English Spelling, Sounds, and Grammar PDF

Chapter 2 English spelling, sounds, and grammar In Old and especially in Middle English, variable spellings are common, even within the writings of one scribe. After 1500, the first English dictionaries and word lists start appearing, but the English spelling system remains irregular. The main reason for this is that changes occurred in the spoken language which were not reflected in the spelling because by then the spelling had been standardized. In addition, over time English – or rather its speakers – borrowed many words from other languages. As a result, English spelling does not represent the way the language sounds: Plato could also be spelled play-dough, at least in American English. To indicate the pronunciation of words, we therefore need a phonetic alphabet. Section 1 of this chapter offers examples of spelling irregularities and Section 2 describes some of the reasons behind them. Section 3 introduces the phonetic alphabet, which represents spoken language as accurately as possible, and Section 4 provides background information on phonetics, which helps explain many of the changes that have occurred in English. Section 5 discusses morphology and syntax since they are relevant to the linguistic changes discussed later in the book. 1. English spelling George Bernard Shaw, the well-known Irish writer, is credited to have said that fish could just as well be spelled ghoti. The gh could sound like the last sound in enough, the o like the first vowel in women, and the ti like the middle part of nation. It turns out there isn’t much evidence that Shaw ever used the ghoti example. Zimmer (2010) reports that the true origins of ghoti date back to 1855 and that the word was invented by William Ollier. Shaw never used it apparently but someone misattributed it to him. Shaw considered the English spelling system inadequate and in need of reform. The following quote from the Preface to Shaw’s Pygmalion indicates his views on English spelling: The English have no respect for their language, and will not teach their children to speak it. They cannot spell it because they had nothing to spell it with but an old foreign alphabet of which only the consonants – and not all of them – have any agreed speech value. Consequently no man can teach himself what it should sound like from reading it. When Shaw died in 1950, he left money to devise a new writing system for English. Although such an alphabet was indeed invented and a book was published using it, it never caught on (the alphabet can be seen at www.shawalphabet.com). 16 A History of the English Language The vowels in bold in (1) provide examples of irregular spelling. They are all spelled differently but sound the same. The phonetic symbol we will use for the sound they have in common is [i]. (1) she, Harry, believe, Caesar, see, people, seize, seas, amoeba, key, machine, suite, and quay. There are also many ways to spell what we will represent as [u], as shown in (2). (2) to, too, two, through, threw, clue, Sioux, suit, flu, lieu, Pooh, Lou, and shoe. Sometimes the opposite occurs: what is spelled the same sounds very different, as demonstrated by the sounds spelled as ou in the words in (3). (3) tough, previous, ought, through, dough, and out. As you will notice, most variation in spelling and pronunciation occurs with vowels. The French philosopher Voltaire is quoted as saying that vowels count for nothing (and consonants for very little). For some speakers of American English, there is no distinction in sound between pin and pen and others do not distinguish between Mary and marry, and for some speakers of British English tower and tar sound the same. Some consonants also show a variety of spellings. For instance, the k sound of keep is spelled ck after short vowels (lack, sick, Rick, deck), k after long vowels (week, soak, shake) and before front vowels (keep, kin, kettle), and c before back vowels (cool, could, cold, cup). (We’ll learn what front and back vowels are in Section 4). The same k sound is spelled differently in borrowed words such as psychology and choral. There are many jokes and poems about spelling irregularities, such as the poem below from an unknown source. I take it you already know Of tough and bough and cough and dough? Some may stumble but not you, On hiccough, thorough, slough, and through? So now you are ready perhaps to learn of less familiar traps? Beware of heard, a dreadful word that looks like beard and sounds like bird, and dead is said like bed, not bead or deed. Watch out for meat, great, and threat that rhyme with suite, straight, and debt. Figure 2.1 Some spelling irregularities As we have seen, in English the correspondence between sound and symbol is not straightforward. Therefore, we need a phonetic alphabet, and one is provided in Section 3. However, first we will look at how English spelling became irregular. Chapter 2. English spelling, sounds, and grammar 2. Why English spelling is irregular The English spelling system has been around for centuries. During this time, it has absorbed words from many other languages and has been used for many varieties of spoken English. The result is that the symbols do not accurately represent the sounds. Also, unlike in Modern German and Dutch, for instance, there has never been a spelling reform in English. In later Old English, there is a standard for West-Saxon Old English for the manuscripts produced at Winchester by Ælfric and others. The Middle English period shows much variety, but after 1400, some standards arise, as we will see later in this section. After 1500, there are advocates for spelling reform: John Hart (Orthographie 1569), William Bullokar (Booke at Large 1580), and Richard Mulcaster (Elementarie 1582). We will revisit these attempts in Chapter 7. Even today, some groups advocate spelling reform, e.g. The English Spelling Society (google their website for more information). The arguments against spelling reforms are many. One is that pronunciation varies so much in the English spoken around the world that it would be hard to come up with one spelling system. A practical problem would be that the various governments and newspaper and book editors in areas where English is an official language would have to agree. Alternatively, several different spellings could be ‘allowed’. In that case, however, we would have multiple systems, and English speakers from India, for instance, might no longer be able to read what speakers in the United States write. As has been mentioned already, the main reason English spelling is irregular is that many sound changes have occurred since it was (unofficially) standardized. First, we will discuss standardization and then we will address the sound changes. In Old and Middle English, scribes used a modified Roman alphabet to transcribe their own speech or to copy from other manuscripts. There is often a lot of variation within the writings of one scribe as well as between different scribes from the same area: sealm, selm, salm, spalme, sphalme and many others are listed in the OED for ‘psalm’. Examine, for instance, the variations of shirt and though in the OED. The current online version of the OED lists 23 different spellings of shirt and 29 of though throughout the history of English! Despite the variation, a standard came to be established since scribes often copied earlier manuscripts and many indeed copied the symbols indicative of an earlier pronunciation. At the end of the Middle English period (in 1420 to be precise), scribes working at the Chancery began writing in English rather than Latin and seem to have been following a Midlands variety, i.e. the area to the North of London. The exact source of the standard is still a matter of debate (see e.g. Samuels 1963, Fisher et al. 1984, Benskin 1992; 2004). The Chancery produced a huge number of documents, and this was connected to the rise of London as a major center for trade and politics. Chancery English is characterized by relatively free spelling, but some rules evolve (a) gh at the end of high even though the consonant is no longer pronounced, (b) th endings for third person verbs, as in he doth ‘he does’ even though many varieties already have he does, (c) past tense (e)d endings and not t even though many sound like t, and (d) such and which for su(l)ch and (h)wi(l)ch, respectively (there are many other variations; see Fisher 1996). 17 18 A History of the English Language Chancery English may be the beginning of a written standard, one that does not necessarily represent spoken English. Smith (1996), Hope (2000), and Wright (2000) – among others – have looked at migration patterns into London and confirm that the Midlands was crucial for a steady influx of workers. All medieval cities needed immigration to maintain their population levels and the Midlands was characterized by population growth and the rise of a middle class. It also generated scientific and religious texts such as Wycliffite sermons and texts. Thus, the Midlands and London were important to the development of an unofficial standard, where ‘unofficial’ refers to the fact that English has never had an Academy or any other body regulating it. A major boost to the standard comes after the introduction of the printing press in 1476. William Caxton started the printing press in London, physically close to the Chancery, even though he himself came from Kent and had spent much time abroad. Standardization is automatically established when a document, book, or pamphlet is reproduced the same way many times. Caxton relied on the writings of scribes rather than inventing a new system and was not himself interested in standardizing spelling. One of his first books, Malory’s Morte d’Arthur, was published in 1485, based on an older manuscript. The printed result contains a lot of variation (duke and duk, when and whanne, hyr and hir for ‘her’); it is quite possible that words no longer sounded the way Caxton printed them. Spelling variations stay around for a long time; Shakespeare’s First Folio of 1623 contains many of them (heart and hart, do and doe). Other important developments towards standardization were the publication of the first English dictionaries around 1600 (see Chapter 7) and the King James Bible in 1611. Thus, between the 1430s and the 1480s, a standard began to evolve. However, beginning around 1400 and continuing until after 1600, perhaps the most ‘disruptive’ of all changes – the Great Vowel Shift (GVS) – occurs. This change started around the time standardization was taking place and is a major reason that English spelling doesn’t correspond to the way it sounds. The GVS involves long vowels – such as those pronounced as [a, e, i, u, o] – becoming [e, i, aj, aw, u] respectively. For instance, before 1400 name was pronounced the way it would if you pronounced it in Spanish, Dutch, German, or French. In present day English, the a of name sounds more like the first vowel in Spanish words such as edición ‘edition’ or French école ‘school’. We will come back to the GVS below and in Chapter 7. Other factors that contributed to the irregularity of English spelling are etymological respellings and the incorporation of words from other languages with changed pronunciation but keeping the original spelling. An instance of an etymological respelling is the English word for debt. It is borrowed from French and occurs for the first time in 1300 as dete, dette, and dett, without a b. Because the Latin forms have bs in the same word, the b is introduced by writers wishing to sound learned. The OED says that it was “artificially spelt debte, after which debt has become the English spelling since the 16th c.” The same happens with doubt, borrowed as doute, and fault, borrowed as faute, and soldier borrowed as sauder. There are many spelling variants of these words in Middle English. Thus, soldier appears, for instance, as sauder, sawder, souder, saudiour, soudour, and many other spellings Chapter 2. English spelling, sounds, and grammar but only around 1500 does the -l- come into the word (and we have of course started to pronounce it in later English). Other examples are receipt, perfect, and indict, borrowed as receyt, parfit, and endyte. The Old English word igland ‘island’ is mistakenly considered to be like the French word isle and therefore the s is introduced into the Modern English island. Johnson’s dictionary of 1755 has the following inconsistent spellings: deceit and conceit versus receipt and fancy versus phantom. Spenser (the author of the Fairy Queen) is supposedly responsible for respelling delite as delight so it would form an ‘eye-rhyme’ with light and night. Instances of loan words incorporated into English in terms of the sounds, but not the spelling are listed in (4): (4) Phoenix, suite, xylophone, quota, chagrin, gnomic, euphemism, debris, glacier These words are usually pronounced as if they were English words, although some may attempt to pronounce chagrin as French. They could easily be spelled feeniks, sweet, zailofone, kwota, shagrin, nomick, youfimism, daybree/duhbree, and glayshir. Other languages have borrowed words and changed the spelling: Dutch borrowed cadeau ‘present’ from French but now sometimes spells it kado and, in the Dutch edition of Harry Potter, phoenix becomes feniks. Sometimes, it is not the word that is borrowed but the letter combination. For instance, the French combination qu replaced Old English cw and cwene became queen. This is due to the influence of the Anglo-Norman scribes. Spelling pronunciation is a phenomenon where speakers pronounce words as they are spelled. For instance, pronouncing the [t] in often and the [l] in salmon are hypercorrections that regularize spelling. Above, the [l] in soldier was mentioned as well. It also happens in the incorporation of loans, e.g. pronouncing the [l] in loans from Spanish, such as tortilla and cholla, rather than the expected [j]. The latter sometimes occurs for external reasons (see Hill 1993). As we have seen in this section, English spelling is irregular. We therefore need to have a way to represent spoken English; we will explore this in the next section. 3. The phonetic alphabet The symbols for the vowels are given in Table 2.1. In the text, the pronunciation of the sounds will be provided in square brackets, according to convention. There is a great deal of variation in the way speakers pronounce certain vowels, especially ["], [a] and [#]; hence, the exact number of vowels is debatable. The reason for the organization of the table will become clear in Section 4. The table also provides an English word in which the sound is used. Check their possible pronunciation in American English at http://www.uiowa.edu/~acadtech/phonetics. Several ways to transcribe the vowels of English can be found in the literature. Mine mostly follows the International Phonetic Alphabet (see www.langsci.ucl.ac.uk/ipa), except in the use of [j] for sounds like [aj] and [#j]. 19 20 A History of the English Language Table 2.1 Phonetic symbols for English vowels sound word i i e ε æ teeth miss make bet bat sound % ə (=schwa) word putt roses sound word u ' o # " aj #j aw too book moat hot father mice boys house Some speakers pronounce a [j] sound after long vowels such as [i] and [e], as in [sij] for see, [lejt] for late. It is up to you to decide how to represent these, depending on how you pronounce them or hear them pronounced. Table 2.2 lists the symbols for English consonants and provides words starting with those consonants. (In Chapter 4, we will see that the ‘(’-like symbol is also used to spell the [j] sound in earlier English). Table 2.2 Phonetic symbols for English consonants (*no word starts with [ŋ] or [(] or [’]; see therefore the consonant in bold) sound word sound word sound word p t k f s θ ʃ ʧ h pet ten cat for sorry thigh shoe chirp he b d g v z ð ( d( (’) bet den get very zoo that rouge* judge bottle* (some speakers) m n ŋ l r j w (*) met no sing* late roll yes wit where (for some speakers) Consonants are less likely to change, but – as we will see – Old English lacks a few of the Modern English consonants; for example, it does not have [(], typically found in loans from French, such as rouge and pleasure. Chapter 2. English spelling, sounds, and grammar 4. Phonetics and sound change As we will see, sound change is regular, unlike the resulting spelling system. To understand that it is regular (and that an [m] does not change to a [k], etc), we need some background on how to describe sounds – phonetics. In this section, we will first look at vowels, then at consonants, and then at how sounds affect each other. English vowels can be described using three features: (1) whether the tongue is high or low, (2) whether the tongue is front or back, and (3) the length of the sound. Explore the height differences by pronouncing [i], [e], and [æ]. You should feel your tongue moving down as you proceed from one sound to the next. The same downward movement should happen if you pronounce [u], [o], and ["]. The difference between front and back vowels can be felt by pronouncing [i], [e], and [æ] versus [u], [o], and ["]. The former are pronounced with the tongue in the front of the mouth, the latter with the tongue in the back. The third feature, length, can be observed, for instance, when comparing [i] with [i] and ['] with [u]. There is a slight difference in height and frontness between long and short pairs, which we will not pay much attention to. In English, [i, e, aj, aw, o, #j, u] are long vowels and [i, ε, æ, ", %, ə, #, '] short. Using the features listed above, [e] can be described as a mid, front, long vowel, and [o] as a mid, back long vowel. Most consider [aj, aw, #j] diphthongs, two sounds in one. In languages such as Old English, a fourth feature, round, is relevant for categorizing vowels. Front vowels such as [i] and [e] are produced with the lips in a spread position, whereas back vowels such as [u] and [o] are made with rounded lips. Make these sounds to feel the difference. In addition to these vowels, some languages have rounded front vowels (e.g. [y] in Old English mys ‘mice’, German Küche ‘kitchen’ and French tu ‘you’) or unrounded back vowels (e.g. [+] in Vietnamese). Figure 2.2 presents the vowels for languages with five vowels (mentioned in Chapter 1, Section 2) and Figure 2.3 the vowels for a language like English. These diagrams are shown as if we were looking at the left side of someone’s mouth and could see the tongue’s position through the cheek. Front Back Front High i u High i e o Low Back u i $ e a ε Low Figure 2.2 Five vowel system ə o " % %j æ aj aw # Figure 2.3 English vowels 21 22 A History of the English Language How is this division relevant to language change? In Section 2, we mentioned that the Great Vowel Shift (GVS) is responsible for many of the irregularities in the spelling system. ")JTUPSZPGUIF&OHMJTI-BOHVBHF Based on work by Karl Luick, many linguists have analyzed this shift, e.g. Jespersen (1909, Chapter 8) and Chomsky and Halle (1968, Chapter 6).We will describe the shift using VTJOHSFMBUJWFMZTJNQMFMJOHVJTUJDUFSNJOPMPHZ8IFO<B>CFDPNFT<F> XFTBZUIBUJUSBJTFT relatively simple linguistic terminology. When [a] becomes [e], we say that it raises (and BOEGSPOUT XIFO<F>CFDPNFT<J> JUSBJTFTUIFTBNFIBQQFOTXIFO<P>CFDPNFT<V> fronts), when [e] becomes [i], it raises; the same happens when [o] becomes [u]. Thus, the ѮVT involves UIF(74JOWPMWFTSBJTJOHUIFMPOHWPXFMTѮFUXPWPXFMTUIBUBSFAQVTIFEPVUPGUIF GVS raising the long vowels. The two vowels that are ‘pushed out of the system’ TZTUFNBSF<V>BOE<J>ѮFZCFDPNFEJQIUIPOHT<BX>BOE<BK> SFTQFDUJWFMZ'JHVSF are [u] and [i]. They become diphthongs – [aw] and [aj], respectively. Figure 2.4 represents SFQSFTFOUTUIJTSBJTJOHJOWFSZHFOFSBMUFSNTCVUTFF$IBQUFSGPSNPSFEFUBJM this raising in very general terms but see Chapter 7 for more detail. 'JHVSF ѮFNBJOEJSFDUJPOPGUIF(74 Figure 2.4 The main direction of the GVS 4JODFUIF(74POMZBĒFDUFEMPOHWPXFMT &OHMJTIIBTQBJSTTVDIBTTFSFOFTFSFOJUZ QSP GPVOEQSPGVOEJUZ BOEEJWJOFEJWJOJUZѮFTFDPOEWPXFMJOUIFêSTUXPSEJTMPOHUIFTBNF Since the GVS only affected long vowels, English has pairs such as serene–serenity, proWPXFMJOUIFTFDPOEXPSEJTTIPSUѮFMPOHWPXFMTIBWFTIJѫFEUP<J> BOE<J> found–profundity, and divine–divinity. The second vowel in the first <BX> word <BK> is long; the SFTQFDUJWFMZ CVUUIFTIPSUPOFTIBWFSFNBJOFENPSFPSMFTTUIFTBNFѮVT UIFTFDPOE same vowel in the second word is short. The long vowels have shifted to [i], [aw], and [aj] WPXFMJOTFSFOJUZJTOPUQSPOPVODFEBT<ө>CVUBT<ʆ> respectively, but the short ones have remained more or less the same. Thus, the second vowel in serenity is not pronounced as [i] but as [ε]. Other long and short vowel pairs in TBOFTBOJUZ WBJOWBOJUZ HSBJOHSBOBSZ IVNBOFIVNBOJUZ DMFBODMFBOMJOFTT occur (5). NBMJHONBMJHOBOU DSJNFDSJNJOBM TJHOTJHOJGZ (5) sane/sanity, vain/vanity, grain/granary, humane/humanity, clean/cleanliness, malign/malignant, crime/criminal, sign/signify. 4PVOEDIBOHFTEJEOPUKVTUIBQQFOJOUIFQBTU7BSJPVTDIBOHFTBSFTUJMMIBQQFOJOHUPEBZ 'PSJOTUBODF -BCPW BOE8PMGSBN EFTDSJCFUIF/PSUIFSO$JUJFT4IJѫ UIF Why do sounds shifts, such as the GVS, take place? There has been much debate on this. 4PVUIFSO7PXFM4IJѫ BOE$BOBEJBO3BJTJOHѮF/PSUIFSO$JUJFT4IJѫDBOCFIFBSEJO Some think the GVS was caused by external events (Fennell 2001); others by internal #VĒBMP $IJDBHP $MFWFMBOE %FUSPJU 3PDIFTUFS BOE4ZSBDVTF*UIBTPDDVSSFETJODFUIF factors (Martinet 1955); and yet others through a combination of internal and external TBOEJOWPMWFTTIPSUWPXFMT FHUIF<B>PGGBUIFSBOE$IJDBHPCFDPNJOH<> BOEUIF events (Labov 2010). It is well-known that vowels tend to be distributed relatively evenly <>SBJTJOHJOUVSOѮF4PVUIFSO7PXFM4IJѫIBTUIFCBDLWPXFMTGSPOUJOHBOEUIFGSPOU in the space where vowels can be produced. Thus, languages with just front vowels do not WPXFMTTIJѫJOHQPTJUJPO$BOBEJBO3BJTJOHJOWPMWFTUIFSBJTJOHPGUIFêSTUQBSUPG<BK> BTJO occur. If, for some reason, a vowel shifts, another vowel may enter that space. Long vowels JDF UP<K>BOE<BX>JOIPVTFUP<X>ѮFDIBOHFPDDVSTPOMZCFGPSFWPJDFMFTTDPOTPOBOUT typically raise or front, as in the GVS, but some change must have started it and that initial BOEUIFSFTVMUJTUIBUUIFWPXFMTJOJDFBOEFZFTBSFEJĒFSFOU/PUJDFUIBU GPSSFBTPOTOPU trigger is not clear. VOEFSTUPPE UIFWPXFMDIBOHFTUZQJDBMMZTUBSUBTSBJTJOHPSGSPOUJOH *O4FDUJPO XFFYBNJOFEBTQFMMJOHJSSFHVMBSJUZJOWPMWJOH<L>ѮJTJSSFHVMBSJUZDBO CFFYQMBJOFEVTJOHUIFGSPOUCBDLEJTUJODUJPOJOWPXFMT8PSETUIBUBSFPSJHJOBMMZ&OHMJTI VTFBDCFGPSFBCBDLWPXFM DPPM DPVME DPQQFS BOEBLCFGPSFBGSPOUWPXFM LJOH LJUDIFO LFFQ -FOHUIJTBMTPSFMFWBOUBѫFSBMPOHWPXFM PSBDPOTPOBOU UIFTQFMMJOHPG<L>JTL XBLF XFFL TOBLF XPSL XJOL BOEBѫFSBTIPSUWPXFM JUJTD L TJDL /JDL TBDL -FUTEFTDSJCFDPOTPOBOUTVTJOHUIFUISFFGFBUVSFTUZQJDBMPGUIFN8IJMFWPXFMTMFU UIFBJSUISPVHIDPNQMFUFMZ XJUIDPOTPOBOUTUIFBJSJTDPOTUSJDUFEJOBQBSUJDVMBSQMBDFJO Chapter 2. English spelling, sounds, and grammar Sound changes did not just happen in the past. Various changes are still happening today (as described in e.g. Labov et al. 2005). There are descriptions of the Northern Cities Shift, the Southern Shift, and Canadian Raising. The Northern Cities Shift can be heard in Buffalo, Chicago, Cleveland, Detroit, Rochester, and Syracuse. It has occurred since the 1950s and involves short vowels, e.g. the ["] of father and Chicago becoming [æ], and the [æ] in cat in turn becoming [eə] and [iə]. The changes proceed in a chain shift, stuck → stalk → stock → stack. The Southern Shift has the high and mid back vowels fronting and the front vowels shifting position (http://www.ic.arizona.edu/~lsp/ provides a good chart). Canadian Raising involves the raising of the first part of [aj], as in ice, to [%j] and [aw] in house to [%w]. This change occurs only before voiceless consonants, and the result is that the vowels in ice and eyes are different. In Section 1, we examined a spelling irregularity involving [k]. This irregularity can be explained using the front-back distinction in vowels. Words that are originally English use a c before a back vowel (cool, could, copper) and a k before a front vowel (king, kitchen, keep). Length is also relevant: after a long vowel (or a consonant), the spelling of [k] is k (wake, week, snake, work, wink) and after a short vowel, it is c(k) (sick, Nick, sack). Let’s now describe consonants using the three features typical of them. While vowels let the air through completely, with consonants the air is constricted in a particular place in the mouth. For instance, in pronouncing [p] and [b], the air is constricted by the closing of the lips; in producing a [θ] and [ð] it is constricted by the tongue and teeth. The air is also restricted in a particular manner, e.g. complete or partial constriction as in [p] and [f], respectively. Another difference between vowels and consonants is that all vowels are voiced, but that is not true of all consonants. Therefore, the features relevant to consonants are: (a) manner of articulation, (b) place of articulation, and (c) voicing. The airflow can be restricted through a complete closure, as in [p, b, t, d, k, g]; the resulting sounds are called stops. Fricative sounds such as [f, v, s, z, ∫, (] let the air through. Affricates, [t∫, d(] in English, are mixtures of a stop and a fricative. Nasals and liquids have a lot in common with vowels in that they are voiced and can be syllables on their own. Nasals are formed by letting the air out through the nasal cavity. Liquid is a cover term for [l] and [r], sounds that are perhaps the hardest to define and the most variable across languages. Glides, such as [w, j], are vowel-like and immediately precede or follow vowels. Place of articulation refers to the place where the air is constricted: the lips for labial [p, b, m, f, v], the teeth for dental [θ, ð], the ridge behind the upper teeth for alveolar [t, d, n, s, z, l, r], the front of the palate for alveo-palatal [∫, (, t∫, d(], the palate for [j], the back of the palate for velar [k, g, ŋ], and the glottis for glottal [’, h]. 23 24 1MBDFPGBSUJDVMBUJPOSFGFSTUPUIFQMBDFXIFSFUIFBJSJTDPOTUSJDUFEUIFMJQTGPSMBCJBM <Q C N G W> UIFUFFUIGPSEFOUBM<ʉ £> UIFSJEHFCFIJOEUIFVQQFSUFFUIGPSBMWFPMBS<U E O T [ M S> UIFGSPOUPGUIFQBMBUFGPSBMWFPQBMBUBM<Ϊ i UΪ Ei> UIFQBMBUFGPS<K> UIF A CBDLPGUIFQBMBUFGPSWFMBS<L History of the English LanguageH Ǌ> BOEUIFHMPUUJTGPSHMPUUBM<A I>ѮFQMBDFTBSFTIPXO JO'JHVSF 'JHVSF 1MBDFTPGBSUJDVMBUJPO Figure 2.5 Places of articulation .PTU DPOTPOBOUT DPNF JO QBJST PG B WPJDFE BOE B WPJDFMFTT TPVOE 7PJDFE TPVOET BSF NBEFXIFOUIFWPDBMGPMETJOUIFMBSZOYWJCSBUF'PSJOTUBODF <G>BOE<T>BSFWPJDFMFTT Most consonants come in pairs of a voiced and a voiceless sound. Voiced sounds are made BOE<W>BOE<[>BSFWPJDFE5SZGFFMJOHUIJTCZQVUUJOHZPVSêOHFSPOUIFNJEEMFPGZPVS when the vocal folds in the larynx vibrate. For instance, [f] and [s] are voiceless, and [v] UISPBUBOEBMUFSOBUJOHCFUXFFOTBZJOHTTTTTTBOE[[[[[[ѮFQBJSTBSFJOEJDBUFEBTTFUTJO and [z] are voiced. Try feeling this by putting your finger on the middle of your throat and 5BCMF between FYDFQUGPSUIFMJRVJETXIJDIBSFCPUIWPJDFE indicated as sets in Table 2.3 alternating saying ssssss and zzzzzz. The pairs are (except for the liquids which are both voiced). Table 2.3 English consonants (*w is labio-velar) Manner Place stop fricative labial dental alveolar alveo-palatal velar glottal p/b f/v θ/ð s/z ʃ/( t/d k/g ’ h affricate nasal liquid m ʧ/d( n ŋ glide w* l and r j To simplify matters, in Figure 2.3, the labio-dentals [f] and [v] are listed as labials, and the palatal glide [j] as alveo-palatal. The liquids can be further divided according to manner: [l] is lateral and [r] is retroflex (in English) How are manner, voice, and place of articulation relevant in language change and variation? The manner in which a sound is produced changes when stops become fricatives, as shown in Latin pater corresponding to English father. This change is part of Grimm’s Law which we will come back to in the next chapter. Stops often become fricatives between two Chapter 2. English spelling, sounds, and grammar vowels, as in Latin faba ‘bean’ to Portuguese fave. In these cases, one sound assimilates to become more like another; this particular kind of assimilation is referred to as frication and it happens because fricatives are more like vowels in letting some air through. Nasalization of vowels (often indicated by a [~] above the vowel) is common when they precede nasal consonants, e.g. for some speakers of English, the vowels in fake and fame are different. It is another instance of manner assimilation on the part of the vowel. Voiceless consonants surrounded by vowels can also assimilate and become voiced: the [t] in the words in (6) starts to sound more like a [d] in American English (and then it is called a ‘flap’ and represented as [ɾ]): (6) literature, spaghetti, butter, bottle British English has an alveolar stop [t] or a glottal stop [’] in butter and bottle. Devoicing occurs as well, as described by Grimm’s Law: Latin ager corresponds to English acre, with voiced [g] changing to voiceless [k]. As for assimilation in place of articulation, nasals provide good examples in English fast speech. They adapt their place of articulation to that of a following stop: in Paris is pronounced [impærIs] and in Canada [iŋkænədə]. Another frequent assimilation in place of articulation involves palatalization – velar consonants such as [k] and [g] becoming palatalized to [ʧ] and [j] respectively. This usually occurs because of the presence of a front vowel. Palatalization happens frequently to Old English words: cirice with an initial [k] becomes church with an initial [ʧ] (in Scotland, one can still hear kirk); cinne becomes chin; ceop turns to cheap (the latter changes from the noun ceop ‘buy’ in god ceop ‘a good bargain’ to the adjective cheap); and ceorl becomes churl (but Dutch keeps kerel). These and other examples are summarized in Table 2.4. The actual change occurs in Old English but I have given pre-change and post-change words. Table 2.4 Palatalization in the history of English Old English [k] cirice cinne ceop cicen ceose > > > > > Modern English Old English [ʧ] church chin cheap chicken cheese [g] weg cæg geard geornan gellan Modern English > > > > > [j] way key yard yearn yell Other processes that occur to facilitate pronunciation are dissimilation, deletion, insertion (or epenthesis), metathesis, and l/r shift. Dissimilation occurs when there are too many sounds of the same kind in a row. For instance, in fifths, many speakers make the middle fricative [θ] into a stop. They can also apply deletion to fifths and pronounce it like [fifs]. Insertion or epenthesis helps makes consonant clusters easier to pronounce: [milk] becomes [milək] in many dialects. Many of these processes occur when languages borrow words that 25 26 A History of the English Language do not fit their phonetic system. Latin astru, for example, is borrowed into Basque as asturu with an inserted vowel. Languages such as Japanese have a strict Consonant-Vowel syllable pattern and borrow English words by inserting vowels into consonant clusters. Metathesis, a process that switches sounds, also changes consonant clusters. For instance, the initial [spə] cluster in spaghetti is often switched to [pəs], as in [pəsgεti]. Other instances are precise becoming [pərsajs], prescription becoming [pərskrip∫ən], and relevant turning into [rεvələnt]. Related words where one has undergone metathesis but not the other one are third/three, nutrition/nurture, promiscuous/mix. Examples of metathesis from Old English to Modern English can be found in Table 2.5. Table 2.5 Metathesis from Old to Modern English Old English beornan beorht tux acs(ian) hros > > > > > Modern English Old English ME brennen bright tusk ask horse wæps hriddel gærs cerse irn(an) Modern English > > > > > wasp riddle grass cress run Rhotacism is a process involving the sound [r] and the change of certain consonants into [r]. You can see rhotacism in the alternation between was and were, rise and rear, opus and opera, and between flos (‘flower’ in Latin) and floral. A final change to be mentioned is the l/r switch: from peregrinus to pilgrim, pruna to plum, and arbor to arbol. The reason for this switch is that [l] and [r] are both liquids and are very unstable over time. This is also shown by their many variants. For instance, Dutch is said to have 13 different [r] sounds in different varieties of Dutch and [l] sounds are variable as well, witness the ‘dark’ [l] in Scottish English, an [l] pronounced very far back in the mouth. 5. Some grammatical terminology This section provides some terminology and context for how words and sentences are built. These fields are called morphology and syntax respectively and this knowledge will become relevant when we start examining Old English in Chapter 4. English changed from a synthetic language, with many endings on nouns and verbs indicating grammatical functions such as subject and object, to a mostly analytic language with almost no endings. In English, words can be formed by means of prefixes, such as pre- and anti-, or suffixes, such as -ness and the plural -s. First, we will look at the prefixes and suffixes that add to the meanings of words (and can change their categories). These word-building rules have not changed much in character since Old English and involve derivational prefixes and suffixes. There are also other ways to construct new words, such as compounding and shortening. Examples of some English word formations are given in (7): Chapter 2. English spelling, sounds, and grammar (7) mark-ed-ness, human-ity, friend-ship, product-iv-ity, comput-er, double-speak, nanny state, green-house-effect, PC, yuppy The first five words involve derivational markings and their affixes are indicated; doublespeak, nanny state (‘too much government interference’), and greenhouse effect are compounds; PC is short for ‘politically correct’; and yuppy is short for ‘young urban professional’. There are many other word formation rules and we will discuss these in later chapters as they become relevant. As mentioned, derivational endings are present in Old, Middle, and Modern English more or less to the same extent. Words can also be marked as being the subject or object of a sentence, for plural and possession by means of inflectional markers, more commonly known as grammatical endings. The endings that mark these functions have changed considerably over time, however. As mentioned, indicating the functions of subject and object is essential; otherwise, we would not know what the sentences of a language mean. There are two basic strategies for representing the functions of subject and objects: (a) by means of word order, pronouns, and grammatical words such as of and by and (b) by means of markers on nouns (case) or on verbs (agreement). Languages using (a) are called analytic and languages using (b) synthetic. I have summarized some characteristics in Table 2.6, where nom stands for nominative, acc for accusative case, and 2sg for second person singular agreement. Table 2.6 Characteristics of analytic and synthetic languages Analytic Synthetic use of prepositions, e.g. the leg of the table use of word order to indicate subject, e.g. The man saw his friend. use of endings, e.g. the table’s leg no markings on the verb to indicate subject but frequent pronouns, e.g. They leave tomorrow. use of case to indicate subject, e.g. De-r Mann sah sein-en Freund (German) [the-nom man saw his-acc friend] verb is marked for subject, and pronoun may be there e.g. þriowa me onsæc-est. (Old English) [thrice me deny-2sg] ‘You will deny me three times’. Synthetic languages indicate subjects either by a marking on the subject, called nominative case, or by marking the person and number of the subject on the verb, called agreement. Old English, a synthetic language, has both of these, but Modern English, an analytic language, has limited case and agreement. In (8), the subject is marked by being a nominative she rather than an accusative her, and the verb is marked by a third person singular agreement marker -s: (8) She walks regularly. 27 28 A History of the English Language It is more common, however, for Modern English nouns and verbs not to be marked for case and agreement, as in (9a). The word order has to be strictly observed, however, and (9b) is ungrammatical (indicated by *): (9) a. Rabbits eat mallow without hesitation. b. *Mallow without hesitation rabbits eat. In many languages, objects are signaled by a special case marking – him, me, us; sometimes, there is a marking on the verb as well. In Old English, there are cases other than nominative and accusative, namely genitive and dative. This will become relevant in Chapter 4; right now we will examine only the background to cases and a few other grammar points. In Old English, objects get accusative, dative, or genitive case. In the glosses, these are acc, dat, and gen, respectively. The nominative will be abbreviated as nom. Nowadays, the dative of Old English is often replaced by the prepositions to or for and the genitive is replaced by the preposition of. The made-up Old English sentence in (10) would be translated as (11) in Modern English, with the ending -e on cyninge ‘king’ replaced by the preposition to. (If the ending cannot clearly be separated, I use a period between the word and the grammatical abbreviation; if it is clear, I use a hyphen). (10) þæt folc geaf cyning-e aþ-as that people.nom gave king-dat oaths-acc (11) The people gave oaths to the king. In Old English, the main functions of the nominative (nom) are subject – se cyning in (12) – and subject predicate, se cyning in (13). (12) Se cyning for ofer Humbre muþan the.nom king.nom went over Humbre mouth ‘The king went across the mouth of the Humber’. (13) þæt is se cyning that is the.nom king.nom ‘that is the king’. (adapted from Chron 867) (adapted from Alfred) The main functions of the genitive (gen) are (a) to express possession, engles in (14), replaced by of in Modern English; (b) to indicate objects after certain verbs, where only part of the object is involved, as in (15); and (c) to indicate measure or number, as in (16). fingr-um awritene (14) mid engl-es with angel-gen finger-dat.P written ‘written by the fingers of an angel.’ (15) Ic gyrnde þa-ra fisc-a I desired those-gen fish-gen ‘I wanted some of that fish.’ (Wulfstan Homilies) (adapted from the Blickling Homilies) Chapter 2. English spelling, sounds, and grammar (16) 7 þær forwearþ cxx scip-a æt Swanawic and there perished 120 scips-gen at Swanwick ‘and 120 ships perished at Swanwick.’ (Chronicle A, for the year 877) The dative (dat) case is used for the object of most prepositions, as in (17), the indirect object, cyninge in (10), the regular object with certain verbs, as in (18), and to express means or manner, as in (19). (17) Her on þys-um gear-e for se micla here Now in this-dat year-dat went the great army ‘In this year, the great army went’. (18) ðæt heafod sceal wisian þæm fot-um the head shall guide the.dat feet-dat ‘The head shall guide the feet.’ (PC for the year 892) (from Pastoral Care 131.22) (19) sweord-e ne meahte on ðam aglæcean wunde gewyrcean sword-dat not might on that creature wounds make ‘with a sword he could not inflict wounds on that creature’ (Beowulf 2904–5) The accusative (acc) is often used as object, aþas in (10), and object of a preposition, ofer in (12) and geond ‘through’ in (20), when the preposition indicates movement. (20) geond þa wud-as and þa feld-as through the.acc woods-acc and the.acc fields-acc ‘through the woods and the fields’. We will use these cases a lot more in the chapter on Old English. A summary of the cases and their main functions is given in Table 2.7. The subject predicate is marked by nominative case in Old English but this is not true for all languages, so I have ignored that in the table. Table 2.7 Cases and their main functions Nominative Genitive Dative Accusative function example subject possession; object object of a preposition; indirect object object I see the examples the roof of the house; I ate of the apple mit mir (German) [with me.dat]; Give me some apples The woman saw them. Verbs in Old English also have endings to indicate which noun is the subject of the sentence. This is called verbal agreement and the endings are considered inflectional endings. As mentioned earlier, we will use the word ‘ending’ rather than ‘inflection’. We will leave Old English verbal endings until Chapter 4, however, since they are more straightforward than case endings. 29 30 A History of the English Language Languages – or their speakers, to be more precise – perceive words as belonging to certain categories. The main lexical categories are Noun (e.g. table), Verb (e.g. see), Adjective (e.g. yellow), Adverb (e.g. quickly), and Preposition (e.g. across). These categories are called lexical because they carry meaning (they have synonyms and antonyms). There are also grammatical categories: Determiner (e.g. the, a, and those), Auxiliary (e.g. might), Coordinator (e.g. and), and Complementizer (e.g. because). These categories are called grammatical since they determine the syntactic relationships in a sentence. Prepositions and adverbs do a little of both. The distinction between lexical and grammatical categories is important because the change from Old to Modern English involves an increase in the number of grammatical categories, as mentioned in Section 2 of Chapter 1, a process often referred to as grammaticalization. The use of grammatical categories is typical for analytical languages. When languages borrow new words, those words are usually nouns, verbs, and adjectives, i.e. lexical categories. Relatively recent examples are the nouns pizza, angst, patio, pita and sudoku. Therefore, the difference between lexical and grammatical categories is often rendered in terms of open and closed categories, the lexical categories being open (new words can be added), the grammatical ones being closed (new words are not easily added). Prepositions are an in-between category and are borrowed very infrequently. 6. Conclusion This chapter provided background information on the spelling, sounds, and grammatical terminology of English. Because the spelling is irregular, we need a phonetic alphabet. We also need to know something about how English sounds are produced since that allows us to explain and describe language change. Syntax and morphology have also been introduced, and the Old English case system explained briefly; we will go into this more in Chapter 4. Keywords irregular spelling; standardization; Great Vowel Shift; phonetics; manner, place, and voicing of consonants; height, frontness, and length of vowels; assimilation; dissimilation; epenthesis; metathesis; morphology; syntax; lexical and grammatical categories; analytic and synthetic; derivational and inflectional endings; nominative, genitive, dative, and accusative case; agreement for number and person; lexical and grammatical categories. Chapter 2. English spelling, sounds, and grammar Exercises Spelling 1. Look at texts A and B below in which certain sounds have been left out. Which words do you recognize and which text is easier to read? What implications does that have for possible spelling reforms? Text A Th Mn M Hv Wtr Scntsts thnk th hv dtctd wtr n th Mn. Sddnl, vsns f ppl lvng n lnr clns tht stp ff t rfl n th w t Mrs r lss fr-ftchd. ftr tw yrs f crfl nlss, scntsts sd ystrd tht rdr sgnls frm n mrcn spccrft ndctd th mn ws nt bn-dr. Th spccrft’s rdr sgntrs sggstd th prsnc f wtr c n th prmnntl cld shdws f dp bsn nr th lnr sth pl. Text B e oo ay ae ae iei i ey ae eee ae o e oo. uey, iio o eoe ii i ua ooie a o o o eue o e ay o a ae e a-ee. Ae o ea o aeu aayi, iei ai eeay a aa ia o a Aeia aea iiae e oo a o oe-y. e aea aa iaue uee e eee o ae ie i e eaey o ao o a ee ai ea e ua ou oe. 2. Which words do you think are most commonly misspelled? Look in the OED to see if these words have had different spellings over the centuries (see also www.barnsdle.demon.co.uk/ spell/error.html for some common ‘errors’). 3. Discuss arguments for and against spelling reform. Phonetics 4. Please write the symbol for a voiced bilabial stop: [ ] voiced affricate: [ ] 5. [k] and [g]: [d] and [g]: t d z k s g k b d æ a i aw Please describe the following sounds in terms of voice, place, and manner. [k] 8. high back long vowel: [ ] low front short vowel: [ ] Circle the sound that does not fit in the sets of sounds and say why. a. b. c. 7. ] What feature (voice, manner, place) distinguishes [b] and [m]: [p] and [f ]: 6. voiceless dental fricative: [ velar nasal: [ ] [ n] [f ] [d] Read the following line aloud as best as you can. Is your pronunciation different? [læŋgwid( ız ə fəndəm/ntəl hjumən fækəlti juzd fər krietiv ikspr/ʃən fes-tu-fes kəmjunəkeʃən, s"jəntifik inkwəri ænd m/ni #ðər pərpəsəz]. 31 32 A History of the English Language Changes 9. How could you best describe the differences between: a. b. c. d. e. Old English hlaf and loaf? German brennen and English burn? Old English thurgh and Modern English through? Early Latin inpossibilis and Late Latin impossibilis? Old English heofod and Modern English head? Grammar 10. Identify the subjects and direct objects in the text (adapted from the Wikipedia article on Gila Monsters): The Gila monster is a species of venomous lizard indigenous to the southwestern United States and northwestern Mexican state of Sonora. It is a heavy, slow-moving lizard and the only venomous lizard native to the United States. The animal produces venom in modified salivary glands in its lower jaw, unlike snakes, whose venom is produced in the upper jaw. The Gila monster lacks the musculature to forcibly inject the venom; instead, the venom is propelled from the gland to the tooth by chewing. Because the Gila monster mainly eats eggs, small animals, and otherwise “helpless” prey, it is thought that its venom evolved for defensive rather than for hunting use. 11. Using the Old English text and its translation below (from King Alfred’s version of Orosius), try to identify a few subjects, objects, indirect objects, and prepositional objects (dative and accusative). What Modern English words might bude and þeah be related to? In Old English, the æ represents the short a, as in cat, and the ð and þ both represent th. Old English Ohtere sæde his hlaford-e, Ælfred-e cyning-e, þæt he eal-ra Norðmonn-a norþmest bude. He cwæð þæt he bude on þæm land.e norþweardum wiþ þa Westsæ. He sæde þeah þæt þæt land sie swiþe lang norþ þonan; ac hit is eal weste, buton on feaw-um stow-um styccemælum wici-að Finn-as. Word-by-word Ohtere.nom said.sg his lord-dat Alfred-dat king-dat that he all-gen Norsemen-gen northmost lived.sg. He said.sg that he lived.sg on that.dat land-dat northward along the Westsea. He said.sg however that that land.nom is very far north from.there, but it is all waste, except.for few-dat places-dat here.and.there live-P Finns-nom.P (i.e. Sami). Free translation Ohtere said to his lord, King Alfred, that he of all Norsemen northmost lived. He said that he lived in that land northward along the Westsea (sea to the west of Norway). He said, however, that that land is very far north from there, but that it is all a wasteland except in a few places where Sami live (nomadically) here and there. Chapter 3 Before Old English Chapters 1 and 2 briefly explained that English originated around 450, when Germanic tribes first settled in Britain. The Germanic dialects that became English have their origins in another language (or set of languages). Germanic belongs to the Indo-European group of languages, which is itself related to other language groups (e.g. Afroasiatic and Uralic). In this chapter, we will discuss the ancestors of English. In Section 1, we take a step back and examine the origins of humans and when they first might have acquired language, probably 80,000 or more years ago. We also consider how people and languages spread. This section is the most speculative because new (archeological) research surfaces almost weekly. Section 2 examines the earliest writings, from over 6,000 years ago. Changes can be observed between these writings and later ones and we assume this reflects how the spoken language changed. Some observations regarding these changes led to the formulation of linguistic laws in the 19th century, as discussed in Section 3. Section 4 shows that Old English inherits its synthetic character from its predecessors; Section 5 reviews a number of methods for reconstructing an earlier stage of a language when we have no access to written material; and Section 6 addresses some broader questions. 1. Origins of language Scholars are uncertain when humans first started using language and how it spread. In this section, we will examine early archeological evidence, genetic findings, and linguistic reconstruction in an attempt to shed some light on these questions. Humans and chimpanzees split off from a common ancestor probably 7 million years ago. They split up in many branches of early humans. Some of these descendants die out and there is a lot of debate as to which of the early hominids that have been found is the ultimate ancestor of modern humans. For instance, the fossil named Ardi was found in Ethiopia and dates to 4.4 million years ago and the famous fossil Lucy was an early (bipedal) hominid that lived over 3 million years ago in Ethiopia. Because we cannot get DNA from most very old remnants, we don’t know if these hominids are the ancestors of modern humans. They precede homo ergaster (found in Africa) and homo erectus (found in Asia). The latter in turn is the ancestor of the Neanderthals in Europe and Asia. Present-day humans, also known as homo sapiens (sapiens), descend from homo erectus. Homo erectus originated two million years ago and early homo sapiens is argued (based on DNA dating and fossil evidence) to have split off 500,000 years ago, long before language originated. The skull of early homo sapiens does not yet resemble that of present-day 34 A History of the English Language humans; that of later homo sapiens does. This resemblance starts developing around 100,000 BP at the point when the culture of homo sapiens becomes diverse. That is why language is assumed to have appeared between 150,000 and 80,000 BP. Among the earliest humans in Europe and Asia are the Neanderthals, who may or may not have had language but who made music, cared for the sick, and buried their dead. A 2010 article in Science provides a draft sequence of the Neanderthal genome and argues that the Neanderthal ”shared more genetic variants with present-day humans in Eurasia than with present-day humans in sub-Saharan Africa, suggesting that gene flow from Neandertals into the ancestors of non-Africans occurred before the divergence of Eurasian groups from each other” (Green et al. 2010: 710, who spell Neandertal without the h). The issue of gene-flow between our immediate ancestors and Neanderthals may remain controversial for some time. More generally, the Out-of-Africa Replacement Model argues that each wave of peoples replaced earlier populations whereas the Multiregional Continuity Model argues there was gene-flow after homo erectus left Africa and that homo sapiens may have emerged in different regions. Some background can be found at http:// actionbioscience.org/evolution/johanson.html. To see how humans are related and how they spread around the world, geneticists have examined relationships in the genetic material of people from different continents, archeologists have studied early habitation sites, and anthropologists have looked at physical characteristics such as teeth. Figure 3.1 represents some of the genetic relationships. The genetic similarities and differences between the various populations suggest that humans migrated from Africa to Australia and Oceania, then to Asia, then to Europe and to America. Africa Asia America Europe Australia/Oceania Figure 3.1 Genetic relationships between people (from Cavalli-Sforza 2000: 39) This figure implies the greatest affinity between the people of Asia and those of the Americas, indicating that the Americas were settled by Asian (Siberian) peoples. Looking at changes in mitochondrial DNA (inherited through the female), the ancestors of modern humans may have moved as in Figure 3.2. Mutations in Y-chromosome DNA (passed from father to son) point to similar migrations. Hypotheses about genetic relationships, as in Figure 3.1, and migrations also predict relationships between languages. This means that language likely originated in one place: it is monogenetic rather than polygenetic. The reason behind this monogenetic assumption is that the world languages have similar properties. Chapter 3. Before Old English EUROPE Z H,U,X T,U,V,W G ASIA I,J,K AFRICA Y X M B B N M L2 A, D A,C,D NORTH AMERICA F B L3 M X M L1 N A,C,D B AUSTRALIA A,C,D SOUTH AMERICA Figure 3.2 Mitochondrial DNA in various populations (from http://en.wikipedia.org/wiki/Human_mitochondrial_DNA_haplogroup) If humans already had language by the time they started migrating from Africa, Figure 3.1 predicts that the languages of Asia are closest to those of the Americas, and the languages of Africa and Australia the most distinct since they developed independently of each other over a longer period of time. Linguistic work confirms this prediction. For instance, Greenberg, Ruhlen, and others have linked the genetic tree in Figure 3.1 to a tree of linguistic relationships between most of the world’s languages, i.e. Figure 3.3. This grouping is controversial, as will be discussed later. If a language is not attested in early writings, we call it a proto-language; all the languages/families below are proto-families but I have added a few contemporary languages and families to make it clearer. Eurasian Dene-Caucasian (Navajo, Chinese, etc) Nostratic Eurasiatic (e.g. IE, Uralic, Eskimo-Aleut) Afroasiatic (e.g. Hebrew) Figure 3.3 Linguistic relationships (Ruhlen 1994: 192) Amerind (e.g. Hopi) Dravidian (e.g. Tamil) 35 36 A History of the English Language The connection of the branches indicates which languages are most closely related. The Afroasiatic languages include North African and Semitic languages: Arabic, Hebrew, and Berber. They are argued to be related to Dravidian, which includes languages such as Tamil, Malayalam, and Brahui. Dravidian is the name for a group of languages that was at one point spoken in Northern India but that was later pushed to the South of India (and to Sri Lanka). If you consider this migration, the assumption that the Dravidian and Afroasiatic families are related is not far-fetched. Eurasiatic incorporates eight language families from Indo-European to Altaic. It includes English, Italian, Bulgarian, Japanese, Korean, Finnish, and Inuktitut. Together with Afroasiatic and Dravidian, Eurasiatic forms a well-known super-family – Nostratic – an idea advocated by Illich-Svitych and Dolgopolsky in the 1960s and by Shevoroshkin, Ivanov and others more recently. The next most closely related family is Amerind. This is a large (and controversial) language family; Greenberg suggests that it includes most of the languages of the Americas – Hopi, Nahuatl, and Quetchua. In this scenario, Na-Dene (including Apache and Navajo) is grouped with Sino-Tibetan and North Caucasian into Eurasian. The only language families not accounted for in a superfamily are the three African language families – with the exception of Afroasiatic – the Australian families, and some Pacific ones. The linguistic representation in Figure 3.3 mirrors the genetic ones in Figures 3.1 and 3.2 only up to the split between Asia and Europe. As mentioned, it leaves out three African language families as well as the Polynesian and Australian languages. If Figure 3.1 is correct, Australian and Polynesian languages are the first to split off and are therefore most distantly related to the Eurasian languages. There is a lively debate on how much to include in Eurasian and Nostratic (e.g. Greenberg 2000) and whether this kind of reconstruction is possible at all. One linguistic argument against such a reconstruction is that reconstruction going more than 6,000 years back is not possible because most words will have been replaced in such a long period (e.g. Kaufman 1990; Ringe 2002). Another argument is that genetic and linguistic similarity need not go hand-in-hand. Thus, languages can be replaced without the relevant genes being transmitted and vice versa. For instance, the language of the Sami in Northern Scandinavia is in the same family as that of the Finns but the genes of the two peoples are different. The most significant debate about reconstructions such as the one in Figure 3.3 is probably the one surrounding the existence of Amerind(ian). The assumption that Na-Dene and Eskimo-Aleut are language families of the Americas is mostly accepted but the one that Amerind is one family is not. The alternative to Amerind is to posit 200 or so separate language families. In short, genetic and linguistic reconstructions indicate that certain people and languages are more closely related since they separate at later points. Some researchers (Cavalli-Sforza, Greenberg, and Ruhlen) argue that genetic and linguistic relationships go hand-in-hand; others (Ringe and Kaufman) criticize such an approach. The debate about the origin of language has always been full of speculation so much so that the Linguistic Chapter 3. Before Old English 37 Society of Paris banned discussions on this topic in 1866. The controversy regarding the languages of the Americas is still very strong. Next, we turn to more tangible evidence of linguistic relationship – written records. 2. Earliest writings In this section, we discuss older writing systems and writing systems in general. Early writings provide evidence of linguistic change, which will be discussed further in Sections 3 and 4. There is a large gap between 70,000 years ago, when humans probably had started to use language, and the time from which we have historical evidence for language in the form of writing. Writing found in Henan, China dates back 8,000 years and the Vinca inscriptions found widely over southeastern Europe date back 7,000 years.

English Spelling, Sounds, and Grammar PDF

Document Details

Tags

Related

Summary

Full Transcript

Upgrade to continue