Printer Friendly

The pre-verbal I- in early Middle English: an analysis of the formal parameters of the prefixed verbs.

ABSTRACT

In contrast to the majority of works on the pre-verbal i-/y- in Middle English which usually attempt to explain the loss of the prefix, the present study focuses on the formal traits of the prefixed verbs. The intended aim of the study is to ascertain whether in the early Middle English period verbs had to meet any formal criteria for the use of the prefix. The linguistic data collected from seven prose texts composed at the beginning of the thirteenth century in West Midlands show that the morphological status of the EME i- extends beyond that of the preterite participle marker. The morphological dualism of the prefix--often blurred by its opaque semantics--becomes even more evident in a comparative analysis of the verbs taking the pre-verbal ge- in Modern Dutch and German with those preceded by i- in Early Middle English.

1. Introduction

The number of different, sometimes contradictory, statements on the formal circumstances of the use of the pre-verbal i- in Middle English shows that establishing the real morphological status of the prefix on the basis of purely formal traits of the prefixed verbs is problematic. By analysis of the use of the prefix in several selected texts of Early Middle English the current study confronts the traditional view of the prefix as a meaningless marker of the preterite participle with linguistic data in search of any systemic prerequisites within the verb for the use of the prefix. The study is therefore an attempt to ascertain whether purely formal parameters of the verb could motivate the use of the prefix, or whether this type of prefixation was fully independent of the formal factors. (l)

The potential dependence(s) between any of the parameters typical for the verb and the use of the prefix will be formalised according to such criteria as form, paradigmatic membership, mood, tense, etc.

Traditional descriptions of Middle English such as Brunner (1965), or Jordan (1974) usually present the prefix as an important dialectal feature of the South and South-West Midlands. As far as its use is concerned, Mustanoja (1960: 446) designates the prefix as "a sign of the past participle". Similarly Mosse (1952: [section] 95 n. iv) makes mention only of the preterite participle as the sole form that "might be preceded by the pre-verb y-, i-". (2) In their brief profile of Middle English dialects Bloomfield and Newmark (1965: 215) explicitly define the morpheme as "the past participle prefix".

These statements, implying that the use of the prefix was strictly confined to one form, appear to be an overgeneralisation when compared to Visser (1966: [section] 1126), who argues that this particular prefix "cannot be called a marker of the past participle form, since it was also prefixed to infinitives in Old and--under the forms [z.sup.e-], y- and i- --in early Middle English". The paradigmatic distribution of the prefix according to Kuhn and Reidy (1968: 1) includes more than two forms as the prefix precedes various forms.

Discussing the dialectal distribution of the prefix, Brunner (1965: [section]68) also purports that in border areas (it is not stated exactly where) verb forms are found "with both i- and -en", which may be construed as pointing at the ablaut alteration in preterite participles of strong verbs as the paradigmatic determinant of the use of the prefix. (3) Functionally the prefix in Middle English is treated on the same footing as the preterite-participle -(e)n ending not only by Brunner (1965). The loss of the preterite-participle marker -(e)n according to Welna (1996: 2.69. n. 5) was compensated by means of the functionally analogous prefix i-/y-. (4)

That the formal distribution of the prefix within the verbal paradigm also poses a moot point is evident in the statement produced by Wyld (1956: 37), who, in contrast to those mentioned above, points at the preterite as the form with which the prefix is often used in southern Middle English and admits that this preverbal element "indeed may be used before any part of a verb, often with no particular force, though it also has the function of making intransitive verbs transitive" (5) In addition to that, Wyld (1956) speaks about frequent omission of the prefix (probably due to high redundancy), especially in strong verbs. The most recent account of the prefix and its use with the verb in Middle English has been put forward by Iglesias-Rabade (2003:331), arguing that the prefix could convey the idea of a perfective action when prefixed to a verb and therefore usually appears in preterite participles, with the reservation that "some verbs are also prefixed by i-/y- in the present and preterite system". The ambiguities and inconsistencies in the above statements as to the precise parameters of the prefixed verbs prove that this aspect in defining the prefix deserves a separate treatment.

2. Previous studies on the ME i-/y-

In contrast to the various studies on the Old English ancestor of the prefix, which usually attempt to explain the semantic properties of the morpheme, those dealing with the Middle English forms i-/y- more often focus on the plausible reasons for the loss of the prefix rather than on the systemic characteristics of the prefixed verbs.

One of those who deliberately consider the formal background of the verbs preceded by the prefix and try to define its function on the basis of some selected parameters is Pilch (1955a). In his oft-quoted study on the use of i- in preterite participles Pilch (1955a: 284) points at syllable stress as a systemic factor plausibly determining the use of the prefix in forms other than preterite participles. More precisely, he claims that i- might precede those infinitives in which the main stress falls on the first syllable, e.g. i-haten in the infinitive haten, or i-heren in the infinitive of heren. This principle is also applicable to preterite participles of loanwords, e.g. i-crhnet, or i-flut. Furthermore, compound verbs with prefixes of Romance origin and those which were not stressed on the first syllable soon began to form their preterite participles without the prefix, e.g. confermed in contrast to y-confermed. Regarding the metre of the verbs which in infinitive are followed by an adverb, Pilch (1955a: 285) points out that by prefixation in preterite participle these forms assimilate their metre to those of the common preterite participles, e.g. y-don and (of)-y-don. As most of the texts analysed by the scholar are not consistent in the way they treat the forms with the above features this observation cannot be regarded as an overall tendency affecting all verbs with this stress pattern, working without exceptions.

Although the analysis of several prose texts dated to the Middle English period was to provide more information about the reasons for the loss of the pre fix, the findings allowed Pilch (1955b) to formulate a few interesting statements about its functions. According to Pilch (1955b: 39) the ME i- in a similar way as the OE ze- could form verbs semantically different from the simplex and the morphological process remained productive throughout the whole Middle English period. Thus, for Pilch (1955b: 41, 42) the ME heren and (i)heren are two different verbs, each having its own distinctive meaning. Furthermore, these two verbs are used by the scholar to illustrate the paradigmatic distribution of the prefix in Middle English, which was not confined to one form only. As a significant systemic factor presumably controlling the use of the prefix Pilch (1955b: 43) points at the phonetic environment, i.e. the fact that i- occurs more often after proclitics terminating in sounds other than those represented by <e>, &lt;i&gt;, or <u>. (6) With regard to the formal parameters of the verb, he points out that the prefix was retained longest in the imperative, subjunctive, and after some auxiliary verbs. The only form which hardly ever takes the prefix in both Old and Middle English is the present participle, whereas in preterite participles the prefix is almost a regularity. On the basis of these cursory observations Pilch (1955b: 44) builds up the following paradigm of use of the prefix: (i) in present indicative the prefix never occurs as i-, sporadically found as y-, (ii) in present and past subjunctive the prefix occurs as y-, (iii) in bare infinitives preceded by an auxiliary only as y-, (iv) never in present participles, (v) in most instances of preterite participles, usually as y-. Another significant remark is that the finite forms can be identified not only by means of the personal pronouns and inflexions, but also by means of the prefix in question and therefore in his reasoning the prefix constitutes a formal index of these forms. Finally, Pilch (1955b: 63, 64) states that at each stage of the development of the prefix its use is conditioned by semantic, phonetic, and grammatical factors, which tend to full regularisation of the use.

In a few lines of his paper on the pre-nominal use of the prefix, Stanley (1982) refers also to the pre-verbal i-. The first of his observations is based on the records of the prefix in the Middle English dictionary, Kurath et al. (1952) (henceforth MED). Stanley (1982: 25) points at its specific use "as a marker for the past participle, a use that remained active longest", completely disregarding the vast number of infinitives and finite forms recorded therein. Stanley (1982: 32) explains the increase in frequency of the prefix found in the Northumbrian gloss to the Lindisfame Gospels by the fact that i- was a "semantically empty prefix", which is evinced by the usefulness of the prefix in glossing "element by element rather than sense by sense". The brief description of the pre-verbal i-presented in Stanley (1982) to a large extent seems to be based on surveys carried out by other scholars and the produced statements in fact reiterate the traditional interpretation of the morpheme.

Higuchi (1998) examined all the instances of preterite participles preceded by the Chaucerian y-, attempting to prove that the prefix indeed served significant stylistic and grammatical functions. As the main source of data the scholar used The Riverside Chaucer edited by Benson (1988), a text which seems more useful for literary rather than linguistic analysis. Higuchi (1998: 202) regards the fluctuation in the use of the prefix displayed by both prefixed and unprefixed preterite participles as sufficient to claim that y- served a "stylistic function", however without explaining the term "stylistic". The single occurrence of ifelaschiped, a preterite participle derived from the noun felaschip for him is a tangible proof that "the prefix y- was a living affix in Chaucer's days" and "far from being a meaningless appendage, plays a gramanatical role of converting a noun to a verb". The lack of y- in preterite participles used in the pre-nominal position is interpreted as another grammatical property of the prefix, namely as the verbal marker of the preterite participle y- prevents these forms from being used as adjectives. In "split passive" constructions Higuchi (1998: 204) accounts for y- as the preterite participle marker, which "indicates more clearly" that the form taking it is a preterite participle. As the prefix is more preferable in preterite participles used in passive constructions than in perfective ones the scholar suggests that y- was capable of conveying "passive meanings". All in all, in Higuchi's (1998: 206) view the prefix in Chaucer's English constitutes a significant element of the verbal system and its use in preterite participles may depend on stylistic purposes, derivational necessities, semantic differentiation, and syntactic conditions and hence the use of y- "is, therefore, not necessarily optional".

Thorough examination of all the instances of verbs with the prefix recorded in the MED, presented in Lewis (2005), shows that this kind of corpus can also be a useful source of information about the prefix. One of the few remarks relevant to this study is that the prefixed verbs have their origin either in Old English or are based on a pattern established in Old English. Moreover, Lewis (2005: 106) also emphasises that many verbs inherited the prefix not only in their preterite participles, but above all in the finite forms, infinitives, and sporadically even in present participles. Nevertheless, as he further points out, the retention of the prefix by particular forms is uneven. The corpus contains many instances of prefixed verbs of both English and foreign origin, which excludes selective prefixation on this level. Another important feature of the prefixed preterite participles, especially those appearing after the year 1300, mentioned by Lewis (2005: 109), is that they "could be derived from a variety of sources". The vast number of prefixed verbs in various forms and of differentiated origin listed by Lewis (2005:110-125) shows that in Early Middle English the prefix was a commonplace in many paradigms.

This short description of the prefix however is still incomplete as it presents the prefix only from the diachronic perspective. There are in fact two sources which turned out to be extremely helpful in completing this description, i.e. two modern West Germanic languages which have retained the distant relatives of the ME i-.

3. The reflexes of the PGmc *ga-

The retention of the prefix in the pre-verbal position can be observed only in the case of Modern Dutch and German. (7) In both languages the prefix usually occurs as ge-, which is a reflex of the common PGmc ancestor *ga- regarded as the ultimate source of the prefix in all Germanic dialects. (8)

According to various accounts of the prefix in Modern Dutch and German the morphological status of the pre-verbal ge- is at least twofold, i.e. the prefix serves either as a derivational formative (this function is today unproductive in both languages) or as a grammatical index of the preterite participle. In their characteristic of the derivational system of German Eisenberg et al. (2005: 699) enumerate ge- beside such native prefixes as be-, ent-, er-, ver-, and zer-. Describing the inventory of verbal prefixes of Dutch, Donaldson (1981: 182) lists the prefix with be-, er-, her-, ont-, and ver-. Attached to a verb, the prefix usually slightly modifies the meaning of the verb, e.g. Germanfrieren 'to be/get cold' and gefrieren 'to freeze over', or Dutch denken 'to think' and gedenken 'to recall'. As Booij (2001 : 15) rightly points out many of the prefixed verbs in Dutch have a root that does not exist as an independent word, e.g. gebruiken 'to use', or genezen 'to heal'. Moreover, this inseparable prefix in contrast to the other prefixes has no concrete semantic content and its meaning is hardly defmable without the meaning of the stem. The verbal systems of the two languages besides inflexions and ablaut alterations employ the prefix as an additional marker of the preterite participle. The prefix however is attached only to preterite participles of morphologically simple verbs. As the ablaut vowel of some strong preterite participles in German overlaps with that of the infinitive, e.g. fahren--gefahren, schlafen--geschlafen, the prefix is the sole formal marker differentiating these forms. In both languages the use of the prefix is not restricted by the paradigmatic traits of the verb, which is evinced by verbs having both strong and weak preterite participles, e.g. German gemolken/gemelkt. The rules governing the use of the prefix in preterite participles of inseparable and separable verbs are also parallel* Nevertheless, there is a group of preterite participles which are immune from prefixation, i.e. preterite participles of morphologically complex verbs such as German besungen and verbs in which the first syllable is unstressed, e.g. erobert. This group includes also preterite participles with final stress such as modernisiert. (9) Another formal parallelism is that apart from common verbs the prefix appears in preterite participles of anomalous and modal verbs, which can be illustrated with Dutch gekund, gehad, gemoeten, or geweest.

In Dutch however the rule prefixing all preterite participles of morphologically simple verbs applies also to verbs with final-syllable stress, which, in contrast to those in German, take the prefix.

That the prefix is still active in the grammatical systems of the two languages can be observed in preterite participles of apparently recent loanwords from English such as in Dutch gejogd 'jogged', geupgraded 'upgraded', or in German gedownloadet, sometimes also as downgeloadet. The above examples show the high degree of grammaticalisation of the prefix in preterite participle.

4. The data

The present study is based on linguistic data collected from textual material and therefore is corpus-based. For the purpose of the study the number of texts constituting the corpus has been restricted to a selection of literary works. The corpus is therefore a compilation of seven texts of the early thirteenth century, Ancrene Riwle, Ancrene Wisse, Seinte Katerine, Seinte Iuliene, Seinte Margarete, Sawiles Warde, and Hali Mei[eth]had. For the sake of metre and rhyme only prose texts are analysed. (10) The language of the manuscripts is localised to West Midlands and traditionally labelled the AB language. (11) The provenance of the texts confined to one area should exclude the dialectal factor from influencing the analysis. The collection of the data consisted in careful selection of all the verb forms prefixed with i- from approximately 209970 words of the whole corpus and then classifying these verbs into several discrete groups according to the formal features of the verb. This in turn produced a database of 2222 verbs preceded by the prefix, arranged into eleven groups.

5. The analysis

5. 1. Distribution and designation of the prefix

The analysis of the data begins with establishing the exact distribution of the prefix amongst the i- forms and measuring of the degree of prefixation in particular categories. The findings clearly show that the use of the pre-verbal i- in Early Middle English was not confined to one verb form. Although the prefix is frequently a component of the morphological structure of many verbs its distribution is uneven and in some categories of the verbal system the prefix is virtually absent, which has been illustrated below:

Despite the high frequency of the prefix in the analysed textual material no instance of a present participle or preterite subjunctive with the prefix has been found. The preponderance of the prefixed preterite participles seems to prove that the formal association of the prefix with this form differs from that in the other categories.

Figures showing the extent to which the prefixation with i- affects particular forms reveal similar discrepancies and tendencies. These statistical dissimilarities also corroborate the strong preference for the use of the prefix with preterite participles:

That the morphological status of i- attached to preterite participles is distinct from that appearing with the verbs in the infinitive or preterite indicative is also evident from the degree of prefixation of particular categories in the corpus. The values obtained from the statistics form the following picture:

Despite the high frequency of preterite participles amongst the prefixed forms, the ratio of the types to the number of occurrences in this category is apparently low. Furthermore, the other forms, e.g. the present subjunctive or imperative seem to be more consistent in prefixing different types of verbs, but on the other hand the number of tokens in these forms is considerably lower. In two cases, i.e. the present and preterite indicative the distribution of the prefix is comparable, nonetheless, the difference in the degree of their prefixation should be kept in mind. As the data indicate the only tendency in the use of i- is that the prefix is more preferable in non-finite forms (i.e. infinitives and preterite participles) rather than in finite ones. The overall ratio of types to the number of tokens equipped with the prefix indicates low type frequency, which can be construed as gradual weakening of the productive force of the prefix, selective prefixation of those items that meet certain formal criteria, or existence of a fixed number of verbs in which i- constitutes part of the morphological substance of the lexeme.

In a situation where the form of the prefixed verbs is not a constant, but varies in a haphazard fashion one may ask whether designation of the prefix as a formal marker of any of these forms is possible at all. The results of the preliminary analysis of the data show that the answer to this question is not straightforward. Firstly, interpreted in terms of sheer statistics, the number of the prefixed preterite participles can be referred to as the modal value (henceforth My) and in this case the [M.sub.v] = 1805, which means that preterite participles were prefixed most frequently. (12) Secondly, the degree of prefixation of particular forms corroborates the formal association between the use of the prefix and the preterite participle as a grammatical category. This proves that the morphological status of i- in preterite participles is distinct from that in the other forms. The values representing the number of types and tokens show the morphological dichotomy of the prefix even more clearly. As the data show designation of the prefix as a marker of the preterite participle is justified from the formal point of view, nevertheless, the data also evince that the morphological status of i- extends beyond that of a grammatical marker. The i- attached to infinitives or preterite forms does not seem to be an obligatory element imposed by the form and therefore it stands in opposition to the more form-dependent i- in preterite participles. Thus, designation of the prefix on the purely formal ground is possible, but it must take the morphological opposition into account.

5.2. The infinitive

Although the infinitive as a non-finite form does not convey complex grammatical information, it does not mean that this form cannot be analysed from the formal perspective. The analysis has disclosed several systemic characteristics within this category shared by most of the infinitives preceded by the prefix.

The salient formal feature of all the prefixed infinitives is that they are morphologically simple verbs, i.e. they do not contain any bound morphemes in their morphological structure. Though an average infinitive encountered in the corpus can be either inflected or uninflected, i- occurs solely with the latter, and therefore the environment of the preposition to and the inflexional -ne is not favourable to the use of the prefix. Furthermore, on the semantic level the analysed verbs do not form any cohesive group. The prefix occurs both with common verbs such as iseon, ifinden, and with less frequent ones such as iswiken, or i[thorn]olien. The highest number of occurrences has been recorded in the case of iwur[eth]en, which appears 40 times, beside iseon, having 39 tokens, and iheren, with 21 uses. Moreover, the meaning of the stem seems not to be a formal constraint for the use of the prefix with infinitives. The analysis of the number of syllables in the prefixed infinitives shows that two-syllable verbs constitute 90% of all the verbs assigned to this category. As to the stress pattern of these verbs, it perfectly corresponds to that pointed out by Pilch (1955a: 284), i.e. it is the first syllable of the verb that is always stressed. The data show why some scholars used to perceive the prefix as a converter of transitivity, namely 90% of the prefixed infinitives found in the corpus are infinitives of transitive verbs and therefore the majority of these verbs require an object. Nonetheless, there are also instances of intransitive verbs such as iwur[eth]en, or ilimpen, which frequently appear with the prefix and therefore the morpheme cannot be deemed a converter of transitivity.

To sum up, the analysis of the formal parameters of the prefixed infinitives shows that the use of the prefix with verbs in this form seems not to be determined by any of these. Neither the number of syllables nor the requirement of an object by the majority of these verbs can be regarded as formal prerequisites for the use of the prefix, as there is linguistic evidence, also in the texts analysed for the purpose of this study, which excludes the determinative working of these factors. Both the results presented in this sub-section and those obtained in the first phase of the analysis suggest that the morphological status of i- attached to infinitives is far from inflexional.

5.3. The finite forms

The results of the first phase of the analysis show that the finite forms constitute roughly 7.27% of all the instances of the prefixed verbs, which puts this category on a par with the infinitives. As the finite forms preceded by i- vary in respect of their mood, tense, number, and person, the precise settings of the formal parameters of these forms are the major concern in this phase of the analysis.

The finite forms that appear with the prefix considerably differ in respect of their mood. This category consists of three major sub-groups, i.e. the prefixed finite forms in the indicative ("unmarked mood"), the subjunctive, and the imperative ("marked moods"). With regard to the tense of these forms this category can be further sub-divided into:

[FIGURE 3 OMITTED]

The values for the present and preterite indicative show that the finite forms in the indicative mood, i.e. the "unmarked mood" constitute 80.87% of all the finite forms having the prefix. Be it noted that no instance of a prefixed verb in the preterite subjunctive has been found in the corpus. As for the much lower values in the "marked moods", they. simply reflect the lower frequency in the use of these moods. The non-restrictive nature of the mood in the use of the prefix is corroborated by a number of such verbs as iseon, iheren, icnawen, iwur[eth]en, ifinden, and ikepen, which take i- irrespective of the mood and therefore the use of the prefix seems not to be mood-dependent.

The present characteristic of these forms includes also their tense, number, and person. A more detailed collation of the formal traits of the prefixed finite forms and several arithmetical markers should provide information on the morphological status of i- attached to the finite forms in the aforesaid categories. The prefixed verbs in the present indicative are represented by the following values:

The fact that each cell of this paradigm can be characterised with these markers proves that neither the number nor the person could directly block the use of the prefix in this category. The number of types and tokens show however that the distribution of the prefix amongst particular cells is uneven. The high numbers of the prefixed items in the third person singular and plural correlate with the high number of the verbs having these parameters in the corpus (the former 68.6%, the latter 12.9%). Moreover, the type/token ratios show the deep statistical asymmetry between the singular and plural. Another example of opposition in the paradigm between the singular and plural can be found in the third person, namely in the singular, where the third person has the highest number of types and tokens, the ratio of these two values is the lowest in this part of the paradigm, whereas in the plural, where the number of types and tokens has the highest value also in the third person, the ratio of the two values still remains the highest one. The percentages indicating the extent of the prefixation point at the first person singular and plural as the cells with the highest degree of the use of the prefix. The lowest values of this marker have been recorded in the second person singular and the third person plural, which is another paradigmatic asymmetry.

The picture of the prefixed verbs in the present subjunctive is in many ways different from that presented above. Nonetheless, there are also few similarities, which seem to reflect the statistical disproportions in the analysed material rather than systemic mechanisms:

One of the similarities is the accumulation of the prefixed entities in the third person singular. The most striking difference in this paradigm are the two gaps in the first and second person plural, which cause the singular/plural asymmetry. The type/token ratio in the third person singular shows that in the present subjunctive the prefix is quite frequently attached to different verbs. What should be pointed out here is the apparently low degree of prefixation in this cell, which reached barely 1.73%. In two cells, i.e. the second person singular and third person plural a similar overlap to that in the second person singular in the present indicative can be observed. The low attestation of the prefix in these cells however rules out the paradigmatic conditioning. The formal distinction, based on the systemic means used by the verb to generate its preterite form, is the starting point in the analysis of the prefixed verbs in the preterite indicative. According to the raw statistics, the use of either a dental inflexion or an ablaut alteration to indicate the tense has no bearing on the applicability of the prefix. The number of tokens of both strong and weak forms shows that neither of the two paradigmatic groups is more liable to the prefixation:

In this paradigmatic group the sequence of gaps in the overall paradigm is even more random. This paradigmatic asymmetry brings the inflexional character of the prefix attached to the verbs in the preterite indicative into question. Moreover, in this paradigm the cell which has the highest number of types and tokens is the third person singular. The ratio of the two values in this cell however has turned out to be lower than in the other cells. The measurement of the degree of prefixation in particular cells shows that the prefix can hardly be recognised as an inflexion in any of the cells.

In contrast to the paradigm of the weak forms, that composed of the strong forms seems to be more symmetrical, at least in respect of the number of the gaps in the singular and plural. Most of the oppositions found in the paradigm of the strong verbs are similar to those in the paradigm of the weak verbs:

One of the peculiarities in the paradigm of the strong verbs is the apparently high degree of prefixation in the first person singular in comparison with the value in the third person singular. The close arithmetical parallelism between the two paradigms can be observed in the number of types and tokens and their ratios. As the strong preterite forms differ in respect of the ablaut vowel, the above collation must be complemented with an ablaut breakdown of these verbs:

Despite the sparse number of entities in this sub-group, the above classification shows that a strong preterite form preceded by i- can be a member of one of the three major classes of verbs, i.e. Class I (Bi-alternant stems), Class II (Tri-alternant stems), and Class III (Stems with four vowel alternants). The highest accumulation of occurrences can be observed in Class III, whereas the highest differentiation of the prefixed verbs is a characteristic of Class II. The data suggest that the use of the prefix is determined neither by the ablaut vowel in the preterite nor by the number of ablaut alternants of the prefixed verb. Furthermore, most of these verbs have their prefixed counterparts in other forms such as infinitive, present indicative, or present subjunctive. Like the prefixed infinitives, most of these verbs belong to the core of the lexis of Early Middle English. Thus, the strong preterite forms with i- preceding their stems do not belong to one class of ablaut verbs, but they represent all the three classes.

The last category in the group of the prefixed finite forms contains the imperatives. Although imperatives appear quite frequently in the corpus, the sparse number of the prefixed entities deepens the statistical discrepancies in the group of the finite forms. In this category the prefix is used both when the referent of the verb is singular and plural. The type/token ratios for the singular and plural do not differ considerably. For the former the ratio is 0.29, whereas for the latter 0.66. As the number of types is equal, the difference stems from the divergent number of tokens. The index of the degree of prefixation shows that neither in the imperatives with a singular referent, prefixed in 1.19%, nor with a plural, prefixed in 0.32%, the prefix can be regarded as an inflexion. Also many of the verbs in the imperative belong to the basic resources of lexis.

A feature typical of inflexional processes is that they are less likely to produce large numbers of unpredicted gaps in the system and therefore an inflexional morpheme should not leave empty cells in the paradigm of the category it marks. Besides this, the applicability of a morpheme which serves as an inflexion in a given category is usually very high. A collation of both the gaps in the paradigms of particular categories and the degree of applicability of the prefix in question in particular cells of the paradigms is a concluding remark in this sub-section:

The cells in which the prefix has occurred at least once are marked with percentages indicating the degree of prefixation. The above table shows that i- was capable of producing such gaps in the paradigms of the finite forms. Furthermore, these gaps do not form any regular pattern and the percentages indicating the low degree of prefixation in particular cells of the listed paradigms considerably differ, disproving the prefix to be obligatory. The prefixation of the finite forms is therefore not a fully predictable process, governed by simple rules based on clear-cut patterns, and the prefix attached to finite forms cannot be regarded as an inflexion in any paradigm of the categories discussed herein.

5.4. The participial forms

All the participial forms found in the corpus constitute the most numerous category among the prefixed verbs. This category however contains only preterite participles, (13) The prefixed preterite participles can be sub-divided according to their paradigmatic membership and the temporal properties of some verbs. The percentages of particular sub-groups in the total number of the prefixed preterite participles show that the weak forms prevail in this category:

[FIGURE 4 OMITTED]

The above percentages however represent only one of the arithmetical markers used in the evaluation of the formal parameters of the prefixed preterite participles and hence must be supplemented. The first conclusion drawn on the basis of the more detailed specification presented below is that with preterite participles the prefix is less likely to produce gaps in the system. The prefix is attached irrespective of the temporal properties and paradigmatic membership of the verb. The extent to which i- is employed by most of the discerned sub-groups shows that the prefix is indeed used to encode the grammatical information:

The analysis of the strong preterite participles in respect of their ablaut vowel has provided more corroborative evidence for the inflexional status of i- attached to preterite participles. Firstly, the strong preterite participles take the prefix irrespective of the vowel used in the ablaut sequence to mark the form. Secondly, similarly to the list of the strong preterite forms that of the strong preterite participles also has no gaps, which proves the inflexional nature of the prefix in this category. Thirdly, the fact that the prefix is attached to the preterite participles of the verbs belonging to all the three classes of the strong verbs excludes selective prefixation in this sub-group:

The morphological structure of an average prefixed strong preterite participle can be formalised as: {i-}+{[V.sub.1] [right arrow] [V.sub.2]}+{-en}. The gradual loss of the -(e)n marker in the southern dialect is often interpreted as a result of its redundancy, triggered off by the use of the "compensatory" i-. As the data show not all of the prefixed strong preterite participles in the West Midland dialect retain this morphological pattern. They can be sub-divided into three groups, i.e. those which retain the -(e)n marker, fluctuate between retention and loss, and those which always occur without the marker--quatrum non datum:

The above statistics show that the verbs whose preterite participles retain the marker prevail also amongst the fluctuating verbs. Those which have the prefix, but never take -(e)n constitute the minority. Thus, if i- could compensate the loss of the -(e)n marker and therefore accelerate its entire elimination from the system, the expected proportions should be opposite, i.e. the items without the marker should be the most numerous sub-group. This can also mean that the loss of the -(e)n marker had just begun in the language of the AB texts. (14)

Despite the remarkable consistency in prefixing preterite participles there are entities which are exempt from the prefixation. This small group of unprefixed preterite participles above all consists of morphologically complex verbs such as bitacht, fordemed, aturnet, or ouercumen, forloren, bigotten, to name but a few. Such preterite participles never take the prefix due to the presence of an unstressed element preceding the stem. Also the complex preterite participles of anomalous verbs, e.g. fordon, ouergan, or ofgan do not take i-. There are rare instances of unprefixed preterite participles of simplexes, which for some reasons drop the prefix, especially in perfective constructions in the neighbourhood of the auxiliary habbe.

5.5. The semantics of i-

The morphological dualism of the prefix is also evinced by the disintegration of its semantic content. The comparison of the prefixed verbs with their unprefixed counterparts shows that the original meaning of the prefix, whether it is defined as 'mit', zusammen', or 'out', forward', 'forth', had already dissipated by the early thirteenth century and therefore no literal meaning can be attributed to i-. Although the prefix itself has no concrete semantic substance, it can sometimes slightly modify the meaning of the verb to which it is attached, which can be observed solely in some infinitives such as seon 'to see' and iseon 'to perceive', or heren 'to hear' and iheren 'to hearken', 'to listen up'. In other pairs such as finden/ifinden, kepen/ikepen, or leuen/ileuen any difference in the meaning can be hardly felt. As regards the finite forms, the effects of such a semantic modification are also scarcely palpable. The only change sometimes introduced by the presence of the prefix in a finite form is the "synopsis" of the action expressed by the verb, which can be also imposed by the tense of the verb. Furthermore, the prefix has no bearing on the meaning of the preterite participles, which remain semantically unchanged.

That the morphological status of i- was dual is also evident in its opaque semantics, or more precisely in the semantics of the products generated with it. A significant criterion distinguishing derivational morphology from inflexional is that the products of the former tend to be semantically irregular, whereas those of the latter are semantically regular. (15) The data show that some entities produced with the prefix are semantically irregular, especially when found in infinitive or rarely as finite forms. Only the prefixed preterite participles are semantically regular, i.e. preserve the meaning of the basic form. Thus, the semantic opacity of the prefix, observable only in the prefixed infinitives and finite forms, is a typical feature of derivational morphemes, whereas the entire elimination of the semantic substance from the prefix in preterite participles seems to be the "semantic regularization" of the products of the inflexional i-, a process which had begun long before the Middle English period.

5.6. Synchronic models of i-

The comparison of the formal features of the verbs preceded by the ME i- with the formal features of the verbs taking ge- in Modern Dutch and German has revealed that the morphotactic pattern of the former overlaps with that of the latter. The number of the similarities shows that the functional development of the prefix in English at the beginning of the thirteenth century had been at a stage much resembling that of the pre-verbal ge- in the two sister languages.

First of all, the results of the analysis presented hitherto clearly show that the morphological status of the ME i-, like that of the synchronic ge-, must have been dual, i.e. the prefix could serve both as a derivational formative and as a marker of the preterite participle. As a derivational formative the prefix can be therefore enumerated beside such verbal prefixes as the ME for-, a-, on-, be-, and to-. In the function of an inflexion, the prefix encoded the grammatical category of preterite participle. Moreover, the semantic opacity of the prefix is a feature of all the three languages. The inflexional i- occurs solely with morphologically simple verbs and its use is not determined by the paradigmatic membership of the verb, which is another parallelism. The non-determinative nature of inflexion in the use of the prefix with preterite participles is best illustrated by the pair ischaped/ischapen, a verb which exhibits paradigmatic shift, but nonetheless retains the prefix both in the weak and strong form of the preterite participle. Like in Modern Dutch, the stress of the verb does not determine the use of the prefix.

Thus, these results show that the ME i- had developed the same functional features as ge- in the two modern languages. The rules governing the use of the prefix both in derivation and inflexion seem to be based on the same formal criteria. This can be explained in terms of either a "language drift", or as a result of linear development of the PGmc *ga-. The close formal parallelism between the synchronic ge- and the ME i- is an ample proof of the common ancestry of the morphemes.

6. Conclusions

The use of the pre-verbal i- in Early Middle English was not confined to the preterite participle as suggested in many traditional descriptions, but embraces also a large number of infinitives and finite forms. The results obtained from the analysis of the formal parameters of the prefixed verbs corroborate the morphological dualism of the prefix, which could serve both as a marker of the preterite participle and as a derivational morpheme. The examined parameters of the prefixed verbs however do not form any clear-cut patterns, on the basis of which the use of the prefix in particular categories could be regulated. Furthermore, the settings of the parameters of the prefixed infinitives and finite forms indicate that they are products of derivational morphology, whereas those of the prefixed preterite participles are typical of products of inflexional processes. The only formal prerequisite for the use of the prefix both as a derivational formative and as a marker of the preterite participle is the morphological structure of the verb, i.e. the prefix is attached solely to morphologically simple verbs. Moreover, the semantic opacity of the prefix proves that the function of i- extended beyond that of an optional appendage. The formal parallelism between the ME i- and ge- in Modern Dutch and German is evidence of the common ancestry of the prefixes. This can be also interpreted as a result of similar developmental tendencies within the verbal systems of the three languages.

Thus, apart from the morphological structure of the prefixed verb, which always has to be simple, none of the analysed parameters seems to directly determine the use of the prefix in any of the discerned categories and their variability stems form the morphological dualism of the prefix rather than systemic conditionings.

REFERENCES

PRIMARY SOURCES

d'Ardenne, S. R. T. O. (ed.)

1977a The Katherine group. Paris: Societe d'Edition "Les Belles Lettres".

1977b Seinte Margurete, in: d'Ardenne, S. R. T. O. (ed.), 53-94.

1977c Sawles warde, in: d'Ardenne, S. R. T. O. (ed.), 165-185.

1977d Hall Mei[eth]had, in: d'Ardenne, S. R. T. O. (ed.), 127-165.

1996 [THORN]e liflade ant te passiun of Seinte luliene. EETS OS 248. London: Oxford University Press.

d'Ardenne, S. R. T. O.--Eric J. Dobson (eds.)

1981 Seinte Katerine. EETS SS 7. Oxford: Oxford University Press.

Dobson, Erie J. (ed.)

1972 Ancrene Riwle. EETS OS 267. London: Oxford University Press.

Tolkien, John R. R.--Nell R. Ker (eds.)

1962 Anerene Wisse. EETS OS 249. London: Oxford University Press.

SECONDARY SOURCES

Aronoff, Mark

1976 Word formation in generative grammar. (Linguistic Inquiry Monographs 1). Cambridge: The MIT Press.

Baayen, Harald R.

1992 "Quantitative aspects of morphological productivity", in: Geert Booij--Jaap van Made (eds.), 109-149.

Bauer, Laurie

1983 English word-formation. Cambridge: Cambridge University Press.

2001 Morphological productivity. (Cambridge Studies in Linguistics 95). Cambridge: Cambridge University Press.

Benson, Larry D.

1988 The Riverside Chaucer. Oxford: Oxford University Press.

Blake, Norman (ed.)

1992 The Cambridge history of the English language. Vol. 2:1066-1476. Cambridge: Cambridge University Press.

Bloomfield, Morton--Leonard Newmark

1965 A linguistic introduction to the history of English. New York: Alfred A. Knopf.

Booij, Greet--Jaap van Marie (eds.)

1992 Yearbook of morphology 1991. Dordreeht: Kluwer.

Booij, Greet

2001 The morphology of Dutch. Oxford: Oxford University Press.

Brunner, Karl

1965 An outline of Middle English grammar. (Translated by Grahame Johnston). Oxford: Blackwell.

Burnley, David

1992 The history of the English language. London: Longman.

Burrow, John Anthony--Thorlac Turville-Petre

1992 A book of Middle English. Oxford: Blackwell.

Donaldson, Bruce C.

1981 Dutch reference grammar. Gravenhage: Martinus Nijhoff.

Eisenberg, Peter et al.

2005 Duden, die Grammatik: Unentbehrlich fur richtiges Deutsch. (Duden in 12 Banden, Bd. 4). Mannheim: Dudenverlag.

Emerson, Oliver F.

1924 The history of the English language. London: Macmillan.

Higuchi, Masayuki

1998 "The roles of the ME preverbal y-, with special reference to Chaucer's English", Journal of English Linguistics 26/3: 199-208.

Hiltunen, Risto

1983 The decline of the prefixes and the beginnings of the English phrasal verb. Turku: Turun Yliopisto.

Iglesias-Rabade, Luis

2003 Handbook of Middle English: Grammar and texts. Muenchen: LINCOM.

Jordan, Richard

1974 Handbook of Middle English grammar: Phonology. (Translated and revised by Eugene J. Crook). The Hague: Mouton.

Kuhn, Sherman M.--John Reidy (eds.)

1968 Middle English Dictionary. Part L J. Ann Arbor: The University of Michigan Press.

Kurath, Hans--Sherman M. Kuhn--Robert E. Lewis (eds.)

1952-2001 Middle English dictionary (MED). Ann Arbor: Michigan University Press.

Lass, Roger

1992 "Phonology and morphology", in: Norman Blake (ed.), 123-155.

Lester, G. A.

1996 The language of Old and Middle English poetry. London: Macmillan.

Lewis, Robert E.

2005 "Verbs with i- prefix in the Middle English dictionary", in: Akio Oizumi--Jacek Fisiak--John Scahill (eds.), 105-128.

Lindemann, Richard J. W.

1970 Old English ge-: lts meaning. Charlottesville: The University Press of Virginia.

Milroy, James

1992 "Middle English dialectology", in: Norman Blake (ed.), 156-206.

Mincoff, Marco

1972 English historical grammar. Sofia: Naouka i Izkoustvo.

Mosse, Fernand

1952 A handbook of Middle English. (Translated by James A. Walker). Baltimore: The John Hopkins Press.

Mustanoja, Tauno F.

1960 A Middle English syntax. Part One: Parts of speech. Helsinki: Societe Neophilologigue.

Nist, John

1966 A structural history of English. New York: St. Martin's Press.

Oizumi, Akio--Jacek Fisiak--John Scahiil (eds.)

2005 Text and language in medieval English prose. (Studies in English medieval language and literature Vol. 12: A festschrift for Tadao Kubouchi). Frankfurt: Peter Lang.

Pileh, Herbert

1955a "ME i- beim Partieipium Prateriti", Anglia 73/3: 279-291.

1955b "Der Untergang des Praverbs he- im Englisehen", Anglia 73/1: 37-64.

Pinsker, Hans E.

1963 Historische englische Grammatik: Elemente der Laut-, Formen-. und Wortbildungslehre. Mtinchen: Max Neuber.

Reed, David W.

1950 "The history of inflectional 'n' in English verbs before 1500", University of California Publications in English 7/4: 157-328.

Smith, Jeremy J.

1999 Essentials of Early English. London--New York: Routledge.

Stanley, Eric G.

1982 "The prenominal prefix ge- in late Old English and early Middle English", Transactions of the Philological Society 1982: 25-66.

Strang, Barbara M. H.

1970 A history of English. London: Methuen.

Trobevsek-Drobnak, Franciska

1994 "The Old English preverbal ge- in the light of the theory of language change as strengthening or weakening", Studia Anglica Posnaniensia 28: 123-141.

Visser, Frederieus Theodorus

1966 An historical syntax of the English language. Part Two: Syntactic units with one verb. Leiden: E. J. Brill.

Welna, Jerzy

1996 English historical morphology. Warszawa: Wydawnictwo Uniwersytetu Warszawskiego.

Wyld, Henry C.

1927 A short history of English. London: John Murray.

1956 A history of modern colloquial English. Oxford: Basil Blaekwell.

(1) The grammatical environment of the OE ge- was examined by Trobevsek-Drobnak (1994), who argues that the prefix is more likely to appear in grammatically complex structures.

(2) Also Burrow and Thorlac (1992: 33) concur that the prefix was an attribute of the preterite participle. For Nist (1966: 184) the prefix is "an indicator of the past participle". Cf. also Burnley (1992: 67).

(3) The account of the prefix in Old English given by Emerson (1924: [section]437) presents ge- as a characteristic of the preterite participle of uncompounded verbs with the exception of some strong verbs, which may suggest the opposite, i.e. that in Middle English the form which should manifest higher retention of the prefix is the weak preterite participle.

(4) Cf. also Mincoff (1972: 284, 285).

(5) According to Hiltunen (1983: 62, 63) the common use of ge- in Old English with preterite forms may have been one of the causes leading to the subsequent grammaticalisation of the prefix. Cf. also Wyld (1927: 267) and Strang (1970: [section]155).

(6) According to the scholar i- [right arrow], [THETA]/-- V# i- -- due to contraction with the preceding vowel of similar quality.

(7) Some traces of the prefix can be found in south-west dialects of English, where it has been retained as a-/[??]/, usually attached to preterite participles, e.g. ataken, asaid (cf. Pinsker 1963: [section][section] 171, 186 n. 1), and in few words of Standard English such as afford (< OE 3efor[eth]ian), where it is almost unrecognisable.

(8) In his monograph on the meaning of the OE ge- Lindemann (1970: 22, 25, 63) questions the etymological connections of the prefix with the Lat. com-, and therefore with the IE *co-. As the ultimate source of the prefix in Germanic dialects, Lindemann (1970: 36) points at the deictic IE *gho-, and thus according to his view the IE *gho- > PGmc *ga- > OE ge-.

(9) Cf. also the examples in Aronoff (1976: 98).

(10) Cf. Lass (1992: 147); Lester (1996:115); Smith (1999:117).

(11) On the uniformity of the AB language see Milroy (1992:158).

(12) In statistics this marker is also termed "dominant" and pertains to the value (here the number of tokens of a particular form taking the prefix) occurring most frequently in a sample or population, often defined as a positional parameter informing about distribution of probability.

(13) Also the deverbal nominalisation icorene, 'the chosen', 'the elected' consistently appears with the prefix in all the analysed texts.

(14) In Reed (1950: 278) the beginning of the loss of the inflexional -(e)n in various verbal categories in both the Southern and Central dialects has been dated at the years 1000-1150. As to the strong preterite participle he claims that in the North-West Midlands the process was not completed by about 1500, which is evinced by rare instances of strong preterite participles without (e)n occurring after 1400 in the North Midlands.

(15) Cf. Bauer (1983: 28)

TOMASZ MOKROWIECKI

Adam Mickiewicz University, Poznah
Table 1. The type/token ratio and hapax frequency in the
prefixed forms

Form tykes tokens type/token ratio

inf. 20 154 0.12
pres. ind. 26 73 0.36
pres. subj. 10 20 0.5
pret. ind. 22 58 0.38
imp. 4 10 0.4
pret. part. 367 1805 0.20
pret.-pres. verbs 3 14 0.21
anom. pret. prat. 4 88 0.04
Total/average 456 2222 0.20

Table 2. The prefixed verbs in the present indicative

 Present indicative

 Singular

marker 1st 2nd 3rd

types 4 2 11
tokens 12 2 40
type/token ratio 0.33 1 0.28
% of i-forms 16.43% 2.73% 54.80%
% of prefixation 3.22% 0.49% 0.93%

 Plural

marker 1st 2nd 3rd

types 3 1 5
tokens 7 3 9
type/token ratio 0.42 0.33 0.56
% of i-forms 9.59% 4.10% 12.32%
% of prefixation 4.90% 1.42% 1.13%

Table 3. The prefixed verbs in the present subjunctive

 Present subjunctive

 Singular

marker 1st 2nd 3rd

types 1 1 6
tokens 2 1 15
type/token ratio 0.5 1 0.4
% of the i-forms 10% 5% 75%
% of prefixation 6.25% 0.70% 1.73%

 Plural

marker 1st 2nd 3rd

types -- -- 2
tokens -- -- 2
type/token ratio -- -- 1
% of the i-forms -- -- 10%
% of prefixation -- -- 1.50%

Table 4. The prefixed weak verbs in the preterite indicative

 Preterite indicative

 Singular

marker 1st 2nd 3rd

types 3 -- 6
tokens 4 -- 17
type/token ratio 0.75 -- 0.36
% of i-forms 16.67% -- 70.83%
% of prefixation 2.47% -- 1.46%

 Plural

marker 1st 2nd 3rd

types -- -- 3
tokens -- -- 3
type/token ratio -- -- 1
% of i-forms -- -- 12.50%
% of prefixation -- -- 1.75%

Table 5. The prefixed strong verbs in the preterite indicative

 Preterite indicative

 Singular

marker 1st 2nd 3rd

types 2 -- 6
tokens 13 -- 19
type/token ratio 0.16 -- 0.31
% of i-forms 38.23% -- 55.89%
% of prefixation 23.21% -- 2.13%

 Plural

marker 1st 2nd 3rd

types 1 -- 1
tokens 1 -- 1
type/token ratio 1 -- 1
% of i-forms 2.94% -- 2.94%
% of prefixation 20% -- 0.62%

Table 6. The number of ablaut alternants in the prefixed strong
preterite forms

Number of ablaut alternants Tykes Tokens

Class I Bi-alternant steins

/[??]:w/ ~ /e:w/ ~ /e:w/ ~ /[??]:w/ 1 1

Class 11 Tri-alternant stems

/i/ ~ /a/ ~ /u/ ~ /u/ 1 2
/i:/ ~ /[??]:/ ~ /u:/ ~ /u:/ 3 10
/[epsilon]:/ ~ /a/ ~ /ae:/ ~ /[epsilon]:/ 1 1

Class III Stems with four vowel alternants

/o:/ ~ /a ~ aw/ ~ /[epsilon]:w/ ~ /e:w/ 2 18
/u/ ~ /a/ ~ /u/ ~ /o/ 2 2

Table 7. The gaps and the degree of prefixation in
particular paradigms of the finites

 Singular

Paradigm 1st 2nd 3rd

pres. ind. 3.22% 0.49% 0.93%
pres. subj. 6.25% 0.70% 1.73%
wk. pret. ind. 2.47% -- 1.46%
str. pret. ind. 23.21% -- 2.13%
imp. 1.19%

 Plural

Paradigm 1st 2nd 3rd
V
pres. ind. 4.90% 1.42% 1.13%
pres. subj. -- -- 1.50%
wk. pret. ind. -- -- 1.75%
str. pret. ind. 20% -- 0.62%
imp. 0.32%

Table 8. The arithmetical evaluation of the prefixed
preterite participles

 Preterite participles

marker wk. str. pret.-pres. anom.

types 295 72 3 4
tokens 1345 460 10 88
type/token ratio 0.21 0.16 0.30 0.04
% of prefixation 79.53% 63.89% 0.70% 88%

Table 9. The number of ablaut alternants in the prefixed
strong preterite participles

Number of ablaut alternants Types Tokens

Class I Bi-alternant stems

/[epsilon]:/ ~ /e:/ ~ /e:/ ~ /[epsilon]:/ 3 17
/ou/ ~ /eu/ ~ /eu/ ~ /ou/ 1 1
/au/ ~ /ou/ ~ /ou/ ~ /au/ 4 19
/u/ ~ /o:/ ~ /o:/ ~ /u/ 1 29
/[??]:/ ~ /e:/ ~ /e:/ ~ /[??]:/ 3 31
/a/ ~ /e:/ ~ /e:/ ~ /a/ 2 15
/a:/ ~ /o:/ ~ /o:/ ~ /a:/ 4 10
/ou/ ~ /eu/ ~ /ou/ ~ /ou/ 1 8
/i:[??]/ ~ /[??]:/ ~ /i:[??]/ ~ /i:[??]/ 1 1
/[epsilon]:/ ~ /a/ ~ /[epsilon]:/ ~ /[epsilon]:/ 1 16
/u/ ~ /a/ ~ /u/ ~ /u/ l 2

Class II Tri-alternant stems

/i/ ~ /a/ ~~ /[epsilon]:/ ~ /[epsilon]:/ 3 14
/e/ ~ /a/ ~ /o/ ~ /o/ 10 43
/[epsilon]:/ ~ /a/ ~ /a/ ~ /[??]:/ 2 43
/ei/ ~ /ai/ ~ /oi/ ~ /oi/ 1 1
/i:/ ~ /[??]:/ ~ /u:/ ~ /u:/ 4 34
/i:/ ~ /[??]:/ ~ /[??]:/ ~ /u:/ 1 6
/e:/ ~ /[epsilon]:/ ~ /[??]:/ ~ /[??]:/ 4 15
/i:/ ~ /[??]:/ ~ /[??]:/ ~ /i/ 2 9
/[epsilon]:/ ~ /a/ ~ /[epsilon]:/ ~ /[??]:/ 2 4
/o/ ~ /a/ ~ /u/ ~ /o/ 1 1
/[epsilon]:/ ~ /o:/ ~ /o:/ ~ /[??]:/ 2 5
/i:[??]/ ~ /ai/ ~ /ei/ ~ /ei/ 1 4
/u:/ ~ /[epsilon]:/ ~ /[??]:/ ~ /[??]:/ 1 2
/i/ ~ /a/ ~ /u/ ~ /u/ 3 5
/i/ ~ /o:/ ~ /o:/ ~ /u/ 1 8
/[epsilon]:/ ~ /ou/ ~ /ou/ ~ /ai/ 1 20
/e/ ~ /o:/ ~ /o:/ ~ /[epsilon]:/ 1 1
/i:[??]/ ~ /ei/ ~ /ou/ ~ /ou/ 1 3
/i:/ ~ /[??]:/ ~ /i/ ~ /i/ 1 44
/i/ ~ /a/ ~ /a/ ~ /u/ 1 3

Class III Stems with four vowel alternants

/i:/ ~ /ei/ ~ /u:[??]/ ~ /ou/ 1 2
/e:/ ~ /a/ ~ /u:/ ~ /o:/ 1 4
/e/ ~ /a/ ~ /[epsilon]:/ ~ /i/ 1 11
/[epsilon]:/ ~ /a/ ... /ae/ ~ /[??]:/ 1 3
/e:/ ~ /au/ ~ /ou/ ~ /eu/ 1 20
/e:/ ~ /ei/ ~ /u:[??]/ ~ /ou/ 2 6

Table 10. The -(e)n marker in the prefixed strong
preterite participles

marker {-(e)n) {-(e)n}/{-O}] {-O}

types 47 18 7
tokens 221 192/37 10

Figure 1. Distribution of the prefix among the i- forms

inf. 6.93%
pres. ind. 3.29%
pres. subj. 0.90%
pret. ind. 2.61%
imp. 0.46%
pret. part. 81.23%
pret.-pres, verbs 0.63%
anom. pret. part. 3.97%

Note: table made from bar graph.

Figure 2. The degree of prefixation of particular categories
in the corpus

inf. 3,07%
pres. ind. 1.17%
pres. subj. 1.43%
pret. ind. 2.10%
imp. 0.67%
pret. part. 74,87%
pret.-pres, verbs 0.70%
anom. pret. part. 88%

Note: table made from bar graph.

Figure 3. The prefixed finite forms according to the mood

pres. Ind. 45.0%
pres. Subj. 12.34%
pret. Ind. 35.80%
imp. 6.18%

Note: table made from bar graph.

Figure 4. The percentage of particular sub-groups in the total
number of the prefixed preterite participles

wk. pret. Part. 70.68%
str. Pret. Part. 24.18%
pret.pres.pret. part. 0.52%
anom. Pret. Part. 4.62%

Note: table made from bar graph.
COPYRIGHT 2007 Adam Mickiewicz University
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2007 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:LINGUISTICS
Author:Mokrowiecki, Tomasz
Publication:Studia Anglica Posnaniensia: international review of English Studies
Article Type:Report
Date:Jan 1, 2007
Words:9550
Previous Article:Possible origins of different usages in present-day spoken and written English.
Next Article:The Scandinavian element beyond the Danelaw.
Topics:


Related Articles
On the histories of de-verbal adjectives in Middle English.
Negative-initial sentences in Old and Middle English.
Recycling inversion: the case of initial adverbs and negators in Early Modern English.
Binary correlations of Middle English one-root deverbal coinages in the OED textual prototypes.
Let and the "bare infinitive": an exploratory exercise in traditional (notional) grammar.
Nim or take? A competition between two high frequency verbs in Middle English.
Is reanimation of voices possible? Pragmatics of reported speech in selected Middle English texts.
The analysis of the semantics of Middle English mowen in the light of its rise to epistemicity and eventuality.
Chronological homogeneity in Middle English deverbal word-formation: the evidence from the Oxford English dictionary.
Possible origins of different usages in present-day spoken and written English.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters