Defining Language Universals

Inhaltsverzeichnis

1 Introduction to language universals

2 Explanations for their existence

3 Different types of universal
3.1 Absolute and nonabsolute universals
3.2 Implicational and nonimplicational universals

4 Different approaches to universals
4.1 Greenberg
4.1.1 Concepts of Dominance and Harmony
4.2 Hawkins
4.2.1 Concepts of heaviness and mobility
4.3 Differences between Chomsky and Greenberg
5.3.1 The data base
5.3.2 Abstractness
5.3.3 Explanations for Universals

5 Conclusion

6 Bibliography

1 Introduction to language universals

Language is something uniquely human. Bertrand Russell stated in 1948 that “A dog cannot relate his autobiography; however eloquently he may bark, he cannot tell you that his parents were honest though poor” (Whaley 1997: 4). Language ist not only unique to humans and diverse but there are also commonalities between languages. About five thousand languages are spoken in the world today but there still is a basic unity that undelies their diversities. Many of the world’s languages show similar principles. An example of such a similaritiy is word order. In some languages such as English, French and Italian the word order is subject, verb and object. This is the so called SOV order.

Besides similarities like this, there also exist features between some languages, which are common to all human languages in the world and are called language universals. These language universals classify entire languages into categories which are then based on shared properties.

Language universals are examined within the field of typology. Typology has the task of examining cross-linguistic patterns. This means that “all typological research is based on comparisons between languages”

(Whaley 1997: 7). Through comparing different languages with each other, one necessarily comes into contact with universals, which hold true for a group of languages or even for all languages. This is how language universals are placed within the field of typology.

This paper starts with defining what language universals are and why they exist. In the main body it introduces different approaches to universals by different linguists, namely Greenberg, Hawkins and Chomsky. The approaches of Greenberg and Chomsky will be contrasted with each other. The comparison mainly focuses on the ideas of Bernard Comrie.

2 Explanations for their existence

In Comrie’s book Language universals and linguistic typology one can find three approaches to explaining universals. These are namely the monogenesis, the innateness including other psychological explanations and the functional and pragmatic explanations.

The term monogenesis means that there is a genetic origin which holds true for all languages of the world. The hypothesis is that there once existed and ancestor language from which all the languages of the world developed. This would also implicate that each of the universals have been part of the ancestor language and that they have passed on through all the stages the language went through. This would mean that every universal has to be present in all of today’s languages. But what happens with universals that have a precondition, like implicational universals do? Implicational universals will be explained in detail in the next chapter under 3.2. Another question arises from this hypothesis: How are the different groups of languages explained which were not in contact with each other all through the development of languages? Besides that, there are fundamental differences between languages of today. How does this hypothesis account for these? There is no proof or disproof of this thesis since there is no possibility to recover this one ancestor language. From this we already can conclude that such an explanation cannot be viable.

The next approach to consider is the one of the innateness and other psychological explanations. Innateness was favoured by Chomsky. He referred to it as the Universal Grammar, which enables the child to learn a language in such a short period as 4 or 5 years. But it is not a satisfying explanation for the universals since it is just a name and it cannot be proven. Comrie states that:

Advocates of innateness are simply arguing that in the absence of any alternative coherent explanation for language universals, innateness is the only possibility they can think of. Instead of serving to deepen our understanding of language universals, the absence of any possibility of testing innateness as an explanation serves rather to divert researchers from considering alternatives that my be testable. (Comrie 1981: 24)

The last approach is the one of the functional and pragmatic explanations. The argument concerning this approach is that through certain universals language becomes more functional. This contradicts the numerous instances where language seems to be dysfunctional. “The existence of synonyms seems to be a needless luxury, and even more clearly the existence of homonyms…” (Comrie 1981: 25). Still, universals may have something to do with making a language more functional since for example a functional language with only homophons is hard to imagine. Some strategies to reduce such dysfunctional elements must play a role and these strategies might lead to explanations for language universals. The functional point of view is that a language universal makes it easier to recover the semantic content from the syntactic structure.

So far it has been discussed why language universals exist and three approaches to their explanation, given by Comrie, were laid out. The next chapter will define the different types of universals. The categorization of language universals will go along with the one proposed by Lindsay Whaley except for the term nonabsolute universals which is replaced by the term tendency as Comrie calls these universals.

3 Different types of universal

3.1 Absolute and nonabsolute universals

Absolute universals are statements which hold true for all languages at any time. The term universal already suggests that something has to hold true everywhere and any time. Croft refers to these universals as unrestricted universals and states that they’re number is relatively small. Hawkins calls these universals nonstatistical. In this paper it will always be refferred to the term absolute universal for the cause of simplicity. Examples of absolute universals are the following:

a) All languages have consonants and vowels.
b) All languages are able to form questions.

Absolute universals require deeper explanations, for example why all languages are able to form questions. These universals are absolute because they have no counter arguments in any of the world’s languages. This also includes all the languages for which there is no description and all those which have become extinct. It also has to include all the languages which will come into existence in the future because language is constantly changing. This can be said because by linguists all over the world:

it is assumed that the rules that govern language structure today are the same that governed language structure yesterday and will be the same that will govern language structure tomorrow.

(Croft 1990: 44)

If anyone will ever discover one language for which the supposed universal does not hold true, then it will not be an absolute universal anymore but instead a nonabsolute universal.

A nonabsolute universal is one which admits exceptions. These universals are not essential to all languages but they represent tendencies and they usually hold true. Hawkins refers to them as statistical universals. Comrie calls them tendencies and in this paper they will be referred to as such since the term nonabsolute is somehow contradictory in itself. Lindsay Whaley uses this contradictory term in her book because it is commonly used in typological research. An example for a tendency is the following: „In a basic word order, the subject precedes the object” (Comrie 1981: 19). There are probably less than 1 per cent of the world’s languages which violate this universal. Two examples of such languages are Malagasy, an Austronesion language with VOS word order and Hixkaryana, a Carribean language with OSV basic word order (compare Comrie 1981: 19).

Hiskaryana (OVS): toto yahosIye kamara
man it-grabbed-him jaguar

The jaguar grabbed the man.

Malagasy (VOS): Manasa lamba amin’ny savony ny lehilahy

Washes clothes with.the soap the man

The man washes clothes with the soap.

(Whaley 1997: 82)

3.2 Implicational and nonimplicational universals

Non-implicational universals are for example the absolute universals which were discussed in the previous chapter. The statement that all languages have consonants and vowels makes no reference to any second condition that has to be fulfilled for the universal to be true. The tendencies were also non-implicational. The term implicational means that the universal has a precondition that needs to be fulfilled. It has the form of {if p then q}. An example of an implicational universal is:

Universal 3: Languages with dominant VSO order are always prepositional.

Universal 4: With overwhelmingly greater than chance frequency, languages with

normal SOV order are postpositional.

(Greenberg 1966: 78)^[1]

Implicational universals have two properties. They can either be absolute as in Universal 3 or nonabsolute as in Universal 4. Besides that the universals cannot be switched around. For example, universal 3 cannot be reversed to the statement that if a language is prepositional, it also has VSO^[2]. English for example is an example of a prepositional language which is not VSO but SVO^[3]. This means that the universals are unidirectional.

[...]

^[1] Prepositional languages are languages which rather have prepositions than postpositions. An example of a prepositional language is English. The only postposition in English is ‘ago’. An example of a postpositional language is German. Examples of both are:

English: According to the letter he will come tomorrow.

German: Dem Brief zufolge wird er morgen kommen.

^[2] VSO = Verb Subject Object

^[3] SVO = Subject Verb Object