe, and I shal slee al flehs in the which spiryt of lijf is.’
==========
English around 1000 (Translation of Ælfric) Gode ofðuhte* ða ðæt he mann geworhte ofer eorðan … And cwæð: ‘Ic adylgie ðone man, ðe ic gesceop, fram ðære eorðan ansyne, fram ðam men oð ða nytenu, fram ðam slincendum oð ða fugelas: me ofðingð soðlice ðæt ic hi worhte.’ And God cwæð ða to Noe: ‘Wyrc ðe nu ane arc of aheawenum bordum and clæmst wiðinnan and wiðutan mid tyrwan. Efne ic gebringe flodes wæteru ofer eorðan, ðæt ic ofslea eal flæsc on ðam ðe is lifes gast.’
==========
Ic is in fact one and the same word as our modern ‘I’, and only looks so different because its pronunciation has changed so much. In the tenth century, ic was pronounced something like {itch}, but by 1400 the final {tch} had disappeared, and the word came to be pronounced {ee} (as in ‘bee’), and thus to be spelt as just ‘I’. In the writing system, ‘I’ has looked the same ever since, but the actual pronunciation of ‘I’ has continued to meander. During the fifteenth century, there was an upheaval in the pronunciation of many English vowels, which linguists call ‘The Great English Vowel Shift’. As a part of this shift, all long {ee} vowels turned into {ay} (as in modern ‘day’), so by the sixteenth century, ‘I’ came to be pronounced {ay}. And by the eighteenth century, {ay} changed further into the modern pronunciation {eye}.
==========
For cultural reasons that are extraneous to spoken language itself, the system of spelling we use today has remained pretty much frozen for at least 400 years, even though the pronunciation continued to drift during this time. So if one compares the King James passage with the modern translation, one could easily fall under the impression
==========
For cultural reasons that are extraneous to spoken language itself, the system of spelling we use today has remained pretty much frozen for at least 400 years, even though the pronunciation continued to drift during this time.
==========
An American linguist once quipped that ‘a language is a dialect with an army and a navy’, and his point is illustrated by recent cases such as Serbian and Croatian, which before the break-up of the former Yugoslavia were regarded as dialects of one language, Serbo-Croatian, but afterwards were suddenly proclaimed to be different languages.
==========
The biblical assertion that there was a single primordial language is not, in itself, unlikely, for it is quite possible that there was originally only one language, spoken somewhere in Eastern Africa, perhaps 100,000 years ago.
==========
==========
Imagine two public buildings with an overgrown field lying directly between them. The only road connecting the buildings winds its way lengthily around the field, so people who have to walk from one building to the other start crossing the field as a short-cut.
==========
As more and more people cross the field, more and more vegetation is trampled, so that eventually the track turns into a nice clear footpath. The point is that no one in particular created this footpath, and no one in particular even intended to. The path did not emerge from some project of landscape design, but from the accumulated spontaneous actions of the short-cutters, who were each following their own selfish motives in taking the easiest and quickest route. Changes in language come about in a rather similar fashion, through the accumulation of unintended actions.
==========
But in essence, the motives for change can be encapsulated in the triad economy, expressiveness and analogy.
==========
The third motive for change, analogy, is shorthand for the mind’s craving for order, the instinctive need of speakers to find regularity in language. The effects of analogy are most conspicuous in the errors of young children, as in ‘I goed’ or ‘two foots’, which are simply attempts to introduce regularity to areas of the language that happen to be quite disorganized. Many such ‘errors’ are corrected as children grow up, but some innovations do catch on. In the past, for example, there were many more irregular plural nouns in English: one bōc (book), many bēc; one hand, two hend; one eye, two eyn; one cow, many kine.
==========
Expressiveness relates to speakers’ attempts to achieve greater effect for their utterances and extend their range of meaning. One area where we are particularly expressive is in saying ‘no’.
==========
Economy refers to the tendency to save effort, and is behind the shortcuts speakers often take in pronunciation.
==========
But gradually, ‘errors’ like ‘hands’ crept in by analogy on the regular -s plural pattern. So bēc was replaced by the ‘incorrect’ bokes (books) during the thirteenth century, eyn was replaced by eyes in the fourteenth century, kine by cows in the sixteenth
==========
Suppose you see two elderly ladies coming out of the theatre, and from their animated conversation you catch the word ‘wicked!’ Of course, you would automatically assume that the ladies thoroughly disapproved of the performance. But if behind the two ladies there were two teenage girls, and one said to the other ‘wicked!’ you would probably interpret her mood very differently. In a hundred years’ time, when the original meaning of ‘wicked’ has all but been forgotten, people may wonder how it was ever possible for a word meaning ‘evil’ to change its sense to ‘wonderful’ so quickly. But for us who are in the midst of it, the variation does not seem to cause too much angst.
==========
Sometimes, the contradictory meanings even rub shoulders for centuries: a word like ‘fast’, which started off meaning something like ‘secure’, or ‘not moving at all’, later developed the contrary sense ‘moving quickly’. Both meanings have survived until this very day, but we still manage to get along all right, apparently without too many serious mishaps.
==========
To modern ears the change from ‘it likes me’ to ‘I like it’ seems unlikely, but from the perspective of the seventeenth century it was just another case of synchronic variation. Shakespeare may have used ‘like’ in the older sense (‘the musician likes me not’), but in fact, he also uses ‘like’ in the modern – flipped – meaning. In Othello, for instance, the musicians are told: ‘the general so likes your music, that he desires you, for love’s sake, to make no more noise with it.’
==========
But it would be rather unfair to blame it all on irrational nostalgia, since there is a much more serious reason why so many people think that language is constantly decaying. This reason is quite simply that … decay is indeed a pervasive type of change in language, and what is more, it is the aspect of change that is by far the most easily observable to the naked eye. The forces of destruction almost seem to leap out of the pages of practically any language’s history, but the contrary processes, the productive forces of renewal and creation, are much more difficult to spot – so difficult, in fact, that it is only in the last few decades that linguists have fully grasped their significance and have made real headway in understanding them.
==========
Originally, as it turned out, Latin flowers were perfectly regular, and so the forms were indeed flos, flosem, flosis, just as one would expect. But early on in the history of Latin, some time between the sixth and fourth centuries BC, a sound change took place, in which every (undoubled) s between two vowels turned into an r. In itself, this was an entirely regular change, and happened systematically to all eligible candidates. But as a result, an irregularity wormed its way into words like flos. The s in flosem, flosis and so on turned into r, because it was between two vowels, whereas the s in flos remained an s, because it was not. (Incidentally, the consequences of this Latin change from s to r can still be felt in English, not only in the borrowed word ‘flower’, but also in various pairs which are borrowed from different derivations of the same Latin noun. Just and jurisdiction, for instance, both go back to the Latin root jus ‘justice’, but in justus ‘just’, the
==========
Originally, as it turned out, Latin flowers were perfectly regular, and so the forms were indeed flos, flosem, flosis, just as one would expect. But early on in the history of Latin, some time between the sixth and fourth centuries BC, a sound change took place, in which every (undoubled) s between two vowels turned into an r. In itself, this was an entirely regular change, and happened systematically to all eligible candidates. But as a result, an irregularity wormed its way into words like flos. The s in flosem, flosis and so on turned into r, because it was between two vowels, whereas the s in flos remained an s, because it was not. (Incidentally, the consequences of this Latin change from s to r can still be felt in English, not only in the borrowed word ‘flower’, but also in various pairs which are borrowed from different derivations of the same Latin noun. Just and jurisdiction, for instance, both go back to the Latin root jus ‘justice’, but in justus ‘just’, the s remained unaltered, whereas in jurisdiction ‘administration of justice’, the s of the genitive case jusis was caught between two vowels, so it changed to juris. The same applies to rustic and rural, both from the noun rus ‘country’.)
==========
So the irregular flowers were actually the result of a simple sound change. But what of it? To all intents and purposes, the explanation for how the Latin s-r irregularity crept into nouns like flos was just a simple solution for a simple problem, and not even a particularly crucial one at that. Nevertheless, for me the flos-floris affair was a revelation, since it showed that even irregularities, those apparently arbitrary nuisances, are not entirely arbitrary after all.
==========
Take any irregularity (like flos-floris), and if you only trace it back far enough, it seems, it will turn out to have developed because of some change to an originally regular pattern (flos-flosis). And if this is so, then the further back one digs in time, the more regular a language should be. According to this logic, then, there must have been some Golden Age of perfection somewhere deep in the past, when languages could boast a flawless structure unblemished by irregularity.
==========
But in essence, Grimm’s law describes a wholesale erosion of sounds that took place in the Germanic branch of Indo-European. Six (out of the nine) changes are shown below. Grimm’s Law: sound changes between Proto-Indo-European and Proto-Germanic The bottom end of the diagram is by now quite familiar, since two of these changes have already featured earlier (the change from k → ch → h in this chapter, and the change from p → f, as in pisk and fisk, in the previous chapter). In fact, all the three changes at the bottom of the diagram are of a similar nature, since in all of them, a consonant called a ‘stop’ (where the air is blocked completely for a split second by the tongue or the lips) is weakened to the equivalent ‘fricative’ (where the tongue or lips don’t block the air completely but let out a small stream that makes a sound through friction).
==========
The changes at the top (g → k, d → t, b → p) also represent a kind of weakening, but here effort is spared for the vocal cords. The sounds g, d and b are called ‘voiced’ consonants, because when we produce them, the vocal cords start vibrating at the same time as the air is released from its temporary blockage, or just a split second before. The sounds k, t and p are called ‘voiceless’ consonants, because they are produced without the additional vibration of the vocal cords, but with the tongue or lips blocking the air in exactly the same way as in g, d and b. (You can check out the difference by pronouncing each of the pairs while whispering. You will hear that the distinction between the voiced and voiceless sounds almost disappears.) Of course, vibrating the vocal cords is an additional effort that can sometimes be spared, and so a voiced g turned into a voiceless k, and similarly, d turned to t, and b to p.
==========
Grimm’s law explains why the relation between English words and their cognates in non-Germanic Indo-European languages is not always immediately apparent to the untrained eye. English cold, for instance, is related to Italian gelato ‘frozen’ rather than to its false friend caldo ‘hot’. And there are many other such surprising ‘twins’, cognates such as grain and corn, which may look rather different to the casual observer, but nevertheless come via different routes from exactly the same Proto-Indo-European root.
==========
The reason why English has many such pairs is that more than a millennium after Grimm’s changes had taken their course, English started borrowing heavily from Latin and French, and thus developed a two-tier vocabulary of home-grown and borrowed words. Corn is the native English sibling, which underwent the g → k change some time after 500 BC, whereas grain was borrowed from French much later on, and so bypassed this change. As the table overleaf illustrates, there are many other such separated twins in English, such as dent(al) and tooth; can(ine) and hound; pater(nal) and father
==========
But sounds can be gouged out from the middle of a word with almost equal ease, as witnessed by the fate of the Old English hlaf-weard ‘loaf-ward(en)’ or ‘bread keeper’. Hlaf-weard, with its two words and three syllables, was shortened to hlaford, thence to laferd, then lowerd, until it finally ended up as our impoverished modern lord, with just one halfpenny-syllable to his name.
==========
In modern English, for example, words like disturbed or loved are written with -ed at the end, although they are pronounced {disturbd}, {lovd}. The reason for the extra -e is that such words were originally pronounced {disturbèd}, {lovèd} with an audible vowel at the end. At the beginning of the eighteenth century the shortened pronunciation was still rather new-fangled, and as such attracted the wrath of Jonathan Swift. In his splenetic rant of 1712, Swift had this to say about pronunciations such as lov’d and rebuk’t, rather than the correct lovèd and rebukèd: ‘By leaving out a Vowel to save a Syllable, we form so jarring a Sound, and so difficult to utter, that I have often wondred how it could ever obtain … This perpetual Disposition to shorten our Words, by retrenching the Vowels, is nothing else but a tendency to lapse into the Barbarity of those Northern Nations from whom we are descended.’
==========
Consider, for instance, what happened to the portly Latin phrase persica malus ‘Persian apple’, with its five juicy vowels and seven luscious consonants. It ended up in French as a word of just one vowel and two consonants: first, the whole second word was dropped altogether, leaving persica. Then the vowel i disappeared to give persca, which was further shortened to pesca, then to pesche, and finally pêche, ending up on English palates as a rather shrivelled ‘peach’.
==========
In fact, when it comes to shedding syllables, French is a prime example. In modern French, there are three different words that are pronounced the same way, as a rather bare and paltry {oo}: ou, où, and août. The first two have relatively moderate histories in terms of shrinking: ou ‘or’ comes from Latin aut, pronounced {out}, and here only one vowel and one consonant have disappeared. Où ‘where’ comes from Latin ubi, and once again, it has only lost one vowel and one consonant. But the third {oo}, the month of août, comes from no less a forebear than the Latin Augustus. Here, four consonants and three vowels have vanished without trace. At first sight, it might seem remarkable that the august Augustus could have ended up as a mere {oo}, but in French hands, there is nothing to it. Take Augustus, which by late Latin had already been shortened to Agustus; then drop the last syllable to get Agust. By the twelfth century, the g had also eroded away to leave {aoost}. The s was next in line for the chop, so the word came to be pronounced {aoot} and spelt août. Later, {aoot} was shortened to {oot}; and finally,
==========
Some languages are notorious for being rather consonant-heavy, and often the reason is drastic vowel loss. As a sort of comic relief during the war in the former Yugoslavia, an email (based on an article in the satirical magazine The Onion) circulated with the following report: CLINTON SENDS VOWELS TO FORMER YUGOSLAVIA City of Sjlbvdnzv and Island of Krk to be First Recipients
==========
The citizens of the stricken towns are eagerly awaiting the vowels. ‘My God. I do not think that we can last another day,’ Trszg Grzdnvc, 44, said. ‘I have six children and none of them has a name I can pronounce.
==========
As is so often the case, however, the reality is better than any fiction. The Czech phrase for ‘stick (your) finger down (your) throat’ is, very appropriately, strč prst skrz krk.
==========
The devastation that erosion metes out is perhaps most conspicuous with case-systems, which of all monumental structures seem to be most vulnerable. The fate of the case-system in the Indo-European languages is a good example. The prehistoric ancestor, Proto-Indo-European, had eight distinct cases, but only Sanskrit retained the full system, whereas in all the other daughter languages, erosion had started taking its toll even before the earliest records began. In Classical Latin, for instance, the eight distinct cases had already been knocked down to just six, and in practice, no single Latin noun had more than five distinct case endings in the singular.
==========
In the second declension, for instance, the dative and the ablative cases fused, leaving the noun with the following forms (in the singular): But even these five different endings didn’t hold out for long. The -um of the accusative was first shortened to -u, and then further to a laxer -o; the long -ō of the dative and ablative also weakened to a short -o; and the rare vocative coalesced with the nominative. So by around AD 300, only three distinct endings remained: And a few hundred years later, by the time of the earliest records in the Romance languages, only two forms had survived: nominative ann-os, and ann-o for everything else. Later on, even that difference was eliminated. In Spanish, the final -s of the nominative dropped off to give año in all cases, and in French, the whole final syllable disappeared, giving just an in all forms. So not much more than a millennium after Cicero, the majestic Latin case system had been entirely wiped out of existence.
==========
An even earlier casualty was the case system in the Germanic branch of Indo-European, which already by the third century AD had lost four of the original eight cases of Proto-Indo-European. By the time of Old English, in the tenth century, nouns were left with at most three distinct case endings in the singular and three in the plural: But not even this reduced system succeeded in standing up to the onslaughts of erosion for very long, as soon after the tenth century, the final syllables were weakened and the whole edifice started to collapse. By the fifteenth century, the system was in tatters, and only the forms ending with an s remained in any way distinct: It is this depleted system that has survived in modern English, which can only boast two distinct endings, stone and stone+s (the latter written in three different ways, stones, stone’s and stones’, in a vain attempt to talk up the number of different forms).
==========
==========
The British orientalist Sir William Jones reported this discovery to the Asiatic Society of Calcutta in February 1786, in words that were to become legendary in the history of linguistics: The Sanscrit language, whatever be its antiquity, is of a wonderful structure; more perfect than the Greek, more copious than the Latin, and more exquisitely refined than either, yet bearing to both of them a stronger affinity, both in the roots of verbs, and in the forms of grammar, than could possibly have been produced by accident; so strong, indeed, that no philologer could examine them all three, without believing them to have sprung from some common
==========
At the implausible age of twenty-one, Ferdinand de Saussure proposed a revolutionary theory which in one stroke transformed the impenetrable complexity of the distribution of vowels in the daughter languages into a system of almost incredible simplicity. He argued that although verbs showed a perplexing variety of vowels in the attested languages, all verbs in the ancestor language had just one core vowel: e. According to his hypothesis, verbal roots in Proto-Indo-European looked like *sek ‘cut’, *bher ‘bear’, *bhewg ‘flee’, *deyk ‘show’.
==========
Hrozný recognized NINDA as the word-sign for ‘bread’ (and assumed that – an must have been some ending). Now in a sentence with ‘bread’ there is a particular verb one would expect to encounter, and as it happened, the word ettsa-tteni, which appeared just after NINDA, looked suspiciously like the prime candidate in various Indo-European languages: Old High German ettsan, Latin edere, Old English etan, Modern English eat. Was it really possible that ettsa was the Hittite verb for ‘eat’, with an ending -tteni? Of course, the resemblance might just be a coincidence, but then again, what about the next word, watar (assuming that -ma was just another ending)? If the first two words had something to do with eating bread, then it doesn’t require a great leap of the imagination to guess what watar might mean. Hrozný then considered the final word eku-tteni, and concluded that if ettsa-tteni was the verb ‘eat’ with an ending -tteni, then eku-tteni had to be another verb with the same ending. And since eku-tteni comes just after watar, then it wasn’t difficult to put two and two together and work out what the verb eku might be. Hrozný thus decided that the whole sentence must have been about eating bread and drinking water.
==========
From verbal endings in other Indo-European languages (such as Sanskrit -thana), he deduced that the ending -tteni was the second person plural (‘ye’), and so he came to the conclusion that the meaning of the sentence must be ‘you will eat bread, you will drink water’.
==========
The language of the texts from Hattusa, which had defied scholars’ wits for more than two decades, was thoroughly Indo-European, and the main reason why it had taken so long to work this out was that Hittite was nearly a thousand years older than the earliest known texts in the sister languages such as Greek and Latin.
==========
It was an ‘almost unbelievable accident’, he wrote, but Hittite appeared to have retained one of the rogue sounds which Saussure had hypothesized. Hittite was so much older than the other attested Indo-European languages (some of its texts dated from the seventeenth century BC) that it still contained a certain consonant, transcribed as , which – lo and behold – appeared in the very places where Saussure had expected one of the rogue sounds to turn up.
==========
Take the root ‘protect’, for instance, which appeared as pās in the other languages, but which according to Saussure must earlier have been paXs (and ultimately have started out as a regular peXs). In the Hittite texts, this root showed up with an additional consonant precisely where Saussure expected an X – it was written pas. What Saussure had deduced purely on the basis of formal correspondences between the other Indo-European languages was dug up more than thirty years later from the Anatolian earth, scratched on clay.
==========
There is every reason to expect, Schleicher wrote in 1850, that languages should progress and develop more and more perfect structures during their history, but alas … at first sight we observe precisely the opposite. The further back we can follow a language, the more perfect we find it. Latin, for example, is richer in forms than the living Romance languages. The living languages of India that stem from Sanskrit have sunk even deeper from the high level of linguistic perfection of their ancestor. In historical times, as we know from experience, languages as such go backwards.
==========
‘Languages,’ he declared, ‘are natural organisms which emerged independently of man’s will, grew and developed according to certain rules, and in turn become old and die; they also possess that series of symptoms which one is accustomed to understand as “life”.’ So, like any other living organisms, languages have an early period of growth, followed by a period of decay. And the turning-point between these two periods lies exactly at the ‘dawn of history’. As Schleicher explains, ‘precisely the fact that we find language already fully constructed by the first dawn of history provides the proof that language-building actually takes place before history. History and language-building are two opposing activities of the human spirit.’ In prehistoric times, according to Schleicher, a nation was busy constructing its language, and ‘only when a nation has perfected its language, may it make its entrance into history’.
==========
Prehistoric languages must have had scores of irregularities, but these must have vanished without trace. So the image of a flawless language spoken some time in prehistory turns out to have been mainly a mirage. In reality, there never was a Golden Age of perfection.
==========
Removal vans in Athens, like the one in the picture above, don’t bear the word METAΦOPEΣ (METAFORES) on their back end because they are advertising courses in creative writing.
==========
As will soon become apparent, we use metaphors not because of any literary leanings or artistic ambitions, but quite simply because metaphor is the chief mechanism through which we can describe and even grasp abstraction. This chapter will expose the role of metaphor in the making of linguistic structures, by tracing a stream of metaphors that runs right through language and flows from the concrete to the abstract. In this constant surge, the simplest and sturdiest of words are swept along, one after another, and carried towards abstract meanings. As these words drift downstream, they are bleached of their original vitality and turn into pale lifeless terms for abstract concepts – the substance from which the structure of language is formed. And when at last the river sinks into the sea, these spent metaphors are deposited, layer after layer, and so the structure of language grows, as a reef of dead metaphors
==========
We have probed deep enough by now to realize that metaphor is much more than just a frill on the edges of language. The sheer density of metaphors even in the most listless prose may be surprising, but the real extent of metaphor’s involvement in ordinary language is only just beginning to surface.
==========
‘Have’ ultimately derives from a Proto-Indo-European root *kap, which meant ‘seize’. The original sense of *kap survives in the Latin root cap ‘seize’, which found its way into English in the borrowed words ‘capture’ (as well as in ‘captive’, ‘caption’, ‘capable’, ‘recipe’, ‘occupy’, and even ‘catch’).
==========
==========
It turns out that metaphor is not only a chief supplier to our store of words, it also provides the raw materials for the structure of language itself.
==========
The first unmistakable sign of change appears in the writings of a guy called Fredegar. Not much is known about him, except that he lived in the seventh century, in what today would be France, and that he wrote a long rambling history of the Frankish kingdoms. Poor old Fredegar hasn’t had a very good press. The Encyclopaedia Britannica, no less, dismisses his Chronicle of the Frankish Kingdoms as ‘written in barbarous Latin and excessively dull’.
==========
Fredegar describes a battle between the Byzantine Emperor Justinian and the Persian king Kavadh, which was fought around a border town called Daras. Fredegar offers a quaint explanation for how Daras got its name. Having defeated the Persians, Justinian had Kavadh brought to him in fetters, and demanded that he cede large territories. But Kavadh wouldn’t hear of it. He kept saying non dabo (‘I won’t give’), and Justinian kept answering daras (‘you will give’). And according to Fredegar, the town Daras was founded on the precise spot where the argument took place. CHAIRMAN: As I’m sure you know, Dr de Troy, daras is not the correct Latin for ‘you will give’. It should be dabis. DE TROY: But that’s exactly the point. Although Fredegar was writing in Latin, in this instance he chose a word from his ‘barbarous’ vernacular to explain the town’s name. Daras is in fact the first recorded example of the future tense in the Romance languages: it’s the contraction of the phrase dare habes ‘to.give you.have’. JOURNALIST: Do you mean that the whole phrase dare habes was reduced to just daras? That’s pretty drastic, isn’t it? DE TROY: Not as drastic as the month Augustus, which ended up as a mere ‘oo’. Daras is a fairly mild reduction compared to that, wouldn’t you say?
==========
Yes, but what about Latin? Could the whole case system really have emerged in this way, from postpositions? DE TROY: It must have done. One of the few case endings that still betrays something of its origin is the ablative plural ending -ibus, as in consul-ibus ‘by the consuls’. This -ibus probably contains traces of an Proto-Indo-European postposition *bhi, which is related to what ended up in English as the preposition by.
==========
In Latin, not all nouns have the same case endings. Different nouns have different sets, and that makes the whole system so much more elaborate. But if, as you claim, those case endings all came from postpositions, then shouldn’t all nouns have had exactly the same endings? DE TROY: Actually, it’s quite likely that in the earliest stages of Proto-Indo-European, there really was just one set of endings for all nouns – one size to fit all, as it were. But what made the system so much more complex was … none other than the forces of erosion. The postpositions fused with the nouns to become case endings, but the erosion did not stop at that. In the process of reduction, the case endings also merged with the final syllable of the nouns, and this is ultimately what produced so much variety.
==========
Once upon a time, in the days before records of Latin began, there must have been a phrase hoc die, which meant ‘(on) this day’. By the time of attested Latin, this phrase had eroded and fused into one word, hodie ‘today’. Later on, in Old French, hodie was ground down into a meagre hui, but the French soon found that they couldn’t utter this paltry syllable with enough emphasis, so they piled up more words, and started saying au jour d’hui, literally ‘on the day of this-day’. But with repeated use, this became a set phrase, and so it fused into one word again: aujourd’hui. And nowadays in colloquial French, the same cycle is beginning all over again. A mere aujourd’hui is not deemed to have sufficient presence, and so to emphasize it, the French have started saying au jour d’aujourd’hui – literally ‘on the day of on-the-day-of-this-day’. As you can imagine, this usage is frowned upon by purists, but things have now sunk so low that you can find the phrase in practically any French dictionary, even if still labelled ‘colloquial’.
==========
Take an English phrase like ‘up above’, and you’ll discover a no less hyperbolic history. Old English ufan meant ‘on up’ – it was the locative case of the preposition uf ‘up’. But this little ufan was not considered nearly sturdy enough, so it was reinforced by another preposition, be ‘by’, to give a beefier be-ufan ‘by on up’. But before long, be-ufan was assaulted by the forces of erosion, and ended up as a mere bufan. Naturally, the syllabically-challenged bufan had to be pumped up again, this time by the preposition an ‘on’, to give an-bufan ‘on by on up’. Later on, anbufan was ground down by erosion, and – to cut a long story short – ended up as the modest above.
==========
So perhaps the easiest way of understanding these cycles of piling up, fusion and erosion is to imagine the forces that work on language as a kind of tireless compressing machine. Erosion keeps pounding at words, making them shorter and shorter. But shortened words are piled up into longer expressions, and the same forces of erosion then hack away at the pile, fuse the words and condense them into a more compact word once more. And so a new cycle begins all over again.
==========
Erosion is a highly useful compacting mechanism which allows us to convey ideas faster and more efficiently. Erosion checks the excesses of expressiveness, just as expressiveness repairs the excesses of erosion.
==========
In earlier stages of English, all verbs maintained a distinction between singular and plural in the past tense: he ‘herde’ but they ‘herden’ (see Chapter 3). In modern English this distinction has long been levelled out, so that verbs nowadays only have one form: he/they ‘heard’. There is, however, a single but notable exception. Perhaps the most common verb of all, the ubiquitous ‘be’, has clung on to this long forgotten distinction, and it still shows a difference between ‘he was’ and ‘they were’.
==========
The reason why such frequent words can sometimes cling on to outmoded traits that have long been shed elsewhere in the language is their extreme familiarity. The most common words are heard so often that they can quickly become indelibly imprinted in the minds of new generations of learners, and thus withstand even drastic overhauls in the rest of the language.
==========
Fortunately, there are some other relics strewn around the Semitic verb which give us clues about the early days of its evolution, and suggest that the thousand-mile march towards the root-and-template system may have started as early as eight millennia ago, with just one small step.
==========
Semitic, Berber and Cushitic are members of what scholars nowadays call the Afro-Asiatic language family. No one can say for sure when the Semitic branch of Afro-Asiatic started diverging from the Cushitic branch, but based on the linguistic distance between the languages, linguists believe that it must have been at the very least 8,000 years ago. While none of the other Afro-Asiatic languages has a root-and-template system like that of Semitic, many of them do show a suspiciously familiar vowel mutation between the tenses.
==========
Let’s now put together everything we have uncovered so far: some time in the prehistoric period, the ancestor of the Semitic languages must have started out with ‘normal’ verbs, with sturdy stems like mūt, nīk, ktum or ptil, which had both consonants and vowels to their name. The
==========
Let’s now put together everything we have uncovered so far: some time in the prehistoric period, the ancestor of the Semitic languages must have started out with ‘normal’ verbs, with sturdy stems like mūt, nīk, ktum or ptil, which had both consonants and vowels to their name. The first step in the evolution of the root-and-template design may have been taken as early as 8,000 years ago, when, for some strange reason, a vowel mutation emerged in the future tense: the vowel of the ancient stem changed to a: In itself, this mutation pattern may not seem such a huge leap forward.
==========
Let’s now put together everything we have uncovered so far: some time in the prehistoric period, the ancestor of the Semitic languages must have started out with ‘normal’ verbs, with sturdy stems like mūt, nīk, ktum or ptil, which had both consonants and vowels to their name. The first step in the evolution of the root-and-template design may have been taken as early as 8,000 years ago, when, for some strange reason, a vowel mutation emerged in the future tense: the vowel of the ancient stem changed to a:
==========
Nevertheless, the a-mutation is a defining moment in the evolution of the Semitic verbal design, since the kernel of a new concept has been formed, from which the notion of the consonantal root will later spring: the idea that a verb can keep the same consonants, but change the vowels between them to mark nuances like tense.
==========
So, for instance, a haphazard sequence of effort-saving changes (which had nothing whatsoever to do with meaning) turned the German plural noun gastiz to gest. But later generations of order-craving minds thought they spotted a pattern in pairs like gast-gest, and presumed that this pattern must be there for some purpose – marking plurality. In consequence, they extended this model to other nouns, creating a more regular pattern.
==========
At the opening of the book, I said that the available written records of any language extend at most 5,000 years into the past, and that the languages around by that time already have pretty much the full repertoire of complex features found in today’s languages. The use of subordination seems to be one of the only exceptions to this claim, since the earliest attested stages of many languages do show a significant difference in this respect from what we are used to in modern written styles. The use of subordinating conjunctions in the earliest stages of ancient languages such as Sumerian, Akkadian, Hittite or Greek seems to have been much less developed, and in consequence, Caesar’s principle had to be relied upon much more consistently. This is why the style of some ancient texts can seem so monotonous to us.
==========
No matter how hard one tries to trace their historical origin, the pointing words in any language never seem to emerge from anything that was not a pointing word to start with. Unlike grammatical words, which over and over again can be seen to develop from nouns and verbs, pointing words appear to have been pointing words all along.
==========
All this suggests that pointing words must have emerged directly as vocal accompaniment to an actual pointing gesture. And it is also likely that the pointing words could have acted as a bridge between the visual processing system and language.
==========
The link between the pointing words and pronouns may not seem immediately obvious, but it may become more transparent when one takes into account that many languages don’t just have a two-way distinction between ‘this’ and ‘that’, but rather a three-way one: ‘this’ (pointing to something near the speaker); ‘that’ (pointing to something further away from the speaker, nearer the addressee); and ‘that yonder’ (pointing to something far away from both speaker and addressee).
==========
Classical Latin is one example, with the triad hic-iste-ille, and Japanese has a similar distinction with koko ‘here’- soko ‘there (near addressee)’- asoko ‘yonder’. From this three-way system, it is much easier to imagine how the triangle ‘me-you-him’ could have evolved, and some languages even illustrate the link quite transparently. In Vietnamese, for instance, it’s not uncommon for speakers to use ‘here’ to refer to themselves, and ‘there’ to refer to the addressee:
==========
Third person pronouns (like ‘he’ and ‘she’) are generally nothing other than pointing words for distant objects (‘that yonder’), which have been divested of their emphatic force. A well-known example is Latin ille ‘that one (yonder)’. In earlier stages of Latin, ille was used only for ‘real’ physical pointing, that is, for pointing at objects in the distance. This physical pointing was then extended metaphorically to the ‘space of conversation’ (see Chapter 4), and ille thus came to ‘point’ at a previous mention of a person or thing in the discourse, meaning ‘that one (which I mentioned a moment ago)’. But with time, both the emphatic force of ille and its form were eroded, and it eventually ended up in French simply as il ‘he’.
==========
We saw in Chapter 1 that from an English perspective, Turkish seems to arrange its words quite consistently back to front, whereas from a Turkish perspective, it is of course English that has got it entirely the wrong way round. What is beginning to emerge here is how this mirror-image effect ultimately goes back to just one basic choice that languages make at some stage during their history, about the arrangement of the action-word and the patient (‘take stone’ or ‘stone take’). This preference can influence not just whether prepositions will emerge (as in English) or postpositions (as in Turkish), but can ripple through the structure of a language, and have far-reaching repercussions on the order
==========
We saw in Chapter 1 that from an English perspective, Turkish seems to arrange its words quite consistently back to front, whereas from a Turkish perspective, it is of course English that has got it entirely the wrong way round. What is beginning to emerge here is how this mirror-image effect ultimately goes back to just one basic choice that languages make at some stage during their history, about the arrangement of the action-word and the patient (‘take stone’ or ‘stone take’). This preference can influence not just whether prepositions will emerge (as in English) or postpositions (as in Turkish), but can ripple through the structure of a language, and have far-reaching repercussions on the order of many other elements.
==========
Time and time again, and in language after language, ‘give’ verbs turn into prepositions meaning ‘for’ or ‘to’.
==========
To start with, where do the names for properties actually come from? Property-words did not feature among the primary terms which were taken as given at our starting point, because they can be derived from the raw materials that are already at our disposal: words for physical objects and for simple actions. To get a sense of the process involved, one need only call to mind a few colour-words, such as orange, olive, violet, silver, claret, burgundy. All of these are commonly used as normal property-words nowadays (‘her dress is orange’, ‘ultra-violet light’, and so on) but their origin is still transparent. Initially, they referred to particular objects (fruit, flowers, metals, wines) which happened to have a striking property, namely their colour, and so the original thing-word came to be used more generally
==========
To get a sense of the process involved, one need only call to mind a few colour-words, such as orange, olive, violet, silver, claret, burgundy. All of these are commonly used as normal property-words nowadays (‘her dress is orange’, ‘ultra-violet light’, and so on) but their origin is still transparent. Initially, they referred to particular objects (fruit, flowers, metals, wines) which happened to have a striking property, namely their colour, and so the original thing-word came to be used more generally
==========
To get a sense of the process involved, one need only call to mind a few colour-words, such as orange, olive, violet, silver, claret, burgundy. All of these are commonly used as normal property-words nowadays (‘her dress is orange’, ‘ultra-violet light’, and so on) but their origin is still transparent. Initially, they referred to particular objects (fruit, flowers, metals, wines) which happened to have a striking property, namely their colour, and so the original thing-word came to be used more generally to denote the colour itself.
==========
While these examples may all refer to colours that are rather fancy, the origin of more basic colour-words is often the same. ‘Red’, for instance, sometimes comes from ‘blood’, as in the Semitic languages, where dam ‘blood’ is the source of adam ‘red’ (and hence, incidentally, via the Hebrew word for ‘red dust’, the name of Adam, ‘man’). In a similar vein, the word for ‘green’ often comes from something to do with ‘leaf’, or (as in English) from something to do with ‘grow’.
==========
… ‘small’ often derives from ‘child’ or ‘baby’ (think for instance of ‘baby grand piano’), ‘strong’ can come from ‘a youth’, ‘sharp’ can come from ‘tooth’ or ‘shard’.
==========
Plural markers often develop from the quantifier ‘all’, or from intrinsically plural terms such as ‘people’. One example can be seen in southern dialects of American English, where a plural form of the pronoun ‘you’ has emerged, y’all, obviously from the fusion of ‘you all’.
==========
Definite articles invariably come from pointing-words like ‘that’. English ‘the’ is in fact just an eroded form of ‘that’. French le has a similar origin, the Latin ille ‘that’. (I mentioned earlier that ille was also the source of the French pronoun il ‘he’. The reason for the divergence in form is different paths of erosion: when ille was used together with a noun, as in ille liber ‘that book’, the stress pattern was such that erosion hacked at it mostly from the front, and so it ended up as le livre ‘the book’. But when it was used on its own, ille ‘that (one)’ was eroded from the other end, and so turned into French il ‘he’.)
