Compounds

Compounding – one of the most prominent word creation processes – involves the creation of compound words by ‘combining’ two or more free morphemes which can also be realised as independent words. The compound, however, has an independent identity as single words. The use of compounds is prolific in English and it is virtually impossible to list all of them. However, recognising that any two random words may not produce a compound word, attempts to define the rules and processes which govern the production of compounds have been made via three approaches: the syntactic approach, the semantic approach and the generative approach.
The Syntactic Approach
The syntactic approach describes compound words in terms of the grammatical categories to which their constituent words belong.
Two-word compounds can be divided into those with a noun, adjective or verb as the second word (which cannot be anything else except in as few exceptional cases as in seen in words which end with ‘up’ or ‘out’). Usually, the compound belongs to the same category as the word in the second position as can be seen in ‘darkroom’, ‘waterproof’ and ‘ballpen’. These compounds can further be divided into noun compounds (comprising two nouns, an adjective and a noun, or a verb and a noun), adjective compounds (comprising two adjectives, or a verb and an adjective) or verb compounds (comprising two verbs, or two adjectives, or a noun and a verb).
However, not all combinations of verbs, adjectives and nouns actually work as compounds, and some kinds of combinations like two nouns (as seen in school-girl) are far more productive than others such as those compounds which comprise two verbs (as seen in sleepwalk). In all these cases, the word on the right is called the ‘head’ and determines the category to which the compound itself belongs.
There are, however, also compounds which involve adverbs etc. such as ‘income’, ‘overdose’ and ‘overflow’ which do not have a head and whose grammatical category cannot be determined by the categories to which their constituent words belong. Thus, the notion of a syntactical head of a compound does not always help in the classification of compound words.
The Semantic Approach
This approach describes compounds in terms of the relationships between the meanings of the words combined. It says that the meaning of the compound is always greater than merely the sum of the meanings of its constituent words. Nonetheless, the meanings of the constituent words usually indicate the meaning of the compounds as can be seen in ‘blackboard’ and ‘girlfriend’. However, there are exceptions to this rule. For example, there is nothing in the constituents of ‘tallboy’ to indicate that it is used to store clothes it is therefore exocentric.
Further, a compound is distinguished from a phrase by semantic distinction reinforced by phonetic distinction. In a compound, the primary stress is carried by the first constituent. Thus, ‘black’ in ‘blackboard’ is stressed while precisely the opposite happens in a phrase: ‘board’ would be stressed if a board which had been painted black were to be spoken of.
The unpredictability of the additional meaning element also arises from the arbitrary nature of the relationship compounds assume between their constituent elements. The existence of exocentric compounds such as ‘tallboy’ or ‘redtape’ whose meanings have nothing to do with their constituents semantically decreases the value of this approach although the semantic approach does appear to work in the case of endocentric compounds whose meanings are dictated by those of their constituent elements.
The Generative Approach
Here, compounds are generated freely from sentential structures and are therefore self-explanatory. They do not need to be defined by dictionaries. The same rules which restrict combinations of words in sentential structures are considered to restrict them in compounds as well. The meanings of compounds are stated in terms of the semantic relations which exist between the underlying sentence relations. Thus, the ‘woman who cleans’ becomes ‘the cleaning woman’.
However, this approach fails to explain exocentric compounds such as ‘tallboy’ or ‘hotdog’ and does not explain the additional meaning in endocentric compounds. For example, it does not explain why every female friend is not a ‘girlfriend’. This approach classifies such words as non-compounds with accidental compound character.

Morphology: Inflectional v. Derivational

The inflectional morphology of a language is the study of the ways in which bound grammatical morphemes combine with stems to be realised as grammatical words. On the other hand, the derivational morphology of a language is the study of the ways in which bound lexical morphemes combine with stems to be realised as lexical words.
Classical grammarians of Latin and Greek generally divided grammar into accidence, word formation and syntax. They did not pay much attention to derivation because they did not really consider it to be a part of grammar.
There are three main differences between inflection and derivation. Firstly, inflection refers to the ways in which bound grammatical words combine with stems to form grammatical words as mentioned earlier while derivation ultimately leads to the formation of lexical words. Both grammatical and lexical words ultimately surface as phonological and orthological words in which bound lexical morphemes can usually be identified as having been affixed. These affixes can be divided into inflectional and derivational affixes. Those which realise bound grammatical morphemes (such as –s, –es on plural nouns, ’s on possessive nouns and –d and –ed on the past participle forms of verbs) are called inflectional affixes and have no fixed, concrete meaning of their own while those which realise bound lexical affixes (such as –ish, –al, –able and –ness) are called derivational affixes.
Inflectional affixes never change the grammatical category of the stem: they are all suffixes which form the outer layer of complex words and modify the meaning of the steam in regular ways. This is not the case with derivational affixes which may be either suffixes or prefixes (such as de–, re– and –ize). It is possible for both inflectional and derivational morphemes to occur in the same word. The latter always constitutes the outer layer as no affix can be added after the inflectional affix has been added. Thus, derivation may have an input in inflection but inflection cannot have any input in derivation. For example, in both ‘deindustrialising’ and in ‘depixelating’ the derivational affix ‘de–’ occurs along with a final ‘–ing’ inflectional affix after which no other affix can be added to either word.
Similarly, if there is both compounding and inflection in a word, the latter must follow the former.
In words in which compounding, derivation and inflection all occur, the inflection is last and compounding is first as can be seen in the words ‘kickstarted’ [(kick + start) + ed] and ‘channelhopping’ [(channel + hop) + ing].
Inflectional morphology not only describes bound grammatical morphemes but also the grammatical rules in which they occur, the paradigm they form and the various orthological and phonological forms in which they eventually surface. Derivational morphology, on the contrary, studies the categories of items with which bound lexical morphemes can be combined, the categories to which the resulting forms belong, the changes in meaning brought on by the process of derivation and the orthological and phonological shapes which bound lexical morphemes acquire.
An inflectional affix occurs solely with all the members of a given class unlike derivational affixes which may occur with the members of more than one class or with only some of the members of any particular class. Thus, there are several differences between derivational and inflectional morphology. The most striking though is that the words created through the process of inflectional morphology such as ‘talk’, ‘talks’ and ‘talked’ are not new words. They are merely grammatical forms of the same words. Derivation, however, creates new lexical words with distinct meanings such as ‘amoral’, ‘disown’ and ‘foreground’.