Borderline between compound words and free word-groups. Inseparability of compound words.

Compound words as inseparable vocabulary units taking shape in a definite system of grammatical forms and syntac­tic characteristics are generally clearly distinguished from and often opposed to free word-groups. Their inseparability finds expression in the unity of their structural, phonetic and graphic integrity.

Structurally the inseparability of compounds manifests itself in the specific order and arrangement of stems which stand out most clearly in all asyntactic compounds. The difference between words and stems even when they coincide morphemically is espe­cially evident in compound adjectives proper. Adjectives likelong, wide, rich are characterised by grammatical forms of degrees of comparisonlonger, wider, richer. The correspond­ing stems lack grammatical independence and forms proper to the words and retain only the part-of-speech meaning, thus compound adjectives with adjective stems for their second components, e.g.age-long, oil-rich, do not form degrees of comparison the way wordslong, rich do. They conform to the general rule of polysyllabic adjectives having analytical forms of degrees of comparison. This difference between words and stems is not so noticeable in compound nouns with the noun stem for the second component, as the paradigm of the compound word coincides with the paradigm of the noun whose stem constitutes its structural" centre.

Phonetically compounds are also marked by a specific structure of their own. No phonemic changes of stems occur in composition, but the compound word acquires a new stress pattern, different from the stress in the words with similar stems, for example, wordskey andhole orhotandhead each possesses its own stress but when the stems of these words are brought together to make up a new compound word'keyhole—'a hole for receiving a key' or'hot-head— an impetuous person', the latter is given one stress, a unity stress on the first component.

Compounds have three stress patterns:

1)a high or unity stress on the first component as in'catnap—"a short, light sleep','hot-head, 'doorway—'the opening into which a door fits in', 'face-ache—'neural­gia';

2)a double stress, with a primary stress on the first component and a weaker, secondary stress on the second component, e.g.'blood-, vessel—'a tube through which blood flows in the body','mad-doctor—'a psychiatrist','washing-,machine, 'money-border—'an order for the payment of money by a post-office'. These two stress patterns are the common­est among compound words and in many cases they acquire a contrasting force distinguishing compound words from word-groups, especially when the arrangement and order of stems parallel the word-order and the distributional formula¹ of a phrase, thusa'green-house—'a glass house for cultivat­ing tender plants' is contrasted toa 'green 'house — 'a house that is painted green','dancing-girl—'a dancer' to'dancing 'girl—'a girl who is dancing','missing-, lists— 'lists of men and officers who are missing after a battle' to'missing 'lists—'lists that are missing','mad-.doctor—'a psychiatrist' to'mad 'doctor—'a doctor who is mad'. The significance of these stress patterns is nowhere so evident as in nominal compounds, built on the n-\-n distributional formula in which the arrangement and order of the stems fail to distinguish a compound word from a phrase;

3) It is not infrequent, however, for both components to havelevel stress as in, e.g.,'ann-'chair, 'icy-'cold, 'grass-'green, etc.

The significance of the stress pattern by itself should not be overestimated though, as it cannot be an overall cri­terion and cannot always serve as a sufficient clue to draw a line of distinction between compound words and phrases. This mostly refers to compound words with level stress as the same stress pattern is typical of word-groups. In this case it is important to look for structural and graphic indi­cations of inseparability.

Graphically most compounds have two types of spelling—they are spelt either solidly or with a hyphen, Both types of spelling when accompanied by structural or phonetic peculiarities serve as a sufficient indication of in­separability of compound words in contradistinction to phra­ses. It is true that hyphenated spelling when not accompanied by some other indications of inseparability may be sometimes misleading, as it may be used in word-groups to under­line the phraseological character of the combination as in, e.g.daughter-in-law, father-in-law, man-of-war, brother-in-arms, etc. which are neither structurally, nor phoneti­cally marked by inseparability.

The two types of spelling typical of compounds, however, are not rigidly observed and there are numerous fluctuations between solid or hyphenated spelling on the one hand and spelling with a space between the components on the other, especially in nominal compounds built on the n+n formula. The spelling of these compounds varies from author to author and from dictionary to dictionary. For example, words— war-path, war-time, money-lender—are spelt both with a hyphen or solidly;blood-poisoning, money-order, wave­length, blood-vessel, war-ship—with a hyphen and with a break;underfoot, insofar, underhand—solidly and with a break. This inconsistency of spelling in compounds, very often accompanied by a level stress pattern (equally typical of word groups) makes the outer indications of inseparability stand out less clearly and gives rise to the problem of distin­guishing between compound words and word-groups.

The numerous borderline cases between compounds and word-groups are connected with one of the most controver­sial problems in word-composition, known in linguistic liter­ature as "the stonewall problem", in other words the prob­lem whether complexes likestone wall, peace movement, summer days regularly spelt with a break should be regarded as compound words or word-groups. The solution of the problem centres on the nature of the first member of such formations. There are two approaches to this problem and linguists, consequently, give different appraisals of the graph­ic and phonetic integrity of such complexes.

Some linguists class such complexes as a specific group of compound words on the ground that the connection between the members of such complexes cannot be regarded as syntac­tic, as the usual means of connection between two nouns typ­ical of Modern English syntax is either the possessive case or various prepositions. They consequently conclude that the connection in formation of the "stone wall" type is asyntactic, hence the members of these complexes are not words but grammatically unshaped elements, i.e. stems. As a junction of two noun-stems they are referred to compound words. The asyntactic structure is taken for a sufficient proof of their inseparability and lack of graphic integrity is disregarded. The proponents of this point of view go on to state that these complexes may also be interpreted as combinations of an ad­jective with a noun, the adjective being formed from the noun-stem by means of conversion for the given occasion, in which case a compound word would remain primary and a word-group secondary. This brings the linguists to the con­clusion that these complexes make a specific group of com­pound words, often termed neutral. They are characterized by. structural instability due to which they can be easily disintegrated into free word-groups under the influence of parallel attributive combinations, level stress and spelling with a break between the components.

The above-cited treatment of these nominal complexes and the disregard of the outer, formal manifestations of in­separability is open to grave doubts. On the one hand, the productivity of conversion in formation of adjectives does not seem convincing because there are very few adjectives of the type in independent use in Modern English; on the other hand it is argued that Modern English nouns in the Common case, singular are used in the attributive function and a purely syntactic nature of the combination of two full-fledged nouns has been almost universally recognized in the last few decades. If we share the opinion, we shall come to the obvious conclusion that there exists a nominal type of free phrases built on the formula N+N and a group of nominal compounds built on the n+n formula which stand in correlative relations to each other. The recognition of nom­inal free phrases deprives "neutral compounds" of theoret­ical validity. Nominal compounds remain a specific class of compounds but in this case the distributional formula even in the most indisputable cases has only a weakened distinguishing force and can by no means be taken for an over­all criterion of their inseparability. It is evident that the hy­phenated spelling or at least fluctuations between hyphenated spelling and spelling with a break become most significant in distinguishing nominal compound words from word-groups. Consequently nominal complexes which are regularly spelt with a space between the components and are character­ized by level stress pattern can hardly be regarded as insep­arable vocabulary units. It is noteworthy that occasional compounds of this type which have become registered vocab­ulary units tend to solid or hyphenated spelling.

Semi-affixes

The problem of distinguishing a compound from a derivative is ac­tually equivalent to distinguishing a stem from an affix. In most cases the task is simple enough: the immediate constituents of a compound are free forms, likely to occur in the same phonic character as indepen­dent words, whereas a combination containing bound forms as its im­mediate constituents, is a derivative.

There are, however, some borderline cases that do not fit in, and so present difficulties. Some elements of the English vocabulary occurring as independent nouns, such as man, berry, land, have been very frequent as second elements of words for a long time. They seem to have acquired valency similar to that of affixes. They are unstressed, and the vowel sound has been reduced to [man], although the reduction is not quite regular: for instance, when the concept "man" is clearly present in the word, there is no reduction. As to land, the pronunciation [lxnd ] occurs only in ethnic names Scotland, Finland and the like, but not in home­land or fatherland. As these elements seem to come somewhere in be­tween the stems and affixes, the term semi-affix has been offered to designate them. Though not universally accepted, it can be kept for convenience's sake.

As man is by far the most frequent of semi-affixes it seems worth while to dwell upon it at some length. Its combining activity is very great. In addition to seaman, airman and spaceman one might compile a very long list: chairman, clergyman, countryman, fireman, fisherman, gentle­man, horseman, policeman, postman, workman, yes-man (one that agrees with everything that is said to him) and many others. It is interest­ing to note that seaman and workman go back to the Old English period, but the model is still as productive as ever, which is testified by the neo­logism spaceman.

The second element, -man is considerably generalized semantically and approaches in meaning a mere suffix of the doer like -er. The fading of the lexical meaning is especially evident when the words containing this element are used about women, as in the following: The chairman, Miss Ellen McGullough, a member of the TUC, said ... ("Daily Worker").

In cases when a woman chairs a sitting, the official form of address­ing her is madam Chairman. Chairwoman is also sometimes found un­officially and also chairperson.

The evolution of the element -man in the 70s provides an interesting example of the extra-linguistic factors influencing the development of the language. Concern with eliminating discriminatory attitudes towards women in various professions led to many attempts to degender, i.e. to remove reference to gender in the names of professions. Thus, cam­eraman is substituted by camera operator, fireman by firefighter, po­liceman by police officer or police person. Person is increasingly used in replacing the semi-affix -man to avoid reference to gender: houseperson, businessperson. The fact that the generic sense of 'human being' is pres­ent only in the word man 'adult male' but not in the word woman which is only 'adult female', is felt as a symptom of implicitly favouring the male sex.

A great combining capacity characterizes the elements -like, -proof and -worthy, so that they may be also referred to semi-affixes, i.e. elements that stand midway between roots and affixes: godlike, gentlemanlike, ladylike, unladylike, manlike, childlike, unbusinesslike, suchlike. H. Marchand points out that -like as a semi-affix is isolated from the word like because we can form compounds of the type unmanlike which would be impossible for a free form entering into combination with another free form. The same argument holds good for the semi-affix -worthy and the word worthy. C f. worthy of note and noteworthy, praiseworthy, seaworthy, trustworthy, and unseaworthy, untrustworthy, un praiseworthy.

H. Marchand chooses to include among the semi-affixes also the ele­ment -wise traditionally referred to adverb-forming suffixes: otherwise, likewise, clockwise, crosswise, etc.

Alongside with these, he analyses combinations with -way and -way(s) representing the Genitive: anyway(s), otherways, always, likeways, side-way(s), crossways, etc. The analysis given by H. Marchand is very convincing. "Way and wise are full words, so it might be objected that combinations with them are compounds. But the combinations are nev­er substantival compounds as their substantival basis would require. Moreover, wise is being used less and less as an independent word and may one day come to reach the state of French -meat (and its equivalents in other Romance languages), which went a somewhat similar way, being developed from the Latin mente, Ablative of mens ('spirit', 'character', later 'manner')."

Two elements, very productive in combinations, are completely dead as independent words. These are -monger and -wright. The existing combinations with the element -monger have a strongly disparaging character, e.g.:If any passages of the present tale should startle the reader's faith, I must be content to bear the stigma of a fictionmonger (Waugh). C f. fashionmonger, newsmonger, scandalmonger, warmonger. Only the words that existed in the language from before 1500 are emo­tionally neutral: fishmonger, ironmonger, -wright occurs in playwright, shipwright, wheelwright.

As -proof is also very uncommon in independent use except in the expression proof against, and extremely productive in combinations, it seems right to include it among the semi-affixes: damp-proof, fire-proof, bomb-proof, waterproof, shockproof, kissproof (said about a lipstick), foolproof (said about rules, mechanisms, etc., so simple as to be safe even when applied by fools).

Semi-affixes may be also used in preposition like prefixes. Thus, anything that is smaller or shorter than others of its kind may be pre­ceded by mini-: mini-budget, mini-bus, mini-car, mini-crisis, mini-planet, mini-skirt, etc.

Other productive semi-affixes used in pre-position are midi-, maxi-, self- and others: midi-coat, maxi-coat, self-starter, self-help.

The factors conducing to transition of free forms into semi-affixes are high semantic productivity, adaptability, combinatorial capacity (high valency), and brevity.