Binary Classification in Synchronic Morphology

{rat} {or} {gen}

lexical syntactic lexical

{ed} {tion}

To begin such a synchronic investigation, we must first discover in what ways morphemes can be synchronically classified. Most useful may be three binary terms, three sets of two words nearly opposite in meaning. One term from each set can be applied to each morpheme as we encounter it (figure 5.1). Each morpheme should be describable by one, but not the other, of the words in each binary set. We have already mentioned the first pair of classifiers: lexical-syntactic. Nearly all words in the lexicon can be identified as either lexical or syntactic morphemes.

Free-bound. The second such binary set is free-bound. Morphemes can be classified as either free or bound, that is, either capable of standing by themselves, as words, or not capable of standing by themselves. A free morpheme can be either lexical, like the word {rat}, or syntactic, like the operator {or}. If the free morpheme is syntactic rather than lexical, other morphemes will not combine with it. Lexical morphemes, however, can combine with other morphemes or other words. For example, to the free lexical morpheme {rat}, we can add the bound allomorph {s}, producing the word “rats”, or the bound morpheme {ty}, producing “ratty”, or another free morpheme {catch}, plus a bound one, {er}, producing “rat-catcher”.

Base-affix. The third useful binary set for synchronic analysis of morphemes is base-affix. A morpheme may be classified as a base or as an affix but not as both – at least, not in the same word. (Like the other terms in the two previous binary sets, “base” and “affix” are relational classifications rather than pure opposites. It is possible for the same morpheme to be free in one use and bound in another, or lexical in one case and syntactic in another.) A base – sometimes called a root – is a morpheme to which other morphemes are attached. The attaching morphemes are called the affixes.

Bases are morphemes that lexically, rather than syntactically, dominate other morphemes. Bases always have more lexical than syntactic meaning but bases are not always free. There are many bound bases in English as well as free bases. An example of a bound base is {gen}, a morpheme that has more lexical than syntactical meaning (gen = beginning), so it is a base rather than an affix; but it cannot stand by itself as a word, so it is bound, not free. Put together with various other attaching morphemes, {gen} gives us words like “generic”, “gene”, and “engender”.

Affixes. The affixes, or attaching morphemes, can be further subdivided by two subsidiary binary sets. The first set, prefix-suffix, merely indicates where the attachment occurs. If the affix is attached at the beginning of a base, as {un} is, then the technical term used is prefix. Prefixes alter the lexical meaning of a morpheme, but do not change its syntactic function. The prefix {un} and its allomorphs {im} and {in}, for example, turn an affirmative base into a negative one: {un} + {well} means “not well”; {im} + {possible} means “not possible”; {in} + {secure} means “not secure”.

If the attachment appears at the end of a base, as does {ed}, then the name for it is suffix. It is in connection with suffixes that the second binary subdivision should be applied, for there are two kinds of suffixes: inflections and derivations. As we indicated earlier, inflections alter the syntactic function of a base but do not change the lexical meaning very much. The class of inflections is very small (plural, tense, possession, comparison) and very stable; no new forms are being added. Derivational morphemes, however having more lexical meaning than do the inflections, belong to a larger and a more open class. New derivations can easily be added. Like inflections derivations can change a base’s syntactic function, but they can also alter the base’s lexical meaning somewhat, in a way that the inflections do not. For example, take the base {friend}. This is a noun of the type called [±count], that is, having the feature “capable of being counted individually”. Such a noun can be made plural by the addition of the plural allomorph {s}: we can speak of one “friend” or of many “friends”, and we can count the number of friends if we wish, one by one. The basic lexical meaning of {friend} has not been much altered by the addition of an inflection. But if to the singular base, we add the derivation {ship}, to produce the word “friendship”, we have a different sort of change. The syntactic meaning has not been altered, for the word is still a noun, but now the lexical feature shifts from [+count] to [–count]; “friendship” is ordinarily a mass noun, one that refers to a quality or a concept rather than to a number. Ordinarily, such nouns cannot function as plurals: we say “friendship is” rather than “friendships are”.

As another example, if we add the derivational suffix {tion} to the base {derive}, the suffix will change the function of the word from verb to noun. If we then add another suffix, {al}, to the newly produced noun “derivation”, we get a second a second category change: the word “derivational” is an adjective. Add one more, {ly}, and there is a third change in function, to adverb. Derivational suffixes, like {ship}, {tion}, {al}, and {ly}, are added to a base before the inflectional suffixes: inflections always tag along last.

Now let us try a little practice with synchronic analysis. Refer to the binary chart (figure 5.1) and keep in mind that the principle of minimal-pairing applies in morphology as well as in phonology: investigating types of morphemes, we should use morphemes that are as syntactically similar as possible – preferably identical – but are different in lexical meaning. Let us use the word “geography” as a test case, to discover how many morphemes it contains and of what sort they are.

Let us substitute {bio} (life) for {geo} (earth); we get a recognizable word,

biography”. So {geo} is a morpheme. Substitute {logy} (study of) for {graphy} (writing about), and the result is another familiar word, “geology”. Finally, change {y} (activity or product)[3] to {ic} (characteristic of), and we have a new word, “geographic”. Since we cannot change any other part of “geography” and still get a standard English word – that is, we cannot subdivide the meaningful forms of “geography” any more than we already have – these must be the only three morphemes in the word: {geo}, {graph}, {y}. One of these morphemes never appears by itself as a separate word; therefore, it is not a free morpheme, but a bound one. Is it a bound base or an affix? It has very little lexical meaning, serving mainly to direct traffic, so it must be an affix. Since it appears at the end, it is specifically a suffix. Since it has some lexical meaning of its own, as well as indicating the function – noun – of the word “geography”, {y} is a derivational rather than an inflectional suffix.

{Geo} is a bound base. It cannot stand by itself lexically as a word and therefore is not free; in other words, it is bound. But morphemes with even less lexical meaning do attach to {geo}, so it is a base. In fact, other morphemes must attach to {geo} in order for there to be anything like a word formed from it, so it has to be a bound base.

Finally, there is {graph}, the most independent morpheme in “geography”. It is, we discover, a free base: free, because it can stand by itself as a word; base, because it can accept other, attaching morphemes. Like other free bases, {graph} is unlimited positionally. It can appear at the beginning of a word (“graphology”), in the middle (“geography”), or at the end (“photograph”).

Around such bases as {graph} is our English lexicon built. One of the reasons for the flexibility of English is that it borrows bases from just about every language there is; {graph} comes from Greek. Another reason for this flexibility is that English constantly uses bases and affixes to produce precise distinctions of lexical meaning. This manipulatability of parts allows for marvelous adaptability to the shifting demands of linguistic contexts, the changes that happen to all languages in response to historical, social, or other pressures. In the next section, we will discuss some specific ways in which morphemes can be manipulated to form different kinds of words.

Diachronic morphology. If we alter our investigative stance from a word’s present morphological construction to its historical development, we move from a synchronic to a diachronic analysis. (It is perfectly possible to do both, but we are here keeping them separate.) Diachronically, we would ask where a given word came from and how has it changed its meanings as it has changed its forms.

Not all words do change their lexical meanings over time. For example “stone” currently, in standard usage, has pretty much the same referent as it did when it was the Old English stan. Most function words, such as prepositions (words like “with”, “out”, “of”, “under”, or “over”), have remained pretty much unaltered in meaning and in form since they entered the language. But enough words do change to make reading older literature difficult, as you know if you have tried Shakespeare or Beowulf. The constancy of change will also make a prowl through the etymologically oriented many-volumed Oxford English Dictionary entertainingly worthwhile; check the O.E.D., for instance, on what has happened to the word “nice”.

Occasionally writers will make deliberate use of a change in a word’s meaning, playing off an older meaning against a contemporary one, like a historical pun. Blake does that with “appalls” in these lines from the poem “London”: “How the Chimney-sweeper's cry / Every blackning Church appalls…”. On one level, the word has the contemporary meaning of “horrifies”, but on another level, it echoes an archaic meaning, “turns pale”. As a contrast with “blackning” (Blake’s spelling), “appalls” suggests the terrible plight of the chimney sweep in an industrialized city, so terrible that even the sooty buildings, and the social institutions symbolized by the buildings (the Church), grow pale with alarm.

We can discover general patterns of lexical change in words, trends that the English lexicon as a whole tends to follow over time. Broadly speaking, there are two ways of classifying the trends, as processes and as directions. If we are concerned with etymological processes, we are investigating morphological changes that can occur at any time to alter a word’s lexical meaning. Directions, on the other hand, reflect the incremental drifts from one stage to the next stage of specifically historical change in meaning.

Processes

Different etymologists will classify the processes and directions in different ways, but this is the one used here for processes:

1) analogy 3) reduplication 5) back-formation 7) shortening

2) compounding 4) derivation 6) base-creation

Analogy. Of these processes, analogy may be the most significant; some linguists have argued that all the processes involve some sort of analogy. (Other linguists disagree.) In any case, analogy in effect matches an already existing pattern with the demands of a new context. The new context produces a new word or a new usage of an old word. For example, “defense” /difεns/ used to be a noun only, but now is commonly also used as a verb /difεns/ by sports commentators and fans, as in “to defense against the Cowboys’ front four”. Everybody uses analogy to some extent, but children in particular are prone to doing so during the overgeneralizing stages. “I toed to town” is a common analogy with the regular {ed} past-tense inflectional affix. My daughter, Joni, was once told to “Behave!” (phonetically: [bi’heiv]). Her answer was, “I’m being have!” (phonetically: [‘biiŋ heiv]), an analogy with linking-verb-predicate-adjective constructions such as: “Be good!” – “I’m being good!”

Many standard constructions and words have come into Modern English by way of analogy. There was, for example, an Old English verb, drincan /drinkon/, “to drink”, and a noun, drenĉ /drεnĉ/, “a drink”. The verb drincan produced a converted form, the noun drinc, which has come down to us as “drink”, while the old drenĉ disappeared as a noun. But drenĉ also produced a conversion, the verb drenĉan, which is now our word “drench”. The conversions from verb to noun to verb were made by analogy of the respective noun-verb derivational suffixes. So Modern English has one verb, “drench”, and one noun, “drink”, as well as the original verb (“drink”).

The other processes involve a systematic jugglingof morphemes:

Free + Free: Compounding, Reduplication

Free + Bound: Derivation

Free – Bound: Back-formation, Shortening

+ Free: Base-creation

Compounding and reduplication. Compounding, discussed earlier when we were examining free bases, simply combines two or more bases into one new word: {road} + {block} = “roadblock”; {stop} + {light} = “stoplight”; {over} + {see} = “oversee”. Note that the resulting compound word means something different from the sum of its parts; it is a new word. Reduplication, the repetition of phonologically similar free morphemes, almost always produces a comic effect because of the near rhyme in key phonemes: “mish-mash”, “helter-skelter”, “zigzag”. Consequently, reduplicated words appear more often in colloquial or slang use than in formal utterance.

Derivation and back-formation. Derivation, possibly the most common etymological process after analogy, and back-formation, which is far less common, are in effect opposites of each other. As we mentioned in the section on bases and affixes, derivation adds affixes to a base, as in {com} (affix) + {pound} (base) + {ing} (affix), “compounding”. Where that process derives words by the addition of affixes, back-formation deprives a base of its apparent affixes, producing a new word in the form of an unadorned base. The verb “edit”, for instance, comes from the Latin noun editor, one who gives out, that is, one who publishes and distributes copies of something. The suffix {or} on the bound base {edit} dropped off, resulting in a back-formed free base. “Sidle”, meaning a sort of crablike shuffle, is backformed from the Middle English adverb sideling, beside or alongside of “Nestle”, to cuddle close to, comes from an Old English noun, nestling, a creature still young enough to be confined to its nest. In all three instancesthe bases – {edit}, {sid<e>l}, and {nestl} – came intoexistence after a longer word had already been in use.

Shortening. Shortening (sometimes called abbreviation) is like the process called synecdoche in poetry: a part is substituted for a whole. In poetry, it is common to find such substitutions as “the crown” for “the king” or “the monarchy”; in diachronic morphology, we find “cab” substituting for “cabriolet”, “extra” for “extraordinary”, or “varsity” for “university”. In varsity”, the graphic shift form <e> to<a> reflects the common British pronunciation of short<e> as [a], where Americans would say [ε]. <…>

Base-creation. As its name suggests, a base-creation makes up bases, using morphemic relations as guidelines. A base-created word often echoes natural sounds onomatopoetically, as in “hiss”, “mumble”, “hum”, or “drizzle”. There is, of course, no such thing as a completely new word, since we are limited generally by the sounds the human speech organs can produce and particularly, by habit as much as anything else, to the sounds with which we are familiar. Thus, while the word “drizzle” (as either noun or verb) is a base-creation, it is also phonologically and etymologically related to the Middle English noun mizzle, a fine mist. Even the brand name Kodak, often cited as an example of “pure” base-creation, did not come out of thin air. Whether its creators intended it or not, “Kodak” follows a familiar English phonological pattern. It consists of two syllables with the accent on the first syllable and an alternation of consonant and vowel sounds. Compare [’kowdæk] with these transcriptions of common nouns that follow the same phonetic pattern of first-syllable stress and consonantal-vocalic alternation: [’nowtis], [’rivεr], and [’tuwtər].[4]

Directions

As with processes, the directions of etymological change will be classified in different ways by different people; we will use this scheme (the classifications here are not mutually exclusive):

1. deterioration 4. concretization 7. radiation

2. elevation 5. extension

3. specialization 6. metaphorization

We can divide these seven into three groups according to the direction of change each manifests. Deterioration and elevation are opposites of each other; specialization and concretization develop in similar ways, the approximate opposite of those followed by extension and metaphorization; and radiation’s directions include nearly all of those followed by the others. We might suggest their relationship by diagrams like those in figure 5.2.

Deterioration and elevation. When a word deteriorates (in etymology, the term has no bad connotations, such as of decay), it changes from a relatively exalted or at least neutral significance in its first recorded usage to a relatively condemnatory or trivial meaning. “Silly”, for instance, originally meant blessed, not its current foolish; “knave” referred to a youth or boy, not to a rascal. The opposite happens when a word’s semantic meaning is elevated from something neutral or deprecatory to something more suggestive of approval. “Knight” originally referred to any young man and carried none of its present associations with romantic gallantry and glory. “Fond” first meant foolish, daft, or crazy, not affectionate (although it is easy enough to see how the meaning changed as it did). A “surgeon” began as a barber who also drew blood, applied leeches, and pulled teeth; his sign, incidentally, was a bloody rag, which survives in the red-and-white barber’s pole.

Specialization and concretization. In specialization and its analog, concretization, the motion of change is from the general or metaphoric to the specific or concrete. The opposite path is followed by extension and metaphorization, which change from the specific or concrete to the general or metaphoric. A word is said to have become specialized if its application has become narrower over time. “Meat” has come to refer particularly to animal flesh rather than to food generally, and “starve” now means death through failure to ingest food instead of death by any means. But a concretized word has moved from an abstract reference to a moreconcrete one. “Multitude” now means a crowd, a collection of tangible, individual, countable bodies. It has the feature [+ count] now. Originally, however, it was a mass [ – count] noun meaning “many-ness”. Similarly, the word “youth”, which originally referred only to the abstract, [ – count] concept “youngness”, now also has acquired a [+count] reference to young people, as in “the youth of this nation are its best chance for the future”.

Extension and metaphorization. Extension reverses specialization: an extended word now has a wider application than it did at first. For example, “bird” currently is a generic reference to any member of the class Aves, rather than its original, specific reference to a young, nesting bird.[5] Similarly, metaphorization reverses concretization: what once had only a specific, concrete reference can now be applied in a metaphoric, nonliteral fashion. “Bright” no longer applies only to a quality of light; we can intelligibly speak of a bright boy. “Sharp” can be applied to tongues or winds, not just to blades; a smile is “cold”, not only the temperature. Many words that technically are the products of metaphorization have been in use so long that they have lost their flash of metaphor: “blanket of snow”, “to labor under an illusion”, “casting aspersions”, “the grasp of a subject”. These are known as dead metaphors.

Radiation. Finally, although radiation is probably most closely related to extension, it actually combines the directions of the last four types. When a word’s meanings undergo radiation, they spread or branch out into near-metaphor, in the sense that the meanings are applied in unusual contexts. But unlike extension and metaphorization, the meanings are not derived chronologically from one another, in a sequential alteration of meaning. Rather, the present meanings of a radiated word coexist with the original ones and all meanings past and present, still refer to essentially the same idea. For instance, all the current applications of the word “paper” – from those referring to an essay or journal to the references to a governmental policy statement – still carry some sense of the original, literal reference to a papyruslike material on which to write. Similarly, “head” still conveys the notion of that part of the body containing the brain, even in such diverse uses as “head of state”, “head of the bed”, or, in the slang of some drug users “head” meaning one who uses drugs extensively, as in “He’s a real dope head”.

Figure 5.2

Date: 2016-04-22; view: 826