Printer Friendly

Composite predicates and modification flexibility in middle English (1).

It has often been stated (Jespersen 1942; Nickel 1968; Live 1973; Brinton and Akimoto 1999) that one of the main factors that has favoured the increasing use of composite predicates in Present Day English has been the flexibility of modification that such structures allow in comparison with simple predicates. It seems to be the case that composite predicates can be more easily modified than simple verbs, since they admit the insertion of an adjectival phrase qualifying its deverbal object.

In this paper we analyse whether the above claim, valid for contemporary English, applies to earlier stages of the language, particularly the Middle English period, by carrying out a contrastive analysis of the modification patterns observed in a corpus of composite and simple predicates drawn from the Middle English section of the Helsinki Corpus of English Texts.


Composite Predicates (henceforth, CPs) are bipartite units consisting of a verb, mainly do, have, give, make and take, and a noun phrase object. (2) The verb supports the grammatical morphemes while the deverbal object is semantically more salient, (3) with the result that the verbal phrase is a semantic unit.

Sentences 1 and 2 below, contain examples of two core CPs, make manred and make mencion:

(1) Hi hadden him manred maked & athes suoren, ac hi nan treuthe ne heolden. (ME1: The Peterborough Chronicle) (4)

(2) Wherfor flys day holy chyrch makyth mencyon of two comyngys of Crist, Godys sonne, ynto flys world. (ME4: Mirk, John. Mirk's Festial)

The flexibility of modification afforded by CPs has often been claimed to be one of the main factors favouring the increasing use of such structures in Present Day English (for instance, Jespersen 1942; Nickel 1968: 15; Live 1973: 34; Corpas Pastor 1996: 70; Brinton and Akimoto 1999b: 2).

In the present paper we try to determine whether this claim applies to earlier stages of the language, particularly the Middle English period (henceforth, ME), by analysing the different possibilities of determination, modification and complementation of CPs. In order to do this we carry out a contrastive analysis of a series of selected pairs of composite and simple predicates.


Our data consists of all the structures with one of the main light verbs, the verb to make, extracted from the ME section of the Helsinki Corpus of English Texts: Diachronic and Dialectal (henceforth, Helsinki Corpus), with the aid of the concordance program WordSmith Tools.

In the 608.570 words of ME prose and verse that make up the ME section of the Helsinki Corpus, we identified a total of 473 tokens, corresponding to 203 different types of CPs. A first look at our data revealed that the nominal element in 32.56% (154 occurrences, henceforth occ.) of the strings is not preceded by any element whatsoever. Nevertheless, 10.15% (48 occ.) of them are followed by some postmodifying element or a complement (as is the case with the deverbal object restitucion in 3 below), so only 22.41% (106 occ.) of the deverbal objects in our data are neither preceded nor followed by any kind of modifying element, like acord in 4 below).

(3) And also in wyse as we wol flat restitucion be maad of fle forsaide goodes. (ME3: Henry V, Letters)

(4) Vn iarmed out he wende . to fle barons wel stille . & anon made acord . & graunted al hor wille. (ME2: Robert of Gloucester. The Metrical Chronicle of Robert of Gloucester)

The chronological distribution of these bare deverbal nouns, shown in Table 1, shows that the rate of CPs with some accompanying element gradually increased throughout the ME period, a statistically significant distribution, according to the Chi-square test, p<0.001.


In order to ascertain whether the use of CPs in ME was favoured by the flexibility of modification of such structures, we analysed the different possibilities of determination, modification and complementation that the nominal element in the CP admits. We addressed each of them individually and studied how they influenced each other in their distribution.

3.1. Determination

Closed-class items, mainly articles and quantifiers, can modify the deverbal noun. Table 2 contains the global figures of the distribution of these closed-class items in our data, while the detailed chronological evolution of these data is reflected in Table 3.

Table 3 shows that the zero article is the most common determiner at the beginning of the period under study. Although a decrease in its use can be detected chronologically, it continues to be widely used throughout the period, and, as a matter of fact, is the most common determiner in all subperiods. The use of the zero article with CPs had already been noticed by Christophersen (1939: 79), Visser (1963-73: 138), Mustanoja (1960: 272), Mosse (1952: 97) or Jespersen (1942: 447-448).

The rate of incidence of the zero article with no accompanying adjective decreased throughout ME, so that, while the ratio between the zero article with adjectival modifier and the zero article with no adjectival modifier was almost 1: 6 at the beginning of the period, the percentages of both categories had become almost equal by the end of it.

The prototypical pattern for CPs in Present Day English has repeatedly been claimed to consist of a verb followed by an indefinite noun phrase, as is the case with example 5 below (see, for instance, Rensky 1964: 295; Nickel 1968: 4; Live 1973: 31; Stein 1991: 1; Dixon 1991: 339; Stein and Quirk 1991: 197; Algeo 1995: 208; Brinton 1996: 187). However, it had already been pointed out by Visser (1963-73: 138) that this was not the case at earlier stages of the language. Our data in this respect confirm this claim. A look at Table 3 reveals that the end of the ME period witnessed an increase in the use of the indefinite article a, which by that time had become the second most common determiner. This may indicate that CPs were slowly approaching the formal pattern just mentioned.

Other shades of referential meaning may be added by the possessive, the negative determiner, or quantifiers, which frequently premodify the deverbal noun in our data, as exemplified in 6, 7 and 8 below, respectively.

(5) Iohn Reynys and Emme, the wyff of Roberd Reynys, made a covenaunt with Cecilie Grene of Hemlyngton in the name of Roberd Reynys of Acle. (ME4: Reynes, R. The Commonplace Book Of Robert Reynes Of Acle)

(6) Swilk houre cumes, als men sal se, fle vmber of Jews law bright sal be When men sal noght mak flaire praiere In flis hill flat flou sese here. (ME3: The Northern Homily Cycle)

(7) Pardee, ye knowen wel that ye maken no deffense as now for to deffende yow, but for to venge yow. (ME3: Chaucer, G. The Tale Of Melibee)

(8) Hens, thevys! 3e haue made many a lesynge. (ME4: Mankind. The Macro Plays)

3.2. Adjectival and prepositional phrase modifiers

The deverbal noun may be accompanied by content adjectives as well as by members of closed-class items, as is the case with solempne oth in example 9, or by participial adjectives, as in 10 below. 28.75% of the tokens in our corpus (136 occ.) are adjectivally modified.

(9) Whan flei had mad a solempne oth, flei went to mete into fle kyngis tent, but flei sey not fle kyng. Aftir mete flei were comaunded for to go with certeyn lordes flat schuld kepe hem. (ME4: Capgrave, J. Capgrave's Chronicle.)

(10) Of that weche seyd payment weell and treuly as it [{is{] aboven made, I bynde me and myn executourys by this present. (ME4: Reynes, R. The Commonplace Book Of Robert Reynes Of Acle)

From the chronological point of view, Table 4 indicates that after the increase in the rates of modification that took place during the transition from ME1 to ME2, these did not undergo significant variation until the very end of the period.

102 tokens of the head of these adjectival phrases correspond to 18 types. Thus, the number of types of modifying adjectives is far smaller than that of tokens, which means that the same words are used repeatedly. Adjectives such as good (12 occ.), new (4 occ.), or much(el) (16 occ.) are frequently used. The most common one is great, which appears in 27 instances, that is, in 19.85% of the total number of adjectivally modified structures. Our results agree with Tanabe's claim that "other than good, frequently used adjectives semantically fall in the category of 'numerous, big'" (1999: 115). The large number of quantifiers that have been seen to accompany the deverbal noun also contribute to this idea of quantity.

Much(el), great and good make up more than 40% (40.44%, 55 occ. out of 136 adjectival phrases) of the total number of occurrences of adjectival modifiers in our corpus. Matsumoto (1999: 83) has argued that the adjectives grete and good modifying CPs act as emphasizers which do not affect the core meaning of the CPs they modify because "the combination of verb + agentive object forms a tightly connected unit; in contrast, for a CP with a 'true' modifier, the connection of verb and agentive object is looser than for a CP rejecting such a modifier".

In order to try to determine the extent to which the behaviour of the CPs in our corpus agrees with Matsumoto's findings, we investigated in further depth the adjectival modifiers of those strings which were recorded four or more times and which only admit the zero article, and which therefore seem to show a high degree of syntactic fixation, indicative of a tighter link between verb and noun. This is the case with make mencion, make serche, make accord, make amend, make sorrow, make bliss and make peace. They all permit adjectival modification. Interestingly, the statistically significant data (p<0.001) in Table 5, shows that 81.82% (18 occ.) of the modified CPs in this subgroup take one of these high frequency emphasizers, therefore confirming the connection between more fixed CPs and emphasizers.

Nor does the core meaning of the CP seem to be affected in cases in which the adjective modifying the deverbal noun is an epithet, as is the case with examples such as woful moan in 11 below, sorry del in 12, both from verse texts, or wunderlice miracle (in a prose text), quoted in 13 below.

(11) Bot after, whanne I understonde, And am in other place al one, I make many a wofull mone Unto miself, and speke so. (ME3: Gower, J. Confessio Amantis.)

(12) Mi douter louede him al to wel; For fli maki sori del. (ME2: Dame Sirith)

(13) & He maket flur[{h{] ure Drihtin wunderlice & manifaeldlice miracles. (ME1: The Peterborough Chronicle)

With regard to the position of this adjectival phrase, it practically always precedes the head noun it modifies (94.12%, 128 occ.), as was the case with closed class items. Only 8 instances (5.88%) of postposition have been found. 7 of them occur in end of line position in verse texts, in such a way that their placement in the line seems to be required by the rhyme (as looks to be the case in example 14 below). The other instance of postposed adjectival phrase has been extracted from a prose text, but it is the adjective alone in 15 below, which cannot precede the noun it modifies.

(14) Wifl Canduleke he wendefl swifle. His kniJttes maden chere blifle, For her lorde in tapynage Was ywent in flat veiage. (ME2: Kyng Alisaunder)

(15) Men forsofle flat befl now beddifl to be mynusched in causon in a litil quantite, flat flere be made a ventosynge alone & infrigidacoun, (^i.e.^), coldnes. (ME3: A Latin Technical Phlebotomy And Its Middle English Translation)

Although it has been claimed (Tanabe 1999: 115) that "one of the advantages of using CPs is that, by the use of adjectives, the degree of emphasis becomes adjustable by utilizing the combination of adjective and adverb", in our data the adjectival phrase most often consists only of the head adjective. Only 9 instances (6.67%) of modified adjectives have been found. All but one of them occur in verse texts, as is the case with example (16) below. Thus, 25% (8 out of 32) of the adjectives in verse texts are premodified, while only 0.96% of the adjectives in prose texts (1 out of 104) are. This statistically significant difference (p<0.001) seems to indicate that verse texts favoured the presence of more complex adjectival phrases.

(16) But hise children alle fyue Alle weren yet on liue,/ flat ful fayre ayen hem neme/ Hwan he wisten flat he keme,/ And maden ioie swifle mikel-/ Ne weren he neuere ayen hem fikel.

Of the 473 CPs in our corpus, 369 were found in prose texts, while the remaining 104 belonged to verse texts. When adjectival modification is taken into account, no significant differences arise regarding the nature of the text to which the CPs belong (p<1), as can be gathered from the results in Table 6.

Participial adjectives may also accompany the deverbal noun. This is the case with 12 tokens in our corpus. 10 of them belong to the very end of the ME period. They are nearly all restricted to official or legal texts, as in 17 below.

(17) the sute flat fle seyd Walter made for supportacion in flis seyd matier was be fle meene of fle seyd Thomas Erpyngham to fle seyd Duk of Gloucestre by whose reule and commaundement fle seyd arbitrement and award was mad in fle fourme aforn seyd. (ME4: Memorandum To Arbitrators 1426-7)

Instances of prepositional phrase and clausal modification have been found. These two always occur in the slot following the deverbal noun.

(19) And whan Je walke togedyr or on sondry, Je schall make none noyse wherthurgh fle dwelleris withinne Jour toun schuld ben dystrobled or lettyd of ther rest ... (ME4: Reynes, Robert. The Commonplace Book Of Robert Reynes Of Acle)

3.3. Relativisation

In addition to all these forms of modification the nominal element of CPs may be relativised, a phenomenon attested 33 times (6.98%) in our data.

(20) fle lesinges flat flou hauest maked, fler flou shalt hem forsake, And shome fle shal bitide. (ME2: The Thrush and The Nightingale)

When both relativisation and the different types of modification are taken into account, the percentage of modified structures increases to 38.90%, and so the difference between modified and non modified deverbal nouns becomes more evident. As was the case when the chronological distribution of the instances of adjectival modification was analysed, a contrast between the very beginning of the period and the following sub-periods becomes evident (see Table 7). This may have been influenced, partially at least, by the limited evidence available from that first subperiod. No instances of prepositional phrase modification or clausal modification occur until ME2, and only one example of relativisation belongs to the very beginning of the period.

Postposed modifiers, including prepositional phrases, clauses and instances of relativisation, seem to favour the presence of an overt determiner, as the statistically significant data (p<0.001) in Table 8 shows. Postmodifying elements particularly favour the presence of the definite article, since 23 out of the 56 (41.07) tokens with postmodification take the definite article (Table 9, also statistically significant, p<0.001). Examples 21a, 21b and 21c, all taken from The Metrical Chronicle of Robert of Gloucester, from ME2, provide support for this claim about the relationship between postmodification and the definite article. In 21a the deverbal object fourme is not preceded by the definite article, but is 21b and 21c, when it takes postmodification:

(21a) & anon made acord . & graunted al hor wille . & made fourme god inou . & suor is wel to holde . flo hii adde al hor wil.

(21b) Ech bar him ek amorwe . out of toune drou . & of fle fourme flat hii made.

(21c) Nom & let hom go aliue . & destruede al flen toun . Ac fle fourme flat he made . aJe fle barons biuore.

3.4. Complementation

The fact that most nominal elements of CPs are abstract nouns favours complementation, which may be either clausal or phrasal. 23.47% (111 occ.) of the deverbal objects in our corpus are complemented,.

(22) Our Lord piteful & merciful made minde of his wondres; he Jaf mete to fle doutand hym. (ME2: The Earliest Complete English Prose Psalter)

(23) And whan sir Percyvale saw hym chace them so, he made grete sorow that hys horse was away. (ME4: Malory, T. Morte Darthur)

The data retrieved with regard to the chronological distribution of complemented CPs, statistically significant according to the Chi-square test, p<0.001, evinces a tendency towards an increasing use of complementation, as Table 10 shows.

Table 11 below shows that the presence of a complement significantly affects the distribution modifier (p<0.005), and so, CPs whose deverbal nouns do not introduce a complement are more likely to be modified or relativised. Bearing this in mind, it seems relevant that rates of modification have remained fairly stable despite the gradual increase in the number of complemented CPs.


After reviewing some of the determination, modification and complementation possibilities of CPs with the verb to make in ME, we focus attention on several structures which had a semantically equivalent simple verb in the ME period:
   to make accord, to make amend, to make arbitrement, to make bargain,
   to make bliss, to make bost, to make chere, to make complaint, to
   make counsel, to make countenance, to make covenant, to make dole,
   to make end, to make fefment, to make feste, to make forward, to
   make fin, to make joy, to make lesing, to make mencion, to make
   minishing, to make mone, to make noise, to make ordinance, to
   make oth, to make peace, to make prayer, to make proclamacion, to
   make replicacion, to make serche, to make sorrow, to make statut,
   to make work.

We analysed in further detail those CPs which occur more than 4 times in our corpus. This is the case with 33 different types and 232 tokens. 93.94% of these high frequency CPs already had an equivalent one-word verb counterpart at some point in the ME period. It is revealing that, of many of these CPs with a semantically equivalent simple verb counterpart, no occurrences of the simple verb, or only an isolated example, were found in the ME section of the Helsinki Corpus.

We selected a number of fairly clear illustrative examples of those CPs for which evidence of a simple verb has also been found, which we contrast next.

We first compare the highest frequency CP in our data, make sorrow, and its cognate one-word verb to sorrow. (6)

Although both noun and verb have been available since OE, the CP was not recorded in the Helsinki Corpus until ME2. In this subperiod only one instance of the CP was found. In contrast, in the ME3 period no occurrences of the simple verb were identified. In the final subperiod of ME, sentences containing both the CP and the simple verb coexist.

As for those CPs for which evidence of a simple verb has also been found, we selected a number of fairly clear illustrative examples, which we contrast next.

Only 5 of the 19 occurrences of the CP make sorrow take no adjectival modification, but three of these take adverbial modification instead. The remaining 14 contain some sort of modification, determination or complementation, or a combination of these.

Syntactically, the CP may appear with no modifier or complement whatsoever, as is the case with example 24a. It may take adjectival modifiers, as in 24b below, as well as adverbial modifiers (see 24c), and it may also take clausal or prepositional complements, as is the case with 24d below. Great and much(el) combined with the zero article frequently modify this CP.

As for the simple verb, in the earlier occurrences (from ME2) it nearly always appears on its own. In contrast, in the data from the end of the period, it is very frequently complemented either clausally or prepositionally, and sometimes adverbially modified, as in 24e below.

(24a) And myche peple of flis citee caam wifl his wydwe and maden sorwe. (ME3: English Wycliffite Sermons)

(25b) And than she made grete sorowe and cryed God mercye. (ME4: The life of St. Edmund)

(25c) For he may make sorow ernestly flat wote & felifl not onli what he is, bot flat he is. (ME3: The Cloud of Unknowing)

(25d) And whan sir Percyvale saw hym chace them so, he made grete sorow that hys horse was away. (ME4: Malory, T. Morte Darthur)

(25e) I sorow moche for your lyf. (ME4: The History of Reynard The Fox)

Since modified simple verbs are attested by the end of the ME period, it does not seem likely that modification alone has favoured the use of the CP. Rather, it may be a matter of choice by the different authors, and so, in the sample from Morte Darthur in the Helsinki Corpus, no instances of the simple verb were found (the CP was recorded 8 times), while in The History of Reynard the Fox the CP was not found, and the verb was attested 5 times. Although the simple verb is used in The History of Reynard the Fox, this does not imply a decrease in modification

Adverbial phrases may also modify the CP make sorrow. Both the adjectives and adverbs that modify this CP coincide in that they convey an idea of quantity (see, for instance, 24b and 24f). Matsumoto (1999: 87) has related the use of adverbial modifiers with CPs to a tighter connection between verb and noun. As a matter of fact, this CP was already claimed to be somewhat fixed referring to the distribution of emphasizers. In relation to the above, the adverbial phrase out of measure has also been recorded as modifying both this CP and to make dole, thus adding to the idea of fixation / convention.

We now turn to the CP make end. (7) As was the case with make sorrow, although both the noun end and the cognate verb end date from OE times, the CP does not occur in our data until the second half of the ME period (ME3). A comparison of the behaviour of the CP and the simple verb in the second half of the ME period reveals that structures with the CP and those with the simple verb usually differ in the distribution of semantic roles.

The subject of the CP is usually an agent and is complemented by an inanimate object, as exemplified in 25a below. In contrast, the simple verb in active sentences usually takes an inanimate patient subject like 25b, as is the case when passivised, as in 24c.

It seems, therefore, that the CP permits an alternative way of distributing the thematic roles in the syntactic structure.

(25a) Ryght as the cony had made an ende of his complaynt cam in corbant the roek. (ME4: The History of Reynard The Fox)

(25b) Here endith the descripcioun of the Astrelabie and here begynne the conclusions of the Astrelabie. (ME4: Chaucer, G. A Treatise on the Astrolabe)

(25c) flis translacioun is i-ended in a florsday, fle ey tefle day of Averyl. (ME3: Trevisa, J. Polychronicon Ranulphi Higden)

Finally, we consider the strings with the CP to make complaint and those with the equivalent simple verb to complain. (8) Due to the date of the first appearance of the noun complaint, all occurrences of the CPs date from the end of the ME period (ME4). All of them belong to The History of Reynard The Fox or to the sample of private letters.

The CP may appear in the following syntactic structures: (1) on its own, (2) modified by an adjectival phrase, as in example 26a, (3) introducing a clausal or prepositional complement, as in 26b, or (4) with a combination of modification and complementation.

The simple verb may appear in isolation or it may be complemented or adverbially modified (as in 3c).

Two of the four CPs recorded belong to the fragment included in the Helsinki Corpus from The History of Reynard The Fox. In this sample of text, 16 occurrences of the simple verb to complain have been found, on only two occasions was the CP used, and on only one of these occasions does adjectival modification occur. Therefore, apart from the influence of easy modification, it seems reasonable to assume that stylistic variation may also have favoured the use of the CP.

(26a) Lapreel the cony and Corbant the roek haue made a grete complaynt also. (ME4: The History of Reynard The Fox)

(26b) A servant of our hath made a complainte of him. (ME4: Paston Letters and Papers of the Fifteenth Century)

(26c) Also to Jow worechepeful and wyse Comunes greuouseliche compleynen alle the kynges tenentJ of the Ryal lordshipe and tounshipe of Chestreton in the Shyre of Cambrigge. (ME3: The Appeal of Thomas Usk)


The analysis of CPs with the verb to make in ME with regard to their flexibility of modification has revealed their versatility in this respect, since CPs are seen to permit a wide array of modification, complementation and determination possibilities through many different structural devices, such as closed-class items, clauses, prepositional phrases, and adjectival phrases.

In addition to the variety of modification possibilities, our analysis has revealed that deverbal nouns are indeed quite frequently modified by either one or a combination of several such modification possibilities.

With regard to the various forms of modification, different interrelations have been revealed operating among them. For instance, postposed modifiers favour the presence of the definite article and complemented CPs, which gradually increase throughout the period, are less likely to be modified than CPs with no complement.

Chronologically, a marked increase in the percentages of modification has been observed in the transition from ME1 to ME2. From then onwards, the rates of modification have remained quite stable, despite the higher numbers of complemented CPs.

The comparative and contrastive analysis of CPs and their equivalent simple verbs in our corpus suggests that, in addition to the importance of the modification flexibility of the CPs other structural or stylistic factors are likely to affect the choice of the CP over the simple verb.
Table 1: Chronological distribution of modified and
unmodified CPs ([chi square] = 33.4, p<0.001.)

                 ME1 (5)       ME2          ME3           ME4

               #      %     #      %      #      %      #      %

NON MODIFIED   17   58.62   20   25.97    42   25.77    27   13.24
MODIFIED       12   41.38   57   74.03   121   74.23   177   86.76
TOTAL          29           77           163           204

Table 2: Determination: global figures

                      #      %

Definite article      53   11.21
Indefinite article    63   13.32
Possessive            33    6.98
Demonstrative         18    3.81
Negative              21    4.44
Quantifier            36    7.61
Zero article         239   50.53
Other                 10    2.11
TOTAL                473

Table 3: Determination: chronological distribution

                      ME1          ME2           ME3           ME4

                   #      %      #      %      #      %      #      %

Definite article    0      --    11   14.29    15    9.20    27   13.24
  article           0      --    12   15.58    19   11.66    32   15.69
Possessive          2    6.90     4    5.19    15    9.20    12    5.88
Demonstrative       1    3.45     5    6.49     1    0.61    11    5.39
Negative            2    6.90     0      --     9    5.52    10    4.90
Quantifier          3   10.34     6    7.79     9    5.52    18    8.82
Zero article       20   68.97    39   50.65    94   57.67    86   41.87
Other               1    3.45    --      --     1    0.61     8    3.92
TOTAL              29            77           163           204

Table 4: Adjectivally modified CPs: chronological distribution

                  ME 1         ME 2         ME 3          ME 4

               #      %     #      %      #      %      #      %

MODIFIED        4   13.79   24   31.17    44   26.99    63   30.88
NON MODIFIED   25   86.21   53   68.83   119   73.01   141   69.12
TOTAL          29           77           163           204

Table 5: Higher degree of fixation and modification
([chi square] = 18.7, p<0.001)

             MORE FIXED   LESS FIXED

             #      %      #      %

EMPHASIZER   18   81.82    37   32.46
OTHER         4   18.18    77   67.54
TOTAL        22           114

Table 6: Modification in prose and verse texts
([chi square] = 0.072, ns)

                 PROSE         VERSE

                #      %     #      %

MODIFIED       105   28.46   31   29.81
NON MODIFIED   264   71.54   73   70.19

Table 7: CPs with relativisation and modification:
chronological distribution

                  ME 1         ME 2         ME 3          ME 4

               #      %     #      %      #      %      #      %

MODIFIED        5   17.24   34   44.16    60   36.81    85   41.67
NON MODIFIED   24   82.76   43   55.84   103   63.19   119   58.33
TOTAL          29           77           163           204

Table 8: Determination and postmodification
([chi square] = 29.8, p<0.001)

                   ZERO ART.E      OTHER

                    #      %      #      %

POSTMODIFIED         9   16.07    47   83.93
NON POSTMODIFIED   230   55.16   187   44.84
TOTAL              239           235

Table 9: Postmodification and definite article
([chi square] = 56.95, p<0.001)

                     DEF. ART     OTHER

                   #      %      #      %

POSTMODIFIED       23   41.07    33   58.93
NON POSTMODIFIED   30    7.19   387   92.81
TOTAL              53           420

Table 10: Chronological evolution of complemented CPs
([chi square] = 8.25, p<0.001)

                     ME1          ME2          ME3           ME4

                   #      %     #      %      #      %      #      %

COMPLEMENTED        3   10.34   12   15.58    38   23.31    58   28.43
NON COMPLEMENTED   26   89.66   65   84.42   125   76.69   146   71.57
TOTAL              29           77           163           204

Table 11: Complementation and modification
([chi square] = 8.60, p <0.005)

                    MODIFIED      NON MODIFIED

                    #      %      #      %

COMPLEMENTED        30   16.30    81   28.03
NON COMPLEMENTED   154   83.70   208   71.97
TOTAL              184           299

(1) An abridged version of this paper was read at the 11th International Conference on English Historical Linguistics (Santiago de Compostela 2000). I want to thank the anonymous Atlantis reviewers for their suggestions, as well as the Spanish Ministry of Education and Science (DGES) for the financial support of a research project on "Idiomatic Verbal Constructions in the History of the English Language", grant number PB96-0955. My special thanks go to Dr. Luis Iglesias-Rabade, of the University of Santiago de Compostela, for his comments on an early draft of this paper.

(2) The term Composite Predicate was coined by Cattel (1984) and has been more recently adopted by Akimoto and Brinton (1999), Matsumoto (1999), Kyto (1999) and Tanabe (1999).

(3) We follow Live (1973) when referring to the nominal element of CPs as deverbal object. Other labels used to refer to this nominal element include "eventive object" (Quirk et al 1985: 750), "agentive object" (Matsumoto 1999: 60).

(4) Examples have been identified in the following way. First, the subperiod to which they belong in the Helsinki Corpus, namely, ME1, ME2, ME3 or ME4, is specified, followed by the title of the work to which the example belongs. A complete description of these texts can be found in Kyto 1993.

(5) ME1 is the first subperiod of Middle English in the Helsinki Corpus, and comprises from the years 1150-1250. ME2 corresponds to the next 100 years. ME goes from 1350 up to 1420, and, finally, ME4 covers 1420 to 1500.

(6) Make sorrow: MED make 8b (b) ~ care (crie, dol, moninge, morninge, sorwe, wo), to utter lamentations, lament, grieve. OED sorrow vb. 1. intr. To feel sorrow or sadness; to regret or grieve.

(7) Make end: MED make 12c (a) ~ an ende, to finish an action or the like, conclude; ~ an ende of, finish (one's work, a war); (b) ~ an ende, to finish a story or speech; ~ an ende of, bring (a story, book) to an end; also, finish telling about (sth.); (c) ~ ende, to come to an end, cease; also, reach an end in space, stop at a point; ~ an ende of, put an end to (sth.), do away with. OED end: vb I. Transitive and absolute senses. 2. a. To bring to an end, conclude, come to a termination of (an action, a speech, a period of time, one's life, etc.; II. Intransitive senses. 5. a. Of a period of time, action, continuous state, series, book, chapter, etc.: To come to an end.

(8) MED complaint 4: a statement of grievances, esp. a formal charge or complaint presented to a court or some other authority. OED complain 8. a. intr. (orig. refl.) spec. To make a formal statement of a grievance to or before a competent authority; to lodge a complaint, bring a charge).


Algeo, John 1995: "Having a Look at the Expanded Predicate". The Verb in Contemporary English: Theory and Practice. Ed. Bas Aarts and Charles F. Meyer. Cambridge: C.U.P. 203-217.

Brinton, Laurel J. 1996: "Attitudes toward increasing segmentalization". Journal of English Linguistics 24:3. 186- 205.

Brinton, Laurel J. and Minoji Akimoto, eds. 1999a: Collocational and Idiomatic Aspects of Composite Predicates in the

History of English. Amsterdam and Philadelphia: John Benjamins.

--1999b: "The Origin of Composite Predicates in Old English". Brinton and Akimoto 1999a. 21-58.

Cattel, Ray. 1984. Composite Predicates in English. Syntax and Semantics, vol. 17. Sydney: Academic Press.

Christophersen, Paul 1939: The Articles: A Study of their Theory and Use in English. Copenhagen: Munksgaard.

Corpas Pastor, Gloria 1996: Manual de Fraseologia Espanola. Madrid: Gredos.

Dixon, Robert M. W. 1991. A New Approach to English Grammar on Semantic Principles. Oxford: Clarendon Press.

Jespersen, Otto 1942: A Modern English Grammar on Historical Principles, vol. VI. Copenhagen: Ejnar Munksgaard.

Kurath, Hans, Sherman M. Kuhn, Robert E. Lewis and John Reidy, eds. 1952--: Middle English Dictionary. Ann Arbor: University of Michigan Press.

Kyto, Merja, compiler 1993 (1991): Manual to the Diachronic Part of the Helsinki Corpus of English Texts. Coding Conventions and Lists of Source Texts. Helsinki: Department of English, University of Helsinki.

--1999: "Collocational and Idiomatic Aspects of Verbs in Early Modern English". Brinton and Akimoto 1999a. 167- 206.

Live, Anna H 1973: "The take-have Phrasal in English". Linguistics 95.3. 1-51.

Matsumoto, Meiko 1999: "Composite Predicates in Middle English". Brinton and Akimoto 1999a. 59-96.

Mosse, Ferdinand 1952: A Handbook of Middle English. Trans. James A. Walker. Baltimore: Johns Hopkins Press.

Mustanoja, Tauno F 1960: A Middle English Syntax. Part I: Parts of Speech. Memoires de la Societe Neophilologique de Helsinki, XXIII. Helsinki: Societe Neophilologique.

Nickel, Gerard 1968: "Complex Verbal Structures in English". International Review of Applied Linguistics 6:1. 1- 21.

Oxford English Dictionary 1994 (1992): CD Rom. Oxford: Oxford University Press.

Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech, and Jan Svartvik 1985. A Comprehensive Grammar of the English Language. London and New York: Longman.

Rensky, Miroslav 1964: "English Verbo-nominal Phrases". Travaux Linguistiques de Prague 1. 289-99.

Rissanen, Matti et al, comp. 1991: The Helsinki Corpus of English Texts: Diachronic and Dialectal. Helsinki: Department of English, University of Helsinki.

Stein, Gabriele 1991: "The Phrasal Verb Type 'to have a look' in Modern English." IRAL 39:1.1-29.

Stein, Gabriele and Randolph Quirk 1991: "On Having a Look in a Corpus." English Corpus Linguistics. Ed. Karin Aijmer and Bengt Altenberg. London: Longman. 197-203.

Tanabe, Harumi 1999: "Composite Predicates and Phrasal Verbs in The Paston Letters". Brinton and Akimoto 1999a. 97- 132.

Visser, F. Th 1963-1973: An Historical Syntax of the English Language. Leiden: E. J. Brill.

WordSmith Tools 1997: Version 2.0, by Mike Scott. Oxford University Press.

Teresa Moralejo Garate

Universidad de Murcia
COPYRIGHT 2002 Spanish Association for Anglo-American Studies (AEDEAN)
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2002 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Moralejo Garate, Teresa
Publication:Atlantis, revista de la Asociacion Espanola de Estudios Anglo-Norteamericanos
Date:Jun 1, 2002
Previous Article:Discriminacion de grupos de conocimiento y adecuacion de las pruebas con las que se realiza.
Next Article:Engler, Balz, and Renate Haas, eds. 2000: European English Studies: Contributions towards the History of a Discipline.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters