HOW TO SOLVE MORPHOLOGY PROBLEMS

HOW TO SOLVE MORPHOLOGY PROBLEMS

When a linguist comes in contact with a new language, one of his major tasks is to discover the meaningful units, or morphemes, out of which the language. Just as with discovering phonemes and allophones, it is important that the linguist have procedure for discovering these minimal units, since it is impossible to isolate morphemes by intuition.

For example, the Classical Greek word [grap^ho:] means ‘I write’, but the linguist has no way of knowing what sound or sequence of sounds corresponds to what English meaning if he considers the word in isolation. It is only by comparing [grap^ho: ] with another form, for instance, [grap^he: ] ‘the writes’ that he is able to determine what the morphemes of these Greek word are.

Comparisson, the is the best way to begin morphological analysis. But of course you will not want to compare just any forms. Comparing a Greek word like [p^hε:mi ] ‘to speak’ with [grap^ho:] will not provide us with much information since the forms are so similar and seem to have no single morphemes in common. What must be compared are partially similar forms in which it is possible to recognize recurring units. In this way we can identify the morphemes of which words are composed.

Now let’s consider our Classical Greek example once more. If we compare [grap^ho:] with [grap^he:] ‘he write’ we note similarities between the forms. The sequence [graph-] recurs in the forms [grap^he: ] and [grap^ho: ]. If we check our English correspondences we find the meaning ‘write’ recurs also. Here we are justified in assuming [graph-] means ‘write’, since [graph-] and write are constants in both. Further, since the final vowels in both forms contrast-and since this contrast is accompanied by a difference in meaning in our English correspondence-we can safely assume the difference between vowels in Classical Greek corresponds to differences I meaning in our English translation. Therefore we assign the meaning ’I’ to [-o: ] and ‘he’ to [-e:].

In sum, then, the initial step in doing morphological analysis is:

1.    Comparing and contrasting partially similar forms.

To give yourself practice, identify and translate the morphemes in the made-up data below, from a hypothetical language:

                   [ ĵapi ]                   house                    [ dru†a ]                tree

                   [ ĵapi† ]                 the house              [ dru†a† ]               the tree

                   [ aĵapi† ]               to the house          [ adru†a† ]    to the tree

Sometimes just comparing and contrasting partially similar forms is not to allow a complete morphological analysis. Consider the following examples:

1.    If we compare the following English words:

Work                broad

Worker             broader

We notice the morphems spelled –er and pronounced [r] for both [bradr] and [wrkr]. However, if we think about it for a minute, it is a apparent that –er has two different meanings even though phonetically it looks like the same morpheme. The –er in worker is the same –er that shows up in words like painter, kiler, lover, and actor. In each of  these case, -er attaches to verbs to form a noun, and means something like ‘one who paints’, one who kills’, one who loves’, etc. The  suffix –er in these case is known as the agentive morpheme.

The –er in broader, on the other hand, is the same –er that shows up in words like wider, longer, colder, prettier, and so on. In each of these case, -er attaches to adjectives to form a new adjective, with the extra meaning ‘more’. The suffix –er in these cases is known as the comparative morpheme.

We will want to argue, then that [r] represent two separate morphemes [r] as an agent marker, and [r] as a comparative marker –even though they are same phonetically, i.e. are homophonous morphemes. The [r] which is added to verbs to yield nouns and the [r] which is added to adjectives to yield new adjectives clearly have distinct meaning.

2.    IF we compare the following set of  words in (a), (b), and (c), we notice that each word has a prefix which means not.

(a)   ‘imbalace’      [imbælens]

(b)   ‘inability’                 [inəbiləti]

(c)   ‘incomplete’   [inkəmplit]

The problem here is the inverse of the problem in example (1). Whereas in examples (1) we had the same phonetic forms representing two different meanings, in examples (2) we have three differrent phonetic forms with the same meaning. Since the phonetic forms of the morpheme meaning ‘not’ here can be predicted on the bais of phonetic environment.

    [Im]    before labials-        [p], [b], [m]

    [Iŋ]     before velars-  [k], [g]

[In]     elsewhere (before vowels and other consonants)

We conclude that even though the forms differ phonetically they belong to the same morpheme, since they have the same meaning. We call [Im], [Iŋ], and [In] allomorphs of the same morpheme. Anoter example of allomorphy in English is the plural morpheme which is realized as either [s], [z], or [əz] depending on the form of the root to which it attaches.

PPROCEDURE FOR DOING MORPHOLOGICAL ANALYSIS

Goal :

Given a set data in phonetic representation, you are asked to perform a constituent morphological analysis of the forms.

Procedure (Keys to Analysis):

1.    Isolate and compare forms which are partially similar.

2.    If a single phonetic form has two distinctive meaning, it must be analyzed as representing two different morphemes (as in example (1) ).

3.    If the same meaning is associated with different phonetic forms all represent the same morpheme (they are allomorphs of the morpheme), and the choice of form in each case should be predictable on the basis of the phonetic environment (as in example (2) ).

Some Cautionary Notes

People frequently assume that language are pretty much the same in terms of what each language marks morphologically. For example, English speakers often assume that all languagemark the plural of nouns with an ending or that the subject and the verb agree in person and number in other language. This is simply not true. For example, Tagalog does not usually mark the plural of nouns (in most case, the number is clear from the context). When it is necessary to be specific, a separate word, mga, is used to indicate plural.

[an bata ]            ‘the child’

[an mga bata ]     ‘the children’

When a number is specifically mentioned, no plural marker appears in Tagalog, though the plural marker is obligatory in English (Three dog is ungrammatical):

          [da I awa]    ‘two’           [dalawaŋ bata]                ‘two children’

          [lima ]                   ‘five’           [imaŋ bata  ]                   ‘five children’

[ŋ] is a “linker” that links numerals and adjective to the nouns they modify, English does not use this type of device).

There is also subject –verb agreement in Tagalog for example in English / eat but he eats. In Tagalog, the same form of the verb would be used, no matter what the person/number of the subject: kumakain ako ‘eat now I = ‘I eat (now)’ kumakain siya ‘eat now he’ = he eats (now).

Other language also make distinctions that we don’t. while English has only singular versus plural, some language have a dual when just two are involved. Consider Sanskritjuhomi ‘I sacriface’ juhavas ‘we(two) sacriface’, and juhumas we (pl) sacriface.

Some languages also have two kinds of first person plural pronouns-that is, English we. Notice that English we in we are going, for example, may include everyone in the group the hearer is addressing (we, every one of us) or it may include some hearer(me and him, but not you). Many language distinguish these two we’s : Tagalog has tayo (inclusive, i.e you and I) but kami (exclusive, i.e. he and I)

Comanche makes a number of distinctions that English doesn’t. In additional to a singular / dual / plural distinction – in  you sg’, -nikw  you (two)’ m  you (pl)’ – and an inclusive / exclusive distinction –taa ‘we(incl.)’ nn we excl.’ – Comanche also make a distinction between visible / invisible and near/ far. Thus, if you are reffering to a thing that is within your view, yu use –ma it (visible)’. If the thing is invisible to you, it (invisible) is used. A near object is designated with – ‘I’ it (proximate)’, but a far object with –‘o’ it (remote).’ (Note : only the subject forms of these pronouns have been given).

The lessons to be learned here is that you cannot assume that another language will make distinctions in the same way that English does. For example, while every language has some method of indicating number, no all language do so in the same way or under the same way or under the same circumstances. As we’ve seen, English uses an affix, Tagalog uses a separate word, and Indonesian reduplicates the word to show plural. Nor can you assume that the distinctions English makes are the only ones worth making. Language must be examined carefully on the grounds of their own internal structures.