Table of Contents
2. The borrowing of derivational suffixes
3. The use of -ment and -ure in English suffixation
3.3 Additional information about the bases from the BNC Baby
3.3.1 -ment derivatives
3.3.2 -ure derivatives
3.4 Discussion of the findings
3.4.1 With reference to Plag‘s Latinate Constraint
3.4.2 With reference to Bauer and Huddleston‘s Class I and Class II affixes
A. List of words containing -ment
B. List of words containing -ure
The Norman Conquest in 1066 was not only an important event in the history of England, but also had a great impact on the English language (cf. e.g. Baugh & Cable 2004: 108 ff.; Faiß 1992: 68). Besides various other changes in the English language system, which will not be discussed here in further detail, many French words were borrowed into English. Some of the French suffixes contained in complex loan words subsequently became productive in English derivation (cf. e.g. Bauer 1993: 225 f.; Faiß 1992: 3; Marchand 1969: 210 f.). However, I argue that the ways in which these suffixes were employed in English word-formation vary greatly (cf. e.g. Marchand 1969: 210 f.). While some of the borrowed suffixes were extensively used in English derivation, others remained largely restricted. This will be illustrated by describing how the borrowed French suffixes -ment and -ure were integrated into the English system of word-formation. Special attention will be paid to the word class and etymological origin of the bases -ment and -ure were combined with. Although sociolinguistic factors might also influence how foreign suffixes were used in derivational processes of word-formation (cf. van Loon 2005: xiii), this term paper will focus on language-internal, morphological factors only.
Next, a short introduction to the phenomenon of the borrowing of derivational affixes into English will be given. In the main part, the cases of the two French suffixes -ment and -ure will be discussed. Using information given in the online version of the Oxford English Dictionary (OED), in various textbooks and in a text sample from the BNC Baby corpus, the occurrence of -ment and -ure in English words and their usage in derivation will be compared. After that, the findings about the use of the two suffixes will be discussed with reference to two theoretical claims: the Latinate Constraint formulated by Plag (1999: 58 ff.) and the classification of affixes into Class I and Class II affixes described by Bauer and Huddleston (2002: 1672 ff.). To conclude, a short summary of the findings as well as of issues which could not be discussed in this term paper will be given.
2. The borrowing of derivational suffixes
Millar (2007: 131) summarizes morphological change as “changes in the morphological structure of lexical items and of inflected forms, and changes in morphological systems.” One such morphological system that can be subject to change is suffixation, for example when new suffixes from other languages enter the language system (cf. e.g. Bauer 1993: 225 f.; Faiß 1992: 3; Marchand 1969: 210 f.; Yule 1991: 63). This can happen when the suffix is “part of a reasonable number of loan words” (van Loon 2005: 200), which the speakers of the recipient language then interpret as consisting of a base followed by a suffix (cf. also Dalton-Puffer 1996: 211). Thereby, the foreign suffix may develop “its own semantic profile”, which eventually allows it to “cross over to indigenous word stock” (van Loon 2005: 200). Combinations of a foreign base with a native suffix, or of a native base with a foreign suffix, are so-called “hybrid formations”, i.e. “complex words which mix elements from the native Germanic part of the vocabulary with elements from the borrowed Romance part of the vocabulary” (Dalton-Puffer 1996: 211). Marchand (1969: 210 f.) stresses that the combination of foreign suffixes with native English bases is only possible under certain conditions: “Before the foreign affix can be used, a foreign syntagma must have come to be familiar with speakers so that the pattern of analysis may be imitated and the dependent morpheme be used with native words. This is ... complicated.” Thus, some foreign suffixes such as -al, -anee or -ity have become productive with English bases only in a few rare cases (cf. Marchand 1969: 211).
Another consequence of the borrowing of derivational suffixes is the occurrence of bound roots (cf. Bauer 1993: 255 f.; Bauer & Huddleston 2002: 1625; Schmid 2005: 41 f.)3 Schmid (2005: 41) gives the following description of them: „morphemartige Wortbestandteile mit identifizierbarer Bedeutung und enger funktionaler Verwandtheit mit freien Morphemen (hoher semantischer Gehalt, wortfamilienbildend), die aber nur gebunden zusammen mit gebundenen Morphemen (Affixen) auftreten“, for example neeess- in neeessary, neeessity or aggress- in aggressive, aggression, etc. However, there is a distinction between words that were borrowed as a whole, such as leisure, and words whose morphological structure can actually be analysed as consisting of a bound root followed by a suffix, such as capture: “In many cases ... it is questionable whether a word that is etymologically divisible into base + affix can properly be assigned a morphological analysis of that kind in Present-day English ...” (Bauer & Huddleston 2002: 1668 f.).
After this general introduction to the borrowing of derivational affixes, next, the use of the French suffixes -ment and -ure in English suffixation will be compared.
3. The use of -ment and -ure in English suffixation
A comparison of the suffixes -ment and -ure is interesting because they share several common characteristics, but the extent of their usage in English derivation varies. At first sight, the two suffixes seem to be similar. Both entered the English language via complex French loan words, both were used to derive nouns from verbs and both are no longer productive today, i.e. they cannot be used to form words in Present-Day English (cf. e.g. Schmid 2005: 171-3). Despite these similarities, the ways in which -ment and -ure were used in English suffixation vary. One of the most obvious differences lies in the number of complex loans containing -ment or -ure compared to the number of lexemes formed in regular derivational processes, which indicates a differing degree of productivity of the two suffixes.
3.1.1 General development
According to the OED Online, -ment (in its earlier forms also: ME -mentt, ME-16 -mente) entered English via complex French loan words, the earliest instance of which was sacrament (1175) (cf. Jespersen 1942: 375). -ment was mainly used to form abstract nouns from French and English verb bases, often denoting the result or product of the action which the verb described, or the action itself. The suffix was used with Germanic bases from the late 14th and the 15th century onwards, e.g. in f eggment (1340), hangment (1440) or onement (1445). According to Bauer and Huddleston (2002: 1703), this shows that -ment “became naturalised as English very early on.” Similarly, Dalton-Puffer (1996: 109) stresses this peculiarity of - ment: “Certainly in comparison with other Romance suffixes MENT has an impressively high share of derivatives which can be analysed as having a Middle English word as their basis.” Derivatives dating from the 14th and 15th century on Romance bases, such as chastisement (1340), were either borrowed as a whole or formed in English as regular derivatives. In the 16th and 17th century, both formations from Romance verbs, such as banishment (1507), as well as from Germanic verbs, such as acknowledgment (1567) or amazement (1590), are attested. From the 17th century onwards, formations with the prefixes en-/em- (the French form of the Latin prefix in-) and be- (a native English prefix) were particularly common, such as betterment (1605), enlightenment (1621), bewilderment (1820) or embodiment (1828).
From the 16th century onwards, -ment was in a few cases also used to derive nouns from adjectives, e.g. in merriment (1574) or oddment (1780)4
Most linguists agree that -ment is no longer productive (cf. e.g. Bauer 1993: 49; Bauer & Huddleston 2002: 1703; Schmid 2005: 173)5.Consequently, all complex lexemes containing -ment are lexicalised, i.e. they “could not be formed with their present meaning by the current rules of word-formation” (Bauer & Huddleston 2002: 1629). Bauer (1993: 88) similarly states that...
... there is an inverse relationship between productivity and ... lexicalization such that the most productive patterns are not lexicalized, and fully lexicalized processes are not productive. This leads to the view... that productivity is not an either/or phenomenon, but presents a cline.
Due to the unproductiveness of the suffix -ment, the only example of a relatively new coinage cited in several textbooks is Englishment, which is, however, not listed in the OED and “ought not ... be given too much weight” (Bauer 1993: 55). Similarly, Plag’s (2003: 76) search in the OED for -ment derivatives formed after 1950 results in only three relevant coinages: endistancement (1961), tracklement (1954) and underlayment (1956).
Surveying the diachronic productivity of -ment, Bauer (2001: 8) notices two peaks in the early 17th century and in the early 19th century. However, he states that there is no obvious explanation for this diachronic variation in productivity, “we merely observe that productivity appears to change” (Bauer 2001: 9). Nevertheless, Bauer (2001: 184) suggests one possible language-internal reason for the decline in the productivity of -ment in the 20th century: he compares the relative productivity of -ation and -ment, i.e., “the percentage which each formation-type contributes to the new nominalisations.” His findings show “something of a trend: a rise in the productivity of the -ation variant as -ment declines in productivity” (Bauer 2001: 184). However, on the whole there is no obvious explanation for the diachronic variation in the productivity of -ment and it is possible that the causes do not all lie within the language system itself.
To conclude, it can be summarised that -ment is no longer productive today and its derivatives are lexicalised.
According to Plag (1999: 73), “the specification of the domain of -ment is extremely difficult, and most of the generalizations that have been proposed in the literature are very crude and often accompanied by numerous counterexamples.” Nevertheless, some linguists do outline common attributes of the bases to which -ment was attached. Concerning the stress patterns of the bases, Jespersen (1942: 376) notes that “a favourite type of stem with this suffix seems to be disyllables with stress on the second syllable”, such as in commencement or refreshment. This tendency could as well provide a possible explanation for the high number of prefixed bases that -ment was combined with, “since prefixed monosyllabic stems will be necessarily disyllabic with ultimate stress” (Plag 1999: 74).
Bauer (2001: 26) lists three examples in which he claims that -ment has been combined with bound roots: augment, sacrament and tegument. However, according to the OED, these forms were actually borrowed as a whole from French and cannot be analysed as consisting of a bound root followed by the suffix -ment. Besides, no further examples of formations in which -ment was attached to a bound root are given. Thus, it seems that these cases are either very rare or, according to Carstairs- McCarthy (2011: 108), do not exist at all.
Bauer (2001: 27) furthermore identifies several examples of -ment derivatives on bases other than verbs, namely basement, bushment, funniment, knowledgment, merriment, oddment, rabblement and scarcement. However, according to the OED, scarcement, knowledgement and merriment were actually derived from verbs. Basement was partly derived from base as a verb and partly from base as a noun, and bushment is an aphetic form of one of the nouns ambushment or f abushment. This leaves only funniment and oddment, which were built on the adjectives funny and odd, and rabblement, whose base was the noun rabble, as clear examples of -ment being combined with bases other than verbs.
Lastly, concerning its morphophonological effects, Bauer (1993: 122) and Dalton-Puffer (1996: 109) class -ment as one of those suffixes which do not affect the stress pattern of the base, i.e. as a stress-neutral suffix.
To sum up, the stress-neutral suffix -ment was mainly combined with free verb bases, which were often disyllabic.
3.2.1 General development
According to the OED, -urei entered English via complex French loan words such as scripture (a1300), figure (1393) or juncture (1589).й The suffix was mainly used to form abstract nouns from verbs, often denoting an action or process or its result or product. In addition, -ure was used to describe a function, state, rank, dignity, office or collective body, e.g. in judicature or legislature. There is a wide range of French and Latin loan words containing -ure, e.g. f clausure or aperture. Furthermore, -ure was used to form derivatives from Latin bases, such as f vomiture or f ructure, as well as from bases of Romance origin, such asf bankrupture or f disembogure. Lastly, the OED lists three -ure derivatives from native English verbs, namely f clefture, f ructure and wafture. However, f ructure was actually derived from the Latin verb f ruct. In the case of wafture, the etymological origin is 6 The OED also lists another suffix -ure2 which only occurs in scientific animal names. not completely transparent, as either the verb waft of Dutch or Low German origin or the verb f waft of Germanic or Old English origin might have been used as a base.
The suffix -ure is no longer productive and the words ending in it are lexicalised (cf. e.g. Bauer 1993: 222; Bauer 2001: 191, 195; Bauer & Huddleston 2002: 1704; Schmid 2005: 173). Marchand (1969: 350) as well as Koziol (1972: 259) list U.S. licensure (1846) as the most recent coinage containing the suffix. Bauer (2001: 180) estimates that -ure reached its maximal productivity in the early seventeenth century. This was a time of controversy about the introduction of the so- called “inkhorn terms”, i.e. words which were consciously borrowed into English from Latin or Greek to “enrich” the English lexicon. As they were “often obscure and even incomprehensible to ordinary readers”, most of them did not persist (Carstairs- McCarthy 2011: 109). Consequently, the suffix -ure probably competed with other nominalising suffixes such as -age, -al, -ance, -ence, -ery, or -y (cf. Bauer 2001: 180). This competition might be a possible explanation for the relatively few derivatives containing -ure compared to those with -ment, and also for the rather high amount of -ure words that are now obsolete (cf. Bauer 2001: 179).
Bauer and Huddleston (2002: 1704) note that most words containing the suffix -ure are in fact loans (cf. also Bauer 2001: 179). Furthermore, they state that the suffix sometimes occurs with bound roots, namely in capture, leisure and treasure. For capture, a morphological analysis as a bound root capt- followed by the suffix -ure seems appropriate, given the morphologically and semantically similar words captable (Adj), f captate (V), captation (N) or caption (N). In the case of leisure, related derivatives such as leisureness (N), leisurable (Adj), leisurely (Adj or Adv), leisurably (Adv), etc., all contain the word leisure as a whole. Therefore, it is questionable whether leisure (N) can actually be analysed as leis- + -ure.
 Throughout this term paper, I refer to the online version of the Oxford English Dictionary. All entries have been accessed in the period between December 22, 2011 and February 12, 2012.
 Burnard, L. & Berglund Prytz, Y. (eds.). (2008). BNC Baby [CD], Release 2.1, Xaira 188.8.131.52. Oxford: University of Oxford, RTS.
 Some linguists instead use the term bound base (e.g. Bauer & Huddleston 2002: 1625; Huddleston & Pullum 2010: 282) or bound stem (e.g. Yule 1991: 61) for this concept.
 According to Bauer & Huddleston (2002: 1703), merriment was actually derived from the now obsolete verb merry, and only appears to be de-adjectival in a synchronic morphological analysis.
 A slightly different approach to the question of the productivity of - ment is suggested by Plag (1999: 15 f.): he states that -ment would in principle be productive with bases containing the prefixes en-/em- and argues that it is the unproductiveness of these prefixes which is the reason for the absence of new forms containing both en-/em- and -ment, rather than the unproductiveness of -ment as a suffix. This matches Bauer and Huddleston’s (2002: 1630) observation that the degree of productivity of a certain affix is “to a large extent... determined by the size of the class of bases available to the process.”