Historical Linguistics:
The Study of Language Change
Chapter 8

    This excerpt taken from Chaucer in the late 14th century shows the vast differences in how English has changed over the last seven centuries. We will seek to help answer how and why languages change over time.
Historical  linguistics   is concerned with both the description and explanation of language change. In this chapter we examine the nature and causes of language change and survey in some detail phonological, morphological, syntactic, lexical, and semantic change. We also explore techniques used to reconstruct linguistic pre-history and briefly discuss interrelated research into language acquisition and linguistic universals.

    All languages undergo change to some degree over time. English has gone through three major periods of change: Old English (roughly 450-1100), Middle English (from 1100 to 1500) , and Modern English (from 1500- to the present)

•  Pa sendan hi ham aerenddracan. (Old)
•  then sent they home a messenger (Middle)
•  ‘Then they sent home a messenger.’(Modern)

1.1 Systematicity of Language Change
     In some instances, language change is consistent. Some words in Modern English such as boat, oath, and stone, were all used with a different vowel sound in Old English as well as Middle English. The vowel sound used in all three words changed at the same time in each stage of English. Another consistent change would be the placement of the subject-verb-direct object used in most sentences in Modern English.

1.2Causes of Language Change
    Since children are not born with language, only the capability to acquire it, language change is passed on from generation to generation. Articulatory simplification (a process that facilitates acquisition) is seen in everyday speech such as the deletion of a consonant in a complex cluster or the insertion of a vowel to break up a complex cluster.
 Ex. Pronouncing ‘fifs’ for ‘fifths’ (deletion ) or ‘athalete’ for ‘athlete’ (insertion of a vowel).

Spelling pronunciation ( one factor in sound change , where a new pronunciation reflects the spelling of the word) shows how a new pronunciation can appear that better represents the way a word is spelled. Take  for example the word ‘often’. The  ‘t’ was pronounced in earlier English, dropped later to sound like ‘soften’, and then reintroduced to be pronounced once again with the ‘t’ in modern English since the ‘t’ was retained in the spelling.
Sound change (a systematic change of sounds that took place over a long period) would be best described in the words ‘assume’ and ‘assure’. Some sounds changed in words even though they had the same spelling as others. There is the /s/ sound in ‘assume’ and a ‘sh’ sound in ‘assure’.
     Two areas of cognitive based change are analogy and reanalysis. Analogy “involves the extension or generalization of a regularity on the basis of the inference that if elements are alike in some respects, they should be alike in others as well”. For example, if it is ‘sting’ and ‘stung’, why not ‘bring’ and ‘brung’ , or children substituting, ‘goed’ for ‘went’.Reanalysis (a source of language change that involves an attempt to attribute an internal structure to a word that formerly was not broken down into component morphemes) accounts for the way we use the word ‘hamburger’. Originally naming a specific meat patty, it derived its name from the city of Hamburg, Germany. The word was later seen to have two components, ‘ham’ and ‘burger’, thus leading  us to naming ‘chickenburgers’, ‘fishburger’ and even ‘burger’. Even though the original word had no meaning as separate components, we use the word ‘burger’ to represent a type of food.
     Another cause of language change is language contact (interaction between speakers of one language and speakers of another language or dialect). This can result in the borrowing (a source of language change that involves adopting aspects of one language into another) of words from other languages as influenced by location or proximity. Amerindian words such as ‘Canada’, ‘moccasin’, ‘totem’, ‘pecan’ and ‘skunk’ have been incorporated as part of the English language. New phonemes or allophones can result from borrowing words from another language. In the early Middle English period, the London dialect had /f/ but not /v/ in the word-initial position. The /v/ was later introduced through contact with French.
Hypercorrection (overgeneralization of particular rules in a language in an attempt to speak ,or write, correctly) is a language change that is found when a speaker who is trying to speak another dialect or language, overgeneralizes particular rules, perhaps by changing a pronunciation or word order. One may say ‘pro/t/igy’ for ‘prodigy’ or misuse the pronoun ‘I’, i.e.. ‘He saw John and I’ instead of ‘He saw John and me’.

2.0  Sound change

     Variation and change in a language are very noticeable in the phonology of a language. The linguistic processes underlying such phonetically conditioned  sound change (Sound change that begins as subtle alterations in the sound pattern of a language in particular phonetic environments) are identical to the one found in the phonology of currently spoken languages. The application of such processes usually brings about an articulatory simplification and, over time, significant changes in the phonology of a language can result. Two types of change aresequential change (Sound change that involves sequences of segments, e.g., assimilation) and segmental change (a sound change that affects a segment).
 
 
 

2.1  Sequential Change
Assimilation (the influence of one segment on another, resulting in a sound becoming more like a nearby sound in terms of one or more of its phonetic characteristics) is the most common type of sequential change. In the following, the four main types of assimilation  (as listed in the ‘catalogue of sound changes’ above) are discussed.

Note:
Partial assimilation  (a phonological process by which neighboring segments become more like each other, e.g., by sharing the same place and manner of articulation.)

Total assimilation (the assimilation of all the features of neighboring segments)
 

3.1 Addition of affixes
      There are several ways to create new words through the addition of affixes.  The first is borrowing from other languages.  The Middle English period had many French words with the suffix –ment (e.g., accomplishment, commencement).  Soon –ment became a suffix in English that was used with bases that were not of French origin (e.g., acknowledgment, merriment).  The suffix –able, which changes a verb into an adjective (e.g., readable, lovable) were first borrowed into English as entire units, soon the suffix was used with new bases. The second is called grammaticalization.  This change occurs when lexical forms become grammatical forms over a period of time.  For example,
The third way,fusion is when two words that are frequently adjacent, over time, they can become fused together to form a single unit consisting of a base and an affix.  The example shows base + affix = suffixation or affix + base = prefixation.  Several Modern English suffixes came from earlier words by this means such as:

3.3 From synthetic to analytic to synthetic
     Morphology in many languages differs greatly.  Linguists use the terms analytic and synthetic languages to distinguish the formation of words. Analytic languages have words with single (root) morphemes.  There are no affixes and categories such as number and tense so a separate word must be used to express this meaning.  Languages such as Mandarin Chinese, Cantonese, Vietnamese, Laotian, and Cambodian use morphemes to indicate a past action which act like an independent word (e.g., Mandarin Chinese)
 

Synthetic languages use words containing a root and one or more affixes.  For example, in the Spanish, words carry endings that distinguish between masculine and feminine, singular, plural and tense.

     When changes do not occur due to borrowing or sound change, the method of fusion keeps the ebb and flow of words over time.  Fusion allows new synthetic forms to appear.  Such as in some Modern English dialects coulda (e.g., I coulda won), represents the fusion of the words could and have.  The –a is treated as a suffix that is no longer related to have.  This is evident in spellings such as could of, which caused confusion over how to represent the pronunciation of coulda in written English.  Fusion provides a language with analytic morphology to become more synthetic over time.  The English language has developed from a synthetic language with many inflectional affixes to an analytic one with very few since the loss of case endings through sound change as shown in the example of the word hound.

3.4 Analogy
Analogy is another manner in which the morphology of a language is effected.  This source of change involves the generalization of regularity, that if elements are alike in some respects, they should be alike in others as well.  The example of the Old English plural of hand was handa.  Vowel reduction changed the word in Modern English plural and singular to hand.  The Modern English plural hands could not be the consequence of sound change.  It is the result of earlier analogy with words such as Middle English hund  ‘hound’ which did form the plural with the suffix –s.  Some other examples of plural forms created on the basis of analogy included eyes (eyen in Middle English) and shoes (shooen).  Every generation of English speaking children produce forms such as sheeps, gooses, and mouses.  As of now these have not been accepted by the adult speakers of Standard English and then are usually forgotten by young language learners as their creativity falls on deaf ears.

3.5 Reanalysis
     Reanalysis can produce new patterns and structure in words. Folk etymology is a type of reanalysis that is not based on a historically correct analysis of a word and does not involve a detailed study of the word on the part of the speaker.  It does produce changes in pronunciations reflecting the new morphological analysis.  For example, the word earwig derives from Old English earwicga (        ), a compound of ‘ear’ and ‘insect’.  Through sound change this word became earwidge (   ).  The second part of the compound was lost as an independent word by Middle English so speakers did not relate it with the meaning insect.  It changed to ‘wiggle’ resulting in Middle English arwygyll (ear + wiggle’).  The final word became –wig and not –widge in Modern English.

4.1 Word Order
      All languages distinguish between the subject and direct object.  This difference is shown through case marking or word order.  Old English with its many case markings had more variety than Modern English.  The subject-verb-object order in simple transitive sentences was such as this example.

When the clause began with  pa ‘then’ or ne ‘ not’, the verb remained in the second position but now the subject came first.
 

This word order is still found in Modern English but has limited use and subject to special restrictions, unlike in Old English
 

When the direct object was a pronoun, the subject-object-verb order was common.
 

The subject-to-object-verb order also was seen in embedded clause, even when the direct object was not a pronoun.
 
 
 

      During the Middle English time period case markings were lost through sound change and the fixed subject-verb-object order was the main way words were formed.  A major change in word order took place between 1300 and 1400 with the verb –object order becoming dominant.

From SOV to SVO

     Languages can be grouped on the basis of the order of subject (S), object (O), and verb (V) in basic sentences.  Most all languages of the world have one of three types:  SOV, SVO, or VSO with the majority of languages being one of the first two types.  Over time languages change from one morphological type to another, as well as from one syntactic type to another.  In the history of English the development from SOV to SVO syntax is evidence that indicates that the earliest form of Germanic from which English descended was an SOV language.  One of the first recorded Germanic sentences gives this example of the word order.

     The change from SOV to SVO is found in many other languages.  Linguists still are not sure why languages change from one syntactic type to another.  The order of verb and object (OV versus VO) has been connected with other word order patterns.

4.2 Inversion in the History of English

5 Lexical and Semantic Change
      Another way language changes are through lexical change of addition and loss of words.   This reflects cultural changes that introduce novel objects and notions, and that eliminate outdated ones.

5.1 Addition of Lexical Items
      The addition of new words comes about through word formation processes or through borrowing.  The addition is usually due to technological innovations or contact with other cultures.  These developments become lexical gaps that are filled by adding new words to the lexicon.  Our mental dictionary has to be readjusted and adapts to the new.
Word formation processes such as compounding and derivation are the most important ways to make new words.  Other types also play a role, such as conversion, blends, backformation, clipping, and acronyms.  These processes have been in use for Old English as well as Modern English.  Even though many Old English compounding and derivational patterns have been maintained in Modern English, words that were acceptable in the Old are not necessarily still in use now in the Modern, even though they are understandable (e.g., ‘elf’ + ‘beautiful’ =  ‘ beautiful as a fairy’).
Not all word formation processes were available to Modern English.  The conversion type was not available to synthetic inflectional languages such as Old English since the change in a word category in these languages is usually indicated morphologically and conversion does not involve the use of affixes.

      Borrowing words from one language to another is an important source of new words.  If you take into consideration the cultural relationship between languages, three types of influence of one language on the other are substratum, adstratum, and superstratum influence. Substratum influence is the effect of a politically or culturally nondominant language on a dominant language in the area. This influence does not usually have an impact on the lexicon of the borrowing language.  Usually borrowed words are restricted to place names and unfamiliar items or concepts.  The speakers of the substratum language who inhabited the area first have an influence on the situation. Superstratum influence is the effect of a politically or culturally dominant language on another language or languages in the area.  The Norman French language had an impact on the vocabulary of English is related to a historical event—the conquest of England by French-speaking Normans in 1066.  As the people learned English over the next decades, they retained French terms for political, judicial, and cultural notions.  Native English speakers who wanted to move up in the social class by imitating the speech of the upper class borrowed these same word.

      In some instances, French loanwords were used with native English words to convey distinctions of various sorts.  When the English word theft was used it was for a minor crime and for a more serious crime the French word larceny was used.  The English kept their own words for domesticated animals but used the French words for the meat from those animals.
 

Adstratum influence occurs where two languages are in contact and neither one is clearly politically or culturally dominant.  In a city such as Montreal with its large number of bilingual speakers, English and French have a great influence on each other (e.g., Montreal English  metro—‘subway’, autoroute—‘highway’).  In adstratum influence common, everyday words are borrowed making it difficult to distinguish between the borrowings from one language to another.  In the early history of English, the Scandinavians settled in part of England and had frequent contact and interaction.  Some loanwords from Scandinavian that most English speakers would not know their origin are for example: (e.g., anger, cake, call, egg, husband, low, root, score, seat, skill, take their, they ugly, window, wing). Here are a few lexical borrowings into English: Italian, motto, pizza; Spanish, tornado, guitar; German, poodle, pretzel; Dutch, cookie, cole slaw; Slavic, tundra, polka; Hindi, punch, shampoo.

5.2 Loss of Lexical Items
     The loss of words are due in part to the changes in society and the object or notion the words refer to have become obsolete.  Some Old English words lost through cultural change were:  peox—‘hunting spear’, eafor—‘tenant obligation to the king to convey goods’.

5.3 Semantic Change
     Although changes in word meaning take place continually in all languages, words rarely jump from one meaning to an unrelated one.  Usually, the changes are step by step and involve one of the following phenomena.
Semantic broadening happens when the meaning of the words becomes more general than its earlier form.

Semantic narrowing happens when the meaning of a word becomes less general or less inclusive than its earlier form.

     In amelioration the meaning of a word becomes more positive or favorable (e.g., pretty/old meaning—tricky, sly, new meaning/ attractive).  The opposite change is called pejoration (e.g., silly/old meaning—happy, prosperous, new meaning—foolish; wench/old meaning—girl, new meaning—wanton woman, and prostitute).
    The weakening of meaning frequently occurs changing the meaning from one of exaggerated meaning to a watered down version, as in the word soon.  Earlier it used to mean ‘immediately’ but now simply means ‘in the near future’.
.

     Semantic change is triggered often by metaphor, a figure of speech that is based on a perceived similarity between distinct objects or actions.  This change involves a word with a concrete meaning that moves to taking on a more abstract meaning.  The original meaning is not lost and the word just adds another dimension to its meaning.

Linguistic innovations spread through the vocabulary of the language and through the population.

6.1 Diffusion through the Language
Lexical diffusion manifests itself in a few words and then slowly spreads through the vocabulary of the language.  In English the ongoing change in the stress pattern of words can be used as either a noun or a verb.  The stress originally fell on the second syllable without regard to it's lexical category.  In the late sixteenth century three such words, rebel, outlaw, and record came to have the stress on the first syllable when used as nouns.  This stress shift began to increase over the next decades.  The entire English vocabulary has not diffused and the stress is still on the second syllable (e.g., report, mistake, and support).  This change is ongoing and may continue to work its way through the language until all nouns in the class we have been considering are stressed on the first syllable.  Another type of change is the sound changes of segments.  In some dialects of Spanish (e.g.,Cuban) the consonantal weakening of (s) to (h) in syllable-final position affects all instances of (s) in those positions.  This change applies across the board to all words.

6.2 Spread through the population
      For a language change to occur the change must be accepted by the linguistic community as a whole. Sometimes the change begins with a small number of words which appears first with a small number of people.  Social pressures play a role in whether or not the change will stay around and may limit the spread of this innovation.  The most notable example of this in the history of English is the loss of postvocalic (r) along the east coast of the United States.

       Some languages share striking similarities.  These languages are compared and analyzed and often found to be genetically related. Genetically related languages are "languages that have descended from a common parent (e.g., German and Italian have both descended from Indo-European)."   The comparison process is called comparative reconstruction. Comparative reconstruction is "the reconstruction of properties of a parent language through comparison of its descendant languages."

7.1 Comparative Reconstruction
      When comparing languages, it is often found that they share many systematic phonetic correspondences. Systematic phonetic correspondences are "sound correspondences between two or more related languages that are consistent throughout the vocabularies."  Some languages that share these correspondences are German, English, Dutch, Danish, and Swedish which are all of members of the Germanic language family.
 

    Because the words share similar forms and meanings it is obvious that they descended from the same source.  These same words in an unrelated language would not share similar phonological form.  The words below are from the Turkish language which is not part of the Germanic language family.
 

    Words that descend from the same family and share similar semantics are called cognates. Cognates are "words of different languages that have descended from a common source, as shown by systematic phonetic correspondences (e.g., English father and German vater)."  Some genetically related languages are distantly related and do not share strong similarities.  Three of these languages are English, Russian, and Hindi.
 

7.2  Techniques of Reconstruction
       Because the process of language change is systematic, reconstruction is not difficult.  Although all components of the Proto-language can be determined, linguists focus primarily on the phonological reconstruction.
    There are two general strategies used to reconstruct the proto-language.  The main one is the phonetic plausibility strategy.  The Phonetic plausibility strategy is "the primary strategy used to reconstruct proto-forms that requires any sound changes posited to be phonetically plausible.  (See also  Majority Rules Strategy.)" The second strategy is the majority rules strategy.  The Majority rules strategy is "a secondary strategy used to reconstruct proto-forms which stipulates that the segment found in the majority of cognates should be assumed to be part of the proto-form.  (See also Phonetic plausibility strategy.)"
    Some sound changes are more plausible than others.  These changes involve palatalization, assimilation, voicing, and consonant deletion.
    When reconstructing proto-Romance, it was found that some languages, such as Sardinian, are more conservative, retaining more of the earlier consonants and vowels. Conservative denotes "a language that show comparatively few changes over time". Some of the word forms in Sardinian were the exact same word in the Proto-forms.  Sometimes it is not possible to reconstruct all of the linguistic aspects of a language.  Some of the earlier languages do not have a traceable written record.

7.3 Internal Reconstruction
     Often a language may be reconstructed without using comparative data. Internal reconstruction is "the reconstruction of a proto-language that relies on the analysis of morphophonemic variation within a single language."  In these, the sound changes create allomorphic and allophonic variations.  These incidents occur in English with w[k] and [sh] as in magic and magician.  Changes like this are phonetically plausible.

7.4 The Discovery of Indo-European
      When Sanskrit was discovered to share amazing similarities to Latin, Greek, Germanic, and Celtic, and a new study began to find the proto-language.  After decades of historical-comparative work and extensive studies in phonetic correspondences, it was discovered that most of the languages in Europe, Persia (Iran),  and the northern part of India were members of the same language family, known as Indo-European.  This language in proto-language. Proto-Indo-European (PIE) is "the protolanguage from which evolved most of the languages of Europe, Persia (Iran), and the northern part of India.

    Another linguist, Karl Verner, made another important discovery.  He found a group of exceptions to Grimm's Law. Verner's Law is "a generalization made by Karl Verner, which states that a word-internal voiceless fricative resulting from Grimm's Law underwent voicing if the original Proto-Indo-European accent did not immediately precede it".

7.5 Reconstruction and Typology
    During the past one hundred years, linguists have acquired extensive information on thousands of languages.  This can be attributed to the  extensive studies that have been performed in the field of linguistic typology.  Linguists have begun to examine languages which are not genetically related, but share similar structures.  They have found that these languages share  common structure such as word order and vowel harmony with languages in other families.
    This information all came about after the discovery of Proto-Indo-European.  Linguists today who are involved in the area of typological studies, are very interested in the role of typological plausibility. Typological plausibility is 'a criterion that guides language reconstruction by referring to universals or existing properties of language."
    Some linguists have determined that according to traditional reconstruction the consonant system of PIE should not be accepted.   It is questionable in at least two areas.  They are that reconstructed forms in PIE are uncommon and that traditional reconstruction posits a group of voiced aspirated stops without a subsequent series of voiceless aspirated stops.  Some linguist insist that all languages do not need a series of corresponding stops.
    Though both forms of reconstruction have their individual strengths, Typological plausibility will most likely remain the secondary form until it is possible to determine what is linguistically possible and impossible.

    Language change generally repeats the same patterns, across time and even across languages.  One theory shows that some of these changes are more natural than others. Naturalness is "the criterion that guides language reconstruction by determining whether or not changes are natural."
    Presuming that naturalness is one of the criterion of language change, evidence should be visible in the process of language acquisition.  One of these examples is the CV (consonant/vowel) syllable.  The following three factors support this theory.

• The CV combination is universal.  Every language in the world has examples of this syllable.
• Within the changes of sound in a language, those syllable types that are less natural are often converted to CV because of its naturalness.
• CV is the predominant syllable combination in language acquisition.

2.  a)  Describe the difference between the two French dialects in the following data.  Assume that the data are in phonetic transcription.
    b)  What sound change would you posit here?  Why?
    c)  State the sound change in the form of a rule?
  Standard French    Acadian French
i)   kav      kav      'cave'
ii)  kurir    kurir       'run'

3. a)  What sound changes differentiate Guaraní from its parent language, Proto-Tupí-Guaraní in the following data?
    b)  State these changes in rule form.
Proto-Tupí-Guaraní     Guaraní
    i)  jukir        juki        'salt'
    ii) kib          ki           'louse'
 

Dr. Olson's Web Site Links!

Next page,please!!	WCB
Find the right chapter!	E-mail Dr. Judith Olson!