English Course

Morphology

1.      MORPHEMES

In linguistic analysis, we have to collect data first, and then identify the minimal meaningful units of language. This minimal meaningful unit is called morphemes. From this minimal meaningful unit, we mean that the unit cannot be further subdivided and still have some meaning assigned to it. For example, we can divide the word “unladylike” into the three parts and assign some meaning to each part.

                        un-      =    not

                        lady    =   (well-behaved) female adult woman

                        -like    =   having the characteristic of

We cannot divide these into smaller meaningful pieces. In –un, we can’t assign a meaning to just vowel u or the consonant n. The meaning is carried by two sounds taken together. This meaningful combination of sounds constitutes a morpheme. And, we can’t assign meaning to “ady” but only the whole morpheme “lady”. 

By minimal, we do not mean smallest. If we take “smallest” literally, then the smallest units of English are limited to –s (plural), –d(past), a (indefinite article), and other morphemes that only consist of single segment. Size is not something important here, a unit does not need to be particularly small to be a morpheme, but it does need to be minimal in the sense of being unanalyzable.

By meaningful, we include either lexical or grammatical meaning. Lexical meaning is the type that we are used to looking up in a dictionary. It includes the meanings of such morphemes as banana, take, and yesterday. When analyzing language, we will recognize lexical meaning because they generally translate easily into some word or phrase in English.

Morphemes by grammatical meaning, do not usually translate directly from one language to another. When we see them, it is usually necessary to describe their meaning with technical linguistic terminology like ”plural”, “1^st person”, “perfective”, and “past”. These meaning are called grammatical because they have more to do with the grammar of a language than with its vocabulary. They usually can not be fully described unless you take into account the larger grammatical context in which they occur. Here, we try to see the morphemes from the language that unfamiliar with us. These are the sentences from Choapan Zapotec (Otomanguean, Mexico).

rao lipi za                     : Philemon eats beans

rao maka bela              : Macaria eats fish

re’en lipi bela               : Philemon wants fish

One way to start is by looking for items that recur in the data with a corresponding constant meaning somewhere in the glosses. This technique is known as finding recurring partials with corresponding constant meaning. In the data, we see that “rao” means several times and correspond with the English word “eats”, so rao probably means “eat”. While “re’en” recurs several times and correspond with the English “wants”. So re’en means “want”.

What we have done is to form a hypothesis about the meanings of each individual word, and check that hypothesis against all the data. It is important to check it out against further data to either confirm or disconfirm it. Morpheme identification becomes more difficult when a word consist of more than one morpheme, so that we need to identify not only the meaning of each piece, but also the boundaries between them. In this case, the best procedure is to use a different technique called finding contrast in a frame. For example, we can look up at these Choapan Zapotec data below :

raowa’             :  I eat                          waowa’           :  I will eat

raolo’               :  You eat                    waolo’             :  You will eat

raobi’               :  He/she eats               waobi’             :  He/she will eat

If we compared the four forms that starting with ra, we see that –wa apparently means “I” and “-lo” means “you”, -bi means “he/she”, and “ba” means “it (animal)”. By comparing “raowa” with “waowa”, “raolo” with “waolo”, etc, we see that –r and –w also contrast. A reasonable hypothesis as to their meaning that –r means “present tense” and –w means “future tense”. Then, we hypothesized that rao was a morpheme that means “eat”. Now,  we see that it containts two morphemes: r indicates present and ao indicates eat. While re’en consist of two morphemes: r that indicates present and e’en that indicates “want”.

In some languages there are morphemes that are noncommittal as to number (singular or plural) or gender (masculine, feminine, or neuter) and mean just “1^st person ( I and we), 2^nd person (you, either singular or plural), and 3^rd person (he, she, it, or they). Here are some examples from Nahuati (Aztec) of Mecayapan (Uto-Aztecan, Mexico, Wolgemuth 1981:47)

                        Nokal         :  my house

                        Mokal        :  your house

                        Ikal            :  his house

Just looking at these words, we can analyze no- as ‘my’, mo- as ‘your’ and i- as ‘his’. But this is not correct. For one thing, ical can also mean ‘her house’ or ‘its house’.

Look at these following forms :

                        Nokalmeh        :  our (excluding you) house

                        Tokal               :  our (including you) house

                        Amokal           :  your (plural) house

By looking at these forms, its clear that –meh means ‘plural’, but only when 2^nd person is not involved (in 3^rd person plural and 1^st person plural exclusive). When 1^st or 2^nd person are the possessor together, the prefix to- is used, and when more than one 2^nd person is possessor (without the speaker), amo- is used. Mo- presumably means just ‘your’ (singular). We see too that i- is not limited to ‘singular possessor’, but means simply ‘3^rd person possessor’. Similarly no- does not mean ‘my’ but rather ‘1^st person plural exclusive’, and that only in 1^st person singular and 1^st person plural exclusive. So, the best gloss is 1^st person exclusive possessor.

The meanings of this rather complicated set of morphemes can be characterized as follows:

            No-                  1^st person exclusive possessor

            Mo-                 2^nd person singular possessor

I-                                           3^rd person possessor

To-                   1^st person plural inclusive possessor

2.      STEM AND AFFIXES

Usually, a word structure is best described as consisting of a base, called either stem or root, and various additions or modifications to this base called affixes. There are two kinds of affixes, which are prefix and suffix. Prefix located at the front of the word while suffix located at the end of the word. There are three rules of thumb to help decide if a morpheme is a stem or an affix.

1.      The richness of their semantics

2.      Whether they belong to open or closed classes

3.      Whether they have bound or free

Stems usually have richer semantics than affixes. Stems usually have lexical meaning, while affixes often have grammatical meaning. Stems usually be glossed by a translation equivalent, while affixes often require technical linguistic terms. Stems are usually members of open classes, while affixes are almost always members of closed classes. ‘Open’ and ‘closed’ refer partly to the number members in a class, but more to the possibility of adding new members to the class. There are hundreds of adjectives in English, but probably less than fifty suffixes that attached to adjectives. New suffixes are much harder to invent, the class of adjectival suffixes is closed to new members.

Always bound, while stem may be either bound or free. When we say morpheme is free, we mean that it can occur by itself as a word. A morpheme that is bound cannot be a word by itself, but must always be attached to some other morphemes. So, of course, affixes must be bound. If an affix were free, that would mean there would be a word that was all affix and no stem. This would go counter to the basic distinction between affixes and stems: that a stem is the central, essential part of a word, and affixes are optional additions.