Languages can be classified into groups based on a number of different linguistic criteria. One such way to categorize languages is by the type and extent of morphology that they use. For example, some languages string many morphemes together while others languages tend to realize most words as independent or mono-morphemic segments. The latter type of language is called analytic and the former is called synthetic. This typology should be seen not as a strict dichotomy between analytic and synthetic, but rather as a scale on which languages can be placed depending on the degree to which they exhibit that type of morphology. Some languages may display different types of morphology in different areas of their grammars and thus it is best to see languages as tending towards a type, rather than strictly adhering to a single one.
Analytic languages are also called isolating languages because each word tends to consist of a single, independent morpheme. Such languages do not make use of affixes, and grammatical markers, for features like tense and case, are generally realized as unattached morphemes. Analytical languages are most common in Southeast Asia (Chinese, Vietnamese), but some such languages are also found among the Austronesian languages (Fijian, Tongan) and some Niger-Congo languages (Gbe, Yoruba). Vietnamese provides an example of this type if language:
1. no se khong doc sach
he FUT NEG read book
'he will not read book'
2. tô: đên nha ban tô:
I come house friend I
‘I come to my friend’s house’
In the first example, person and tense are not marked as affixes on the verb, but stand as separate morphemes preceding it. In the second example, the possessive is not marked on the noun as in English (‘friend’s’), and the first person pronoun does not have a different form in a subject position ( nominative case) than it does a possessor position (genitive case). Because of this, strict adherence to a certain word order is necessary, and is in fact a common feature of analytical languages. If case and other grammatical functions are not marked on words, than only order can assure the correct interpretation. In (2) above, the order of ‘house’ and ‘friend’ determined which possessed the other; had the order been reversed, the resulting structure would have referred to the (semantically implausible) house’s friend. Another example comes from Fijian, in which strict adherence to a VOS word order allows for the correct interpretation:
Fijian: ea taya na ŋone na yalewa
PAST hit the child the girl
‘the girl hit the child’
However, some isolating languages do mark subjects and objects though they do so through the use of separate morphemes rather than affixes. Tongan employs such markers before the noun (bolded in the example):
Tongan: na?e taa?i ?e hina ?a vaka
PAST hit SUB Hina OBJ Vaka
‘Hina hit Vaka’
Therefore, it can be seen that while isolating languages make little use of some morphological processes, they are nonetheless quite capable of showing grammatical relationships.
Synthetic languages differ from analytic ones in that they string several morphemes together into multi-morphemic words. Languages of this type make abundant use of such morphological processes as affixation and compounding. There are several degrees of synthesis seen in languages, from those that combine only a few morphemes to those in which an entire sentence can be expressed by a multi-morphemic word. Languages of the latter type are called polysynthetic languages and are sometimes seen as comprising their own separate morphological class. Many Native American, Siberian, Caucasian, and northern Australian languages are classified as polysynthetic, including Inuktitut, Mohawk, Central Siberian Yupik, Cherokee, Chukchi, and Tiwi. The examples below show that the comparable English translation requires several separate words:
‘I made a dress.’
‘I kept on eating’
The category of synthetic languages can be further subdivided into two types: agglutinating and fusional. In agglutinating languages, each morpheme expresses a single meaning, and several can be strung together. Such a languages might mark tense with one morpheme, person with another, and aspect with a third, leading to a build-up of such morphemes on the stem. Examples of this type of language include Turkish (Altaic), Hungarian and Finnish (Uralic), Swahili (Bantu) and Japanese and Korean. The following Turkish example shows different markers for number, possession, and location.
‘in my houses’
An example from Swahili shows agglutination in verbal morphology. Markers for person, tense, and object are realized as separate morphemes on the stem:
‘he liked me’
Agglutinating languages can produce some very long words from the concatenation of morphemes, each with its own function.
In contrast to agglutinating languages, in fusional or inflectional languages single morphemes simultaneously combine or fuse several meanings in one form. Indo-European languages are familiar examples of this type. For instance, in the verbal morphology of many such languages, tense, person, and number are realized as a single affix on the verb, as in the following example:
Spanish: habl—o habl—as habl—a
speak-1.sg.PRES speak-2.sg.PRES speak-3sg.PRES
‘I speak’ ‘you speak’ ‘he/she speaks’
Fusional morphology can also be seen in case markings, as in the example below from Russian. Here, the affixes on knig, ‘book,’ indicate both case and number in a single, fused morpheme (in some instances, the same marker is used for multiple forms in the paradigm):
Russian: Case singular plural
Nominative knig-a knig-i
Genitive knig-i knig
Dative knig-e knig-am
Accusative knig-u knig-i
Instrumental knig-oj knig-ami
Prepositional knig-e knig-ax
Despite the differences between them, both agglutinating and fusional languages, as synthetic languages, share another characteristic in common: a freer word order. This is due to the fact that such languages often mark features such as case and person directly on words and so are not dependent on word order to correctly interpret a sentence; the morphology provides the necessary information. Consider the following Latin example, which carries the same meaning regardless of word order:
Latin: 1) agricul-a puell-ās vīdit
farmer-NOM girl-ACC saw
‘the farmer saw the girl’
2) puell-ās agricul-a vīdit
3) vīdit puell-ās agricul-a
The study of morphological types can shed light on both the diversity and similarity of languages. By examining the different types, we can recognize the ways in which languages contrast between categories, and compare within them.