This is the first of a sequence of lectures discussing various levels of linguistic analysis.
We'll start with morphology, which deals with morphemes (the minimal units of linguistic form), and how they make up words.
We'll then discuss phonology, which deals with phonemes (the meaningless elements that "spell out" the sound of morphemes), and phonetics, which studies the way language is embodied in the activity of speaking, the resulting physical sounds, and the process of speech perception..
Then we'll look at syntax, which deals with the way that words are combined into phrases and sentences. Finally, we'll take up two aspects of meaning, namely semantics, which deals with how sentences are connected with things in the world outside of language, and pragmatics, which deals with how people use all the levels of language to communicate.
The basic concepts in any field are often the most difficult to define, and the concept of word is no exception. Crystal enumerates five tests, mentioning that the first-listed ones are the most reliable, and the most universally applicable.
What Crystal calls "Phonetic boundary" (#4) is only applicable in lucky cases. For instance, in German, certain word-final consonants are devoiced, so that Hund "dog" is pronounced with a final [t]. However, even in German, this doesn't help diagnose forms that end with a vowel or with a consonant that doesn't show the effect. Most languages lack such word-boundary sound changes entirely. The problems with Crystal's criterion #5, "Semantic units" can be seen by trying to figure out the contribution of on to the sentence Crystal cites, I switched on the light.
Crystal therefore defines a word as "the smallest unit of grammar that can stand alone as a complete utterance, separated by spaces in written language and potentially by pauses in speech." The spaces criterion is not very general, applying only to languages with writing conventions like those of modern European languages. It does not apply at all to writing systems like those of Japanese and Chinese, in which spaces are not used to separate words; and its application is affected by changes in writing conventions, as happened recently with changes in the official rules about writing German compound words.
Notice that this definition does not say anything about meaning. This is appropriate, because words are not the most basic units of linguistic meaning. There is a more basic unit, the morpheme, defined as follows by the Ohio State Language Files (a widely used linguistic workbook): "the smallest linguistic unit that has a meaning or grammatical function."
These definitions suggest many questions. Two important ones are:
The answer to (1) is "yes". Although not all languages have exactly the same categories of words, they all have words, and almost all share the major categories words come in. What traditional grammar calls the "parts of speech", linguists call either word classes (Crystal) or lexical categories (Language Files). Crystal gives the following list of eight, for English:
nounLanguages may differ in some details. Some languages don't really distinguish between verbs and adjectives, for example, and languages with a basic Subject-Object-Verb (SOV) word order tend not to have prepositions. Instead, they usually have "postpositions", which come to the right of the nouns an English speaker would think of them as "introducing". We will learn later that such verb-final languages are better thought of as "head-final, " so that the final position of the verb in the sentence, and the final position of the postposition relative to its noun, are two aspects of the same more general property of the language. Here is an example of a postposition (-se) in Hindi:
Ram cari-se kutte-ko maraAll languages also have morphemes, although the classes they fall into are not as similar across languages as the categories of words are.
Ram stick-with dog hit
"Ram hit the dog with an stick."
MORPHEMES, as minimal units of meaning or of grammatical function, may differ from each other in many respects. Here is an initial list of examples from English:
Bound Morphemes: cannot occur on their own, e.g. de- in detoxify, -tion in creation, -s in dogs, cran- in cranberry.
Free Morphemes: can occur as separate words, e.g. car, yes.
In a morphologically complex word -- a word composed of several morphemes -- one constituent may be considered as the basic one, the core of the form, with the others treated as being added on. The basic or core morpheme in such cases is referred to as the stem, root, or base, while the add-ons are affixes. Affixes that precede the stem are of course prefixes, while those that follow the stem are suffixes. Thus in rearranged, re- is a prefix, arrange is a stem, and -d is a suffix. Morphemes can also be infixes, which are inserted within another form. English doesn't really have any infixes, except perhaps for certain expletives in expressions like un-effing-believable or Kalama-effing-zoo.
Now we are ready to answer the second question, dealing with the relationship between words and morphemes. The best way to explain this relationship is to break the question down into a series of simpler questions. Along the way, we will also consider the relationship between morphemes and syllables.
Can a word = a morpheme?
Yes, all of the following words also consist of one and only one morpheme:
|Word (=Morpheme)||Word Class|
Are there morphemes that are not words?
Yes, none of the following morphemes is a word:
|kempt (?)||bound morpheme|
|shevel (?)||bound morpheme|
(Our discussion of kempt and shevel will be followed up in the lecture notes Morphology II.)
Can a word = a syllable?
Yes, all of the following words also consist of one and only one syllable:
Are there morphemes that are not syllables?
Yes, some of the following morphemes consist of more than one syllable; some of them are less than a syllable:
|under||preposition (> syll.)|
|spider||noun (> syll.)|
|-s||'plural' (< syll.)|
Lastly, are there syllables that are not morphemes?
Yes, many syllables are "less" than morphemes. Just because you can break a word into two or more syllables does not mean it must consist of more than one morpheme!
|kayak||(ka.yak)||neither ka nor yak is a morpheme|
|broccoli||(bro.ko.li) or (brok.li)||neither bro nor brok nor ko nor li is a morpheme|
|angle||(ang.gle)||neither ang nor gle is a morpheme|
|jungle||(jung.gle)||neither jung nor gle is a morpheme|
We must conclude that there is no necessary relationship between syllables, morphemes, and words. Each is an independent unit of structure.