Semantic change

From Wikipedia, the free encyclopedia

In diachronic (or historical) linguistics, semantic change is a change in one of the meanings of a word. Every word has a variety of senses and connotations which can be added, removed, or altered over time, often to the extent that words of one time period mean quite different things to the same words as spoken in a previous time. Moreover, two words derived from the same original may develop in very distinct ways: cognates across languages often look very similar but mean entirely different things.

Semantic change is not to be confused with etymology, another field in diachronic linguistics; etymology is the scientific study of word origin, while semantic change deals with the development of sense. In fact, semantic change is one of the factors that need to be taken under consideration, in order to evaluate a proposed etymology.

1 Limitations of historical semantics
2 Causes of semantic change
3 Types of change
- 3.1 Multiple processes
4 History
5 Notes and References
6 See also

[edit] Limitations of historical semantics

In recent years semantic change has received a large amount of attention, especially within the framework of pragmatics and cognitive linguistics. Nevertheless, the quest for a standard taxonomy of semantic changes, which would adequately account for all attested mechanisms, is still in doubt. Many linguists (chiefly comparatists) would agree with R.S.P. Beekes' assertion, that "studies on change of meaning do not offer satisfactory results" and no-one has so far presented a full-scale method for interpretation and classification of the data^[1].

There are certain linguistic causes behind this deficiency, related to inherent limitations of historical semantics:

By contrast with the limited number of phonemes and morphemes in a linguistic system, the number of senses is unlimited. This bipolar position has been described as resulting from the co-existence of a closed phonetic/morphological system and an open semantic system^[2].

There are no semantic rules or principles that may exclude a certain change of meaning (according to Antoine Meillet's ascertainment^[3]). Therefore, historical grammars and handbooks, after dealing with the development of phonetic and morphological system of the language, tend to proceed with the study of lexical change or origin (loanwords, lexical diffusion, dialectic split etc.), setting aside any need for a logical classification of semantic changes.

Information on the circumstances that contributed to an alteration are rare and scarce. A linguist cannot retrieve (or verify) valuable evidence concerning style, intonation (the suprasegmental parts of language) and the relevant position of synonyms in a lexical field. Accordingly, he is not able to define and evaluate the sociolinguistic background which may have favoured a certain change or a degree of intentionality that lead to the entrenchment of the new meaning.

Polysemy, which constitutes a prerequisite for semantic change, is based upon the construction of a central (nuclear) meaning of a word and of marginal or peripheral senses, that form a radial network^[4]. In fact, change of meaning has been sometimes defined as drift or shift of semantic load from a nuclear meaning to a peripheral one. Unfortunately, texts and inscriptions tend to be silent on the prototypicality^[5] of the nuclear meaning and on the relevant semantic force of one synonym over another.

[edit] Causes of semantic change

A major contribution to the theory of semantic change has been Antoine Meillet's article Comment les mots changent de sens, published in 1921. This article has influenced many later essays, as it addresses the most important aspects of meaning change. Meillet discusses three causes of semantic change:

Structural (structurelles): This category refers to the linguistic structure of lexical items. The limited number of phonemes / morphemes reduces, as such, the possible contexts for these elements. By striking contrast with the morhophonemic part, there are no a priori context limits related to the meaning of a word, concerning its possible connotations and positions in a sentence. In addition, lexical fields allow for powerful semantic interaction among their members, though the results are usually visible only after the conclusion of the process^[6].

Two noteworthy structural mechanisms that affect semantic change are grammaticalisation and reanalysis.

Referential (referentielles): This category includes changes affecting the referent, i.e. the object or thing that a linguistic unit stands for. Normally, progress in technology and culture goes along with changes in items, materials, tools and concepts. Nevertheless, since language abides to the principle of economy (old means - new usages), a certain delay in following that progress is certainly expectable. The system of any given language will most likely extend the semantic field of an existing word in order to cover the new usage rather than create ex nihilo a new lexeme. The importance of the simultaneous function of lexical items and referents had been the object of a linguistic and ethnographic field back in 1910, which was named Wörter und Sachen (“Words and Things”). Though not reflecting mature linguistic analysis, this field focused on producing a descriptive pattern to account for both the word and its cultural background or, briefly, to connect the history of a word to the history of its referent^[7].

[edit] Types of change

The four most widely recognised types of semantic change are extension, narrowing, amelioration, and pejoration. The first two represent changes in a word's scope, while the second pair can also cover changes in a word's individual meanings.

Extension: Extension is the widening of a word's range of meanings, often by analogy or simplification. For example, virtue was initially a quality that could only be applied to men, like our modern word manliness, but in contemporary society, it can equally be applied to women as well. Maverick used to be a rancher's term for an unbranded cow but can now mean a person who doesn't conform to the conventions of a group (Jeffers & Lehiste).

Narrowing: Narrowing is the reduction in a word's range of meanings, often limiting a generic word to a more specialised or technical use. For example, broadcast originally meant "to cast seeds out;" with the advent of radio and television, the word was extended to indicate the transmission of audio and video signals. Today, because of narrowing, very few people outside of agricultural circles use broadcast in the earlier sense (Jeffers & Lehiste).

Amelioration: Amelioration occurs as a word loses negative connotations or gains positive ones. For example, mischievous used to mean "disastrous", where it now only means "playfully annoying".

Pejoration: Pejoration occurs as a word develops negative connotations or loses positive ones. For example, notorious initially meant "widely known". Yet it has gone through the process of extension to now mean "widely and unfavourably known". A much more famous example is of the word gay, which can mean happy or colorful and was used commonly until it became a reference to homosexuals^[8]. While this may or may not have been a euphemisation in itself, the word in the original sense is avoided. Gay is also extended in certain slang vocabularies as a pejorative adjective. See also euphemism treadmill.

Semantic shift: Semantic shift occurs as a word moves from one set of circumstances to another, resulting in an extension of the range of meanings. An example of this is navigator, which once applied only to ships but, with the development of planes and cars, now applies to multiple forms of travel. Another example is Old English, meat, (or rather mete), which referred to all forms of solid food while flesh (flæsc) referred to animal tissue, and food (foda) referred to animal fodder. Meat was eventually restricted to flesh of animals, then flesh restricted to the tissue of humans and food was generalized to refer to all forms of solid food (Jeffers & Lehiste).

Semantic drift: Semantic drift is the movement of the entire meaning of a lexeme to a new meaning, and is particularly evidenced by semantic differences between cognates.

For instance, the English word to starve is cognate with the German sterben ("to die") and in some parts of England, the word can mean "be cold" (since it evolved through the meaning "to die of cold"). Though both words arose from a common West Germanic root *sterb-a- ("to die"), and their meanings are still somewhat related, semantic drift has caused their specific meanings to differ. The same may occur language-internally, especially when one form is specifically agglutinated. For example, English to hurdle is cognate to hard and is agglutinated with the -le frequentative suffix.

A more extreme example is with the English word black, which is cognate with Slavic words for white (Russian белый); the reconstructed Proto-Indo-European root for both is *bhel. English black derives from Germanic *blakaz, a past participle of a verb meaning "to blaze." As an adjective, the word would indicate something that has burned and since what is burnt is generally black, the shift in meaning makes more sense.

Figurative use: Figurative use is a change in meaning that is based on an analogy or likeness between things. For example, a crane is a bird with a long neck, but the word can now also mean a piece of equipment for lifting weights. The earlier examples of maverick and broadcast are also examples of figurative use.

Grammaticalisation: Grammaticalisation is the development of function words and grammatical affices from content words. It often begins with extension of a word to include a grammatical function, and the subsequent narrowing of the word (usually after the word has suffered morphological changes) to a solely or predominantly grammatical use. An example of this is the French word pas, which literally means "step" but is also used with ne in forming negating statements like je ne pense pas ("I don't think so") as well as by itself: ma voiture a un toit ouvrant, la leur pas ("my car is convertible, theirs isn't").

Metonymy: A type of extension, metonymy or synecdoche is the use of a part of an object to refer to a whole. In many languages, the word for head can be used as a substitute for the word for person. In English, we have the phrase "a head", resembling the Latin phrase "per capita", which we also use. The word "poll", originally meaning the top of the head, can refer to the whole head, and a "poll tax" is a fixed tax applied to each person. The convention of using capital cities to represent countries or their governments is another example of metonymy.

Euphemism: A euphemism is the use of a substitute word in an attempt to replace or mask the negative connotations of the normal word for a certain object or action. The substitute word undergoes an extension, while the word replaced may suffer pejoration by dissimilation. For example, snogging was once an alternative word for sex, though it has now been ameliorated in most registers to mean a french kiss.

Political correctness: Political correctness is a real or perceived attempt to refine or restrict language and terms used in public discussion to those deemed acceptable or appropriate. For example, in the United Kingdom in the early 1990s the term "blackboard" became perceived by some as being "politically incorrect", and so some schools instructed teachers to refer to it as a "chalkboard" instead.^[9]

[edit] Multiple processes

Many words go through more than one process. A good example of this is the word punk. It initially meant "a young boy". However it was then pejorated to mean "prostitute". In the 1970s, the word was then ameliorated to describe a music genre. In some areas, this change has gone even further and "punk" can mean someone who doesn't stand up for himself and thus is pushed around and disrespected.

[edit] History

It is certain that semantic change has occurred since human language first arose. However, the study of semantic change only dates back to the 1900s.

The pioneering work was carried out by the French linguist Michel Bréal, who published his book Essai de sémantique in 1899 (Paris, 2nd ed.). Bréal was soon followed by most linguists, who defined mechanisms of semantic change in harmony with his findings. Accordingly, in 1921 the German linguist Hermann Hirt classified semantic changes under six categories: (a) narrowing (Verengerung), (b) pejoration (Verschlechterung), (c) amelioration (Verbesserung), (d) extension (Erweiterung), (e) metaphor (Metapher), and (f) metonymy (Metonymie)^[10].

A few years later, Jost Trier, a German linguist, who presented his findings in 1934, was the first one to point out the importance of lexical / semantic fields. In his studies, he showed how the structure of the German language had changed between 1200 and 1300.

In 1200, the German language had no separate word for cleverness. It only had Kunst for "courtly skills" and List for "non-courtly skills." The language also included the word Wîsheit for any kind of knowledge.

By 1300, however, things had changed. Wîsheit had been narrowed to just mean "religious experience", kunst was beginning to take on the meaning of "art" or "skill", and List had been removed from the language entirely (it had begun to gain pejorative connotations). List has returned to Modern German, where it now means "cunning" or "trick".

[edit] Notes and References

The Cambridge Encyclopedia of Language (1987) David Crystal.
The Cambridge Encyclopedia of the English Language (1995) David Crystal.
Jeffers, Robert J. and Lehiste, Ilse (1979). Principles and Methods for Historical Linguistics. MIT press.

^ R.S.P. Beekes, Comparative Indo-European Linguistics. An introduction, Amsterdam 1995, p. 140.
^ A. McMahon, Understanding language change, Cambridge 1994, p. 185.
^ “La difference de sens entre les mots rapprochés doit être expliquée par des raisons précises, autant que possible par des raisons tirées de faits positivement attestés, car il n’existe pas de régles sémantiques permettant de dire que tel out el développement de sens est exclu” (A. Meillet, Linguistique historique et linguistique générale, Paris 1921, p. 31).
^ D. Geeraerts, St. Grondelaers, P. Bakema, The Structure of Lexical Variation. Meaning, Wording and Context, Berlin 1994.
^ A lexical meaning is called prototypical if it recalls the conceptual structure that best represents the mental space of the term. Prototypical meaning allows for more sufficient comprehension of marginal senses and offers a wider frame for reinterpretation.
^ It is with reason that Hans Krahe supports the inclusive treatment of semantic change: «Man muß ein solches Wort nämlich nicht isoliert betrachten, sondern muß es inmitten einer ganzen Gruppe von sachlich zugehörigen Wörtern sehen» (Einleitung in das vergleichende Sprachstudium, Innsbruck 1970, p. 87). Harry Meier adds that an etymological solution is not firmly established if the whole semantic field has not been taken into consideration (Prinzipien der etymologischen Forschung, Heidelberg 1986, p. 109).
^ The Italian linguist Carlo Tagliavini discussed emphatically the importance of Wörter und Sachen field and indicated that failure to scrutinise parallel histories could also misguide etymology: “Diese Richtung (i.e. Wörter und Sachen) setzt sich ein für das gleichzeitige Studium von Kultur- und Wortgeschichte und halt jegliche allein am sprachlichen Material ausgerichtete etymologische Forschung für sinnlos und gefährlich” (Einführung in die romanische Philologie, München 1973, p. 15).
^ According to reliable sources, the meaning "homosexual" arose from the expressive term gay cat / gey cat, which originated by the 1930s as part of slang used in the community of inmates and unemployed. It referred to a younger hobo (a wandering unemployed) or inmate, who was under the protection of an older one. The Miriam-Webster's Word Histories Lexicon (Springfield MA, 1989) offers full record of sources for this etymology and makes an important notice: “[Α hobo] often needed to attach himself to a veteran of the road in order to survive. It was generally taken for granted that the relationship between the youth [sic] and the older man eventually turned into a sexual liaison” (p. 190). If such relationships were at some degree commonplace in those communities, it is visible why a 1935 handbook of prison slang included the following definition: Geycat... a homosexual boy. The above mentioned reference seems to be the first attested instance of the pejorative meaning.
^ "PC or not PC ... that was 1993's burning question;Review of the Year 1993" The Sunday Times (London); Dec 26, 1993; Maurice Chittenden;
^ H. Hirt, Etymologie der neuhochdeutschen Sprache, München 1921, 2nd. ed.