Thou and ye: a collocational-phraseological approach to pronoun change in Chaucer's Canterbury Tales.
Chaucer's use of the singular or plural form of the second person pronoun to address a single person in his Canterbury Tales usually follows the established standards of his time. However, some ninety instances of pronoun switching do occur, and explanations drawing on pragmatic parameters, rhyme and textual corruption have not been able to explain all of these deviations. Complementary to these approaches, this paper offers a novel explanatory hypothesis. The "collocational-phraseological hypothesis" suggested here takes into account the force of the syntagmatic relationship of words. On the basis of an original electronic compilation of all instances of pronoun switches in the Canterbury Tales and a classification according to three main types, we argue that frequently and/or habitually used lexical combinations (collocations, formulae, quotations) can account for a significant number of the cases in question.
Towards the end of the fourteenth century, the use of the second person plural pronoun to address a single person was generally determined by sociopragmatic conditions such as status and social distance (cf. Finkenstaedt 1963: 73-74; Burnley 2003). Chaucer's literary use of the personal pronouns thou and ye (henceforth: T and Y; including inflected forms) usually accords well with the practices of the time. Skeat's (1894: 175) assessment of that general practice has basically remained unchallenged: Y was reserved for the address of a servant to the lord, for compliment, to express honour, submission or entreaty. However, it has also been noted for a long time that sudden changes of the pronouns in the conversation of the same pair of speaker and listener do indeed occur in Chaucer's works. The Canterbury Tales, on which this paper concentrates, show a considerable number of such cases. (1) The following example from a speech by the yeoman to the summoner in Friar's Tale (1397-1402) is an illustration of the change in question. It shows a single deviation from the singular (thyn etc.) to the plural (vow) in verse (1399): (2)
1) I am unknowen as in this contree; Of thyn aqueyntance I wolde praye thee, And eek of bretherhede, if that yow leste. I have gold and silver in my cheste; If that thee happe to comen in oure shire, Al shal be thyn, right as thou wolt desire. [I am not known in this country; I wish to ask you of your acquaintance and also of sworn brotherhood, if you wish. I have gold and silver in my chest; If you happen to come to our shire, all shall be yours, just as you wish.]
Striking deviations from the norm such as this one seem at first sight to be rather irregular and arbitrary. What complicates the picture is that next to single deviations such as in (1), complete changes of the paradigm or utterly irregular switchings in both directions can be found. Scholarship has so far mainly concentrated on affective-situational or other pragmatic explanations to account for all these changes. In particular, developments in the attitude of the speaker towards the addressee in the course of the text have repeatedly been claimed as the motivating factor for the changes. Demands of rhyme or textual corruption can be held responsible in a few cases; and from time to time reference to "formulaic phrases" has been made. Section 2 of our paper is a brief review of earlier approaches which shows that these explanations fall short of accounting for all of the instances of pronoun change.
We suggest that the syntagmatic relationship of words had an impact on the choice of the pronouns of address. In particular, we argue that frequently and/or habitually used lexical combinations could influence the choice regardless of micro- or macropragmatic considerations. It will be shown that this line of investigation, which we call the "collocational-phraseological hypothesis" and which will be more fully introduced in section 3, can usefully complement pragmatic explanations. Our investigation rests on a consideration of all changes of pronoun of address in Chaucer's Canterbury Tales, which are systematicized and catalogued in section 4. (3) Cases where the collocational-phraseological force was paramount in the choice of the pronoun of address are discussed in section 5. In the concluding part we give a general account of the explanatory power of our hypothesis for the cases studied and discuss some implications of our results.
2. Earlier approaches to pronoun change in Chaucer
The striking switching between formal and informal pronouns of address for a single person has become something of a brain-teaser for linguistic Chaucer research. Of course, scholars working on this topic assume that there is indeed a proper solution to the problem and that Chaucer did not alternate between the forms at random. Agreeing with Blake who states that "we cannot be certain that all these switches are significant" (1992: 539) at all would mean to admit to the inadequacy of the explanatory power of linguistics. More than a dozen studies on the pronoun changes in the Canterbury Tales show that the question is not deemed to be settled. In the recent past some extensive studies have been presented which put forward new and interesting suggestions. (4) The following explanation patterns have been brought forward.
First, a rather technical matter which is pointed out in Nathan (1956: 41-42) must not be ignored (cf. also Koziol 1943: 174; Kerkhof 1982: 228). Our text, that is the electronic Robinson (1957) and the printed Benson (1988) version, may contain scribal errors. Nathan has suggested this explanation for example (1), where Benson's reading draws on Ms. Hengwrt, while other manuscripts--as well as the second edition of the Caxton print, which is believed to be based on a good manuscript--stick to T. Some of the supposed changes may in fact be due to scribal variation; these are questions for textual critics. But Nathan's plea seems to be inspired by the attempt to vindicate Chaucer and to discuss away supposed "errors" (1956: 42: "a slip of Chaucer's pen"; but see Jucker--Taavitsainen 2003: 12; Jucker 2006: 58).
In other cases, a less controversial explanation presents itself. Where a pronoun stands at the end of a line the demands of rhyme can account for a pronoun change (cf. Shimonomoto 2001: 44-45; Karpf 1930: 43, 45). In cases such as Nun's Priest's Tale (3351) metrical demands seem to be a fairly plausible explanation for the change (3350-3352):
2) Why ne hadde I now thy sentence and thy loore, The Friday for to chide, as diden ye? For on a Friday, soothly, slayn was he. [Why did I then not have your judgement and your learning to scold the Friday, as you did? For on a Friday, it is true, he was murdered.]
Pragmatic explanations have been the focus of attention (cf. Burnley 1983: 1922; Honegger 2003; Mazzon 2000; Shimonomoto 2001). Adopting from Hope (1993) the distinction between macro- and micropragmatics, we can describe as macropragmatically based those changes where the social relationship between speaker and addressee changes in the course of the text (comparable to the power semantics in Brown--Gilman's seminal article). (5) This may account for the conversion to Y in Wife of Bath's Tale (1088), where the olde wyf addresses her husband after their marriage (cf. Shimonomoto 2001: 36; Kerkhof 1982: 229). The micropragmatic explanation, though, is more relevant. Here the immediate linguistic and non-linguistic factors of a conversation are concerned. Friar's Tale (1584-1623), where a summoner pesters an old widow to give him money, is a good example. She uses the formal Y-pronoun towards him. As the summoner becomes more and more obtrusive and confronts her with outrageous and completely made-up accusations (1616), the widow loses her patience in (1618) (cf. Nathan 1956: 40; Shimonomoto 2001:11-12; Jucker 2006: 68):
3) "Thou lixt!" quod she, "by my savacioun ... ["You are lying!" she said, "by my salvation ...]
These changes have alternatively been viewed as affective, emotional, emotive, expressive or situational (cf. Becker 2003: 163), and also interactional (cf. Jucker 2006). Virtually all scholars refer to the (micro-)pragmatic level in trying to make changes of the emotional attitude between speaker and addressee plausible.
Given the social dimension of address there is no doubt as to the applicability of pragmatics in dealing with pronoun changes. The explanations given so far are nonetheless not always satisfactory. Pragmatists have not been able to explain all changes convincingly, as Mazzon concedes: "of course, it is difficult to justify all switches pragmatically in any precise way, especially given the distance in time and Weltanschauung that separates us from this work" (2000: 139-140). Shimonomoto concludes her chapter on pronoun changes with the observation that "[w]e should admit there are still a few uncertainties concerning Chaucer's use of ye and thou" (2001: 45). Where no pragmatic explanation is evident, sometimes complex inferences about the socio-cultural background and the speakers' intentions are made. Honegger (2003), for instance, shows that sometimes many factors have to be taken into consideration in order to suggest an explanation. This is particularly evident in his reconstruction of the reasons for Palamon's "confuse" switching in his address to Venus (cf. Honegger 2003: 68-69, 75-78). However, for the reasons mentioned before we have to be aware at all times of Finkenstaedt's warning of a "Beugung des Textes oder uberspitzt raffinierte[n] Deutungen" [distortion of the text or exaggeratedly sophisticated interpretations] (1963: 75; cf. also Bumley 2003:31).
Lastly, sporadic reference to "formulaic phrases" and the like can be found as early as in Kennedy's 1915 study (1915: 84) on the thirteenth century, and then also in Koziol (1943: 172), Harley (1988: 5), Miller (1992: 152), Shimonomoto (2000: 114; 2001: 13), and recently most explicitly in Mazzon (2000: 155, 158) and Bumley (2003: 30-31, 33). Systematic investigation of this aspect has, however, not yet been attempted.
3. The collocational-phraseological hypothesis
Our collocational-phraseological hypothesis claims that certain lexical cotexts may trigger the choice of the pronoun. This hypothesis is supported by the observation that even in a language with a rigid T/Y-system such as Modern German, particular cotexts in fixed or partly fixed expressions may prefer or even demand one of the pronouns. Examples are: Wie du mir, so ich dir (saying, T only), Du ahnst es nicht! (general T), Du sollst nicht toten (maxim, general T), das kannst du dir abschminken (T dominant), wenn Sic gestatten (Y dominant). (6) The appropriate use of the last two examples in Modern German, however, depends on the register, which is in turn based on extra-linguistic factors such as the power relationship or the degree of familiarity between the speakers. We suggest that in a language with a non fully rigid T/Y-system, such as the Middle English of the later 14th century, lexical co-selection constraints could have exercised a significant impact on the pronoun choice. At least in literary texts such as the Canterbury Tales, the effect of this purely syntagmatic force could be restricted to the brief insertion of a deviant pronoun of address, but it could also lead to a general conversion of the paradigm in the conversation between the same pair of speaker and addressee.
In our attempt to pin down what exactly characterizes such combinations of T/Y-pronouns with particular cotexts and how they can be studied, we take recourse to two complex and much-debated lines of scholarship with different, yet interrelated approaches to preferred word combinations. One is the research on collocations in the tradition of British contextualism, the other, which has an equally long history, is phraseology. (7) Needless to say, we do not attempt to discuss a general synthesis of the two approaches with their varied definitions and terminology for our practical purpose. But it does seem feasible to point out some of the postulated core features of collocations and phraseological units in order to explain the effect of the syntagmatic lexical force as it is suggested here.
In his basic definition, Sinclair characterized collocation as "the occurrence of two or more words within a short space of each other in a text" (1991: 170). Typically, these co-occurrences are frequently repeated or statistically relevant in a corpus. (8) In this sense, they may be understood as lexical items habitually used together or in close proximity to one another. In our context, a direct syntactic relationship of the members is given. Collocations with a pronoun are not easily situated in the models that have been suggested for Modern English. The cases under consideration here may best be characterized by the frequent association of a node (e.g. the Middle English word for to tell) with a member of a certain category, that is, one of the two second person pronouns (you). In this description collocations such as tell and you may be compared, for instance, to the lexicogrammatically constrained collocation of the adjective rancid with butter or fat rather than, say, bread (on this case, cf., e.g., Moon 1998: 27), or stamps and collect rather than gather. The difference is, of course, that the choice is limited because the collocate is a pronoun, and as such a member of a closed class. (9) We suggest that the frequent coupling of the T-pronoun with terms for body parts, for instance, leads to a collocational force and the preferred selection of T by the noun, or the frequent co-occurrence of a form of mowen and Y may render this connection preferable regardless of semantic and pragmatic considerations. (10)
One constituent of a collocation may possess a semantic feature different from those actualized in all or most other contexts (such as to jog in to jog one's/somebody's memory 'remind somebody of/about something'). (11) In many studies of collocation, especially in lexicographically oriented ones, this restriction is an obligatory feature. (12) In the terminology of some phraseological studies like for instance those by Glaser (1986, 1998) and Cowie--Mackin--McCaig (1983 : xiii), these units are called "restricted collocations", (13) a sub-type of phraseological units. Phraseological units are here understood as two or more lexical items in syntactic relation with one another. They are lexically and syntactically relatively stable semantic or pragmatic units which are habitually reproduced rather than produced. In a structural perspective, they may allow lexical variation (e.g. clear as crystal/day) and syntactic transformations. Depending on their degree of idiomaticity, which is located along a cline, they may show semantic anomalies (e.g. to kick the bucket 'to die'), lexical (e.g. to peter out 'to dwindle to nothing') or syntactic ones (e.g. to trip the light fantastic). (14) All of these phraseological units are covered in some collocation studies such as Bartsch's (2004). (15) Frequently used formulae such as Happy birthday/may also be seen as belonging to this area of overlap between collocation and phraseology.
However, phraseological research differs from collocation studies as outlined above in at least two respects. Firstly, frequency is no defining criterion of phraseological units. (16) Secondly, phraseological units may be pragmatically non-compositional: They may be as long as whole sentences, such as the proverb One swallow does not make a summer. Therefore, in addition to restricted collocations and other phraseological units which may be employed as equivalents of single words in a sentence, units of sentence length can also be gathered under the cover term phraseology. (17) The main classes in G1aser's categorization (1986: chs. 5 and 6) are proverbs and fragments of proverbs (e.g. a new broom [sweeps clean]), commonplaces (e.g. We live and learn), quotations and allusions to well-known texts ("winged words" such as The time is out of joint, from Hamlet), maxims (e.g. Do it yourself), commands (e.g. Thou shalt not kill), slogans (e.g. Safety first), and habitually used routine formulae (e.g. How do you do?). Many of our examples belong to this large subclass of phraseological units. (18)
For our study of preferred and fixed lexical co-occurrences with pronouns of address, a combination of the partly overlapping and mutually complementing concepts of collocation and phraseology as briefly outlined above is suggested. In the absence of a more elegant term, we refer to it as "collocational-phraseological". What is important for our approach is that both collocations and phraseological units possess a high degree of syntagmatic force which leads to the co-selection of lexical items.
4. Method of investigation and an inventory of pronoun changes in the Canterbury Tales
The first step in our study was a frequency count in the Canterbury Tales on the basis of recent corpus-linguistic technology. We searched the electronic version of the text provided by the Corpus of Middle English verse and prose for the collocates of the pronouns of address using a concordancer. Oizumi's lexical concordance (2003), which lists all orthographic variants, was indispensable in doing so. Only explicit pronouns were counted, inclusive of 150 clitics. Pronoun changes which are implied in a change of verbal endings (cf. Elliott 1974: 383; Karpf 1930: 45) were not taken into consideration, such as the use of beth by Chaucer the Pilgrim to the Host in the Prologue to Sir Thopas (707) (and T in 926).
In order to verify the collocational-phraseological hypothesis, a complete list of all pronoun changes in the Canterbury Tales was needed. Identifying these changes is not as straightforward as it might appear. The main problems and our solutions will now be briefly considered.
The first problem is met in passages in which it is uncertain who the actual interlocutors are (who addresses whom?; cf. Burnley 2003: 30), given that tacit switchings to plural reference are possible. An example which may illustrate this problem is the change to Y in the Prologue to the Wife of Bath's Tale (188), where she addresses the Pardoner. Does this plural form already refer to al this compaignye addressed in (189), and is this the reason why this example has never been mentioned in earlier studies? However, the Wife of Bath definitely refers to the previous utterance by the Pardoner at this place ("sith it may yow like"), and there is a contrast to (189) ("But yet I praye to al this compaignye"). (19) Therefore (188) is included in our list.
Another critical point is to determine a "basic form" of address between two interlocutors in some passages. Sometimes it is indeed not possible to establish such a normal, basic form, and each change has to be counted individually (cf. Type 3, below).
Furthermore, an individual (directed) and a universal (non-directed) use of the pronoun has to be distinguished. In this context, the pronoun use in Prudence's address of her husband in the Tale of Melibee is remarkable, and it is indeed interesting to see how freely she switches back and forth, sometimes within one sentence (1175, 1212, 1220; cf. in more detail Mazzon 2000: 155160; also Pakkala-Weckstrom 2004: 164). These are cases of a switching between an individual Y-form and a universal T-form, or also T-forms reminiscent of universal T in the sermon-like tone of the passages. (20) Some of the eleven instances of switching in Prudence's address of Melibee may be considered as triggered by formulae structuring the text, in particular thou shalt (1175, 1210, 1220, or shaltou 1212; allusion to maxims) (21) and "whan ye han examyned youre conseil" (1211, 1222). As no clear case for the collocationalphraseological argument can be made in this particular communication situation, and the universal T is a distinct type of pronoun, we decided to not take these switches into account. Further examples of this transition from individual to universal pronouns (and vice versa) abound in the whole work, and some occur at a point of pronoun change. (22) Still, we do not regard them as pronoun changes in our sense and thus do not include them in our list.
In order to be able to do justice to these problems, we distinguish between three types of changes which we call "insertion", "conversion" and "switching". In the case of an insertion, there is no change of paradigm of second person pronouns. The term conversion refers to cases which indicate such a change. In switching, no basic form is recognizable according to the micro- and macropragmatic analysis; the changes are irregular and insertion may be combined with conversion.
As versified language is denser than prose, each instance of pronoun change was allowed to entend over maximally a rhyming couplet or two consecutive lines. This convention has two entailments. For one thing, an "insertion" may be as long as two lines, and "conversion" refers to changes of more than two lines in extension. Secondly, "regressive" changes were admitted in three cases. Here, the collocational-phraseological trigger appears slightly later than the first change, either in the same line (Miller's Tale 3287), in a rhyming couplet (Knight's Tale 2249-2250) or in two consecutive lines (Canon's Yeoman's Tale 1119-1120), respectively.
Type 1: insertion, i.e. deviation of a single line or rhyming couplet, or, respectively, two consecutive lines
Type 1A: single insertion, e.g. T-T-T-Y-T-T-T (1 change)
Type 1B: multiple insertion, a block (passage longer than two lines) at the end is possible, e.g. T-T-T-Y-T-T-T-Y-T-T-T ... (-Y-Y-Y) (n x 1 changes)
Type 2: conversion, i.e. general change of the paradigm
Type 2A: single conversion, e.g. T-T-T-Y(-Y-Y) (1 change)
Type 2B: double conversion, e.g. T-T-T-Y-Y-Y-T-(T-T) (2 changes)
Type 3: switching, i.e. irregular alternations, e.g. T-T-T-T-Y-Y-Y-Y-T-T-T-T-Y-Y-Y-Y-T-T-T-T ... (n changes)
Table 1 is a list of all initial points of pronoun changes between T and Y which we identified in the Canterbury Tales. In the last column, the type of change is indicated. Line numbers printed in italics indicate places where we suggest a collocational-phraseological trigger. These cases will be discussed in the following section.
5. Discussion of the data
As postulated in section 3, above, instances of pronoun change must fulfill the criteria of frequency of co-occurrence and/or semantic/pragmatic unity in order to be identified as influenced by the collocational-phraseological force in language. The 31 passages in the Canterbury Tales which meet these criteria fall into three groups. The first is marked by frequency (collocations) but not by semantic/pragmatic unity. The second group in contrast is characterized by fulfilling the requirements of phraseological units (formulae, sterotyped comparison and [fragments of] quotations) but not by frequency. And a third one (routine formulae / collocations) fulfills both criteria of frequency of co-occurrence and semantic unity. This will be discussed first. (23)
5.1. Routine formulae / collocations
The routine formula by youre leve 'with your consent, if you please' occurs ten times and exclusively with Y in the Canterbury Tales. The same is true of the formula of/for youre curteisye 'kindly, if you please' (3 instances). (24) The change from T to Y in the Wife of Bath's address to her husband in the Prologue to the Wife of Bath's Tale (331), and in the speech of the fictitious friar to Thomas in Summoner's Tale (2112) (in rhyme position) are thus explainable. As noted in section 4, above, Alison's change to Y in her conversation with Nicholas in Miller's Tale (3287) can be explained by a regressive effect of the formula for youre curteisye.
In the formula of request I (we)prey thee/yow 'I beseech you' (in parenthetical phrases expressing deference, earnestness, etc.), Y once more outnumbers T (62 : 14). The formula is not as fixed as by your leve, for instance, because the verb is movable and an auxiliary can be inserted. (25) The change in Absolon's address to Alison is particularly remarkable (Miller's Tale 3361-3362); to Karpf (1930: 44) it seems incomprehensible.
4) "Now, deere lady, if thy wille be, I praye yow that ye wole rewe on me ..." ["Now, dear lady, if it be your will, I beseech you to have mercy on me..."]
Nathan (1959: 195) and Koziol (1943: 172) rather consider the T-form in (3361) as deviant, but this argument is weakened by the fact that Absolon uses T in (3726) and (3794-ff.) as well.
Another good example of the collocational-phraseological force of this formula is the prioress's invocation of the Mother of God in the Prologue to the Prioress's Tale (467-487) (cf. Shimonomoto 2001: 43; Kerkhof 1982: 231). Mazzon (2000: 138) sees the co-occurrence with nominal terms of address as crucial at the point of transition--however, the address O blisful Queene is a few lines away (481), it is followed by T, and at the place in question there is no such term (481-487):
5) My konnyng is so wayk, O blisful Queene, For to declare thy grete worthynesse That I ne may the weighte nat susteene; But as a child of twelf month oold, or lesse, That kan unnethes any word expresse, Right so fare I, and therfore I yow preye, Gydeth my song that I shal of yow seye. [My skill is so weak, o blissful queen, in declaring your great excellence that I cannot sustain the weight but I behave just like a child of twelve months, or less, that can hardly utter any word, and therefore I beseech you, guide my song that I will sing of you.]
In Knight's Tale (2255) (Palamon to Venus: "Thanne preye I thee") T marks a return after one of the four insertions which we identified. (26) The decisive factor for the insertion of Y in (2254) is arguably the respectful nominal address term my lady sweete. (27)
Shipman's Tale (395) (merchant to his wife) is a counter-example ("I prey thee, wyf, ne do namoore so"). Here and in two further cases of this kind (cf. if thee lest, below), the pragmatic power of the line is particularly intensive--and it might be argued that Chaucer exploited what may be called a "counter-collocational-phraseological" force to heighten this intensity even further.
God thee/yow + verb in the subjunctive is a suitable frame for wish and request formulae. They occur more than twice as often with Y (23) than with T (11). This would explain the change of address of the Host to the Monk in the Prologue to the Nun's Priest's Tale (2788). (28)
In the routine formula (if/as etc. it) thee/yow like(th) 'if you choose, if you wish' the verb liken predominantly co-occurs with the Y pronoun in the object case (19 Y : 5 T). This can explain the change in the Prologue to the Wife of Bath's Tale (188) discussed in section 4, above. (29)
Almost synonymous with the phrase if that yow liketh is the more frequent (if/as etc.) thee/yow list, where the predominance of the Y pronoun is equally clear (47 Y : 9 T). (30) This formula with predominant Y can account for the changes in Friar's Tale (1399), where the yeoman addresses the summoner in example (1), above (cf. Harley 1988: 5; Nathan 1956: 41-42; Shimonomoto 2001: 12-13), and in Aurelius's invocation of Phebus (Franklin's Tale 1041), for which Shimonomoto sees "no obvious reason" (2001: 44).
In Knight's Tale (2249-2250) (Palamon to Venus), we suggest a regressive change (cf. section 4, above):
6) Youre vertu is so greet in hevene above That if yow list, I shal wel have my love. [Your power is so great in the high heavens that if you wish I will easily possess my love.]
Two counter-examples are the pronoun changes in Summoner's Tale (1985) (friar to Thomas "if thee leste"; cf. Mazzon 2000: 161-163) and in Canon's Yeoman's Tale (1360) (canon to priest "if that thee list it have"; cf. Nathan 1959: 197), but they might have been employed for particular pragmatic purposes, as suggested above.
(I wol/shal etc.) telle(n) thee/yow is used with diminished force or in emphatic expressions and occurs preferably with the Y pronoun (67 Y : 15 T). Obviously, this is due to the fact that I wol telle yow is a frequent metalinguistic formula used by the pilgrims addressing the fellow travellers, that is, an illocutionary formula. However, this does not spoil the collocational argumentation insofar as in establishing collocation as used here, it is irrelevant whether a particular pronoun is used to refer to one or more than one person. The changes in Wife of Bath's Tale (1012) (old wife to knight) and Merchant's Tale (2169) (Januarie to May) fit in well with our explanation. (31) In addition, the French equivalent je vous dy in Summoner's Tale (1832) (Friar to Thomas) also triggers a change. A similar proportion can be discovered with the semantically related say (to) thee/yow (29 Y : 6 T), which is not relevant for any of the changes, though.
5.2. Further collocations
There seems to be a preference for body part terms to collocate with T. (32) This lexical preference may indeed account for some of the identified pronoun changes. To choose an example, eyen co-occurs twice as often with thy (13) as with youre (6). Januarie's move to T towards May in Merchant's Tale (2138-2148), taken from the Song of Songs, is a case in point (cf. Shimonomoto 2001: 30; Mazzon 2000: 151-152). Other changes involving the collocation with a body part term are Friar's Tale (1623), Franklin's Tale (1635), Summoner's Tale (2131), Clerk's Tale (890), (33) Canon's Yeoman's Tale (1236) (el. Mazzon 2000: 163-165), and possibly Canon's Yeoman's Tale (1119-1120) and (1153-1154) (both regressively).
Mazzon (2000: 155, 158) assumes that there is a tendency for thou to co-occur with shal, while ye takes the allomorph shullen. This is certainly true for the Tale of Melibee, which she refers to, as well as for the entire Canterbury Tales. (34) In this case, however, we doubt that the collocational argument is appropriate. Mazzon maintains that shal is stronger, shul(en) rather tentative. MED has nothing about a semantic difference between the forms (cf. shulen (v. (1)). The choice of shal and shul simply seems to be determined morphologically.
The assumption of certain collocations with auxiliaries remains interesting, though. For later centuries Lass (1999: 149) points to this possibility. While no clear picture emerges in the case of shullen, the auxiliaries mowen and willen are more promising with 75 Y : 49 T (mowen) and 74 Y : 46 T (willen). The collocation with mowen may account for Knight's Tale (2312) (Emilye to Diana), and the Prologue to the Wife of Bath's Tale (369) (Wife of Bath to husband). (35) Willen occurs with ye in Knight's Tale (2254) (Palamon to Venus), Prologue to the Man of Law's Tale (42) (Manciple to Cook), (36) and in four more cases that have already been or will be explained with reference to the collocational-phraseological hypothesis. (37)
5.3. Further formulae, (fragments of) quotations, and a stereotyped comparison
The cases in this group are marked by phraseological force rather than frequency, as far as the text of the Canterbury Tales is concerned.
The rare occurrences of the formula faire (be)falle (to) X 'may X prosper, good luck to X' with a second person pronoun always feature Y. It might have prompted the pronoun changes in the Epilogue to the Nun's Priest's Tale (3460) (Host to Nun's Priest) and the Prologue to the Parson's Tale (68) (Host to Parson): (38)
7) "Sire preest", quod he, "now faire yow bifalle!" ["Sir priest", he said, "now good luck to you!"]
Heere may ye se 'By this may you see, i.e. understand' in Friar's Tale (1567) is a formula specific to a particular text type, namely the opening of a sermon, according to Shimonomoto (2001: 13). It is originally addressed to more than one person but seems to be responsible for the insertion of Y in this line, where the yeoman (i.e. the devil) addresses the summoner (1566-1568): (39)
8) "Lo, brother", quod the feend, "what tolde I thee? Heere may ye se, myn owene deere brother, The carl spak oo thing, but he thoghte another." ["Lo, brother", said the devil, "what did l tell you? By this may you see, my own dear brother, the fellow spoke one thing, but he thought another."]
In contrast, the line By swiche ensamples olde maistow leere 'Through these old exemplary stories can you learn' in Nun's Priest's Tale (3106), with which Chauntecleer introduces his explanation of the signification of his examples to Pertelote, should be interpreted as general T. (40) But since T may introduce the exposition in non-homiletic texts, such as fables, (41) it may possibly be another text-type specific formula. The courtly animals address each other in the polite form in all other instances.
Quotations or fragments of quotations (allusions) play a great role in medieval literature. Those quotations which are recognizable by the audience as institutionalized expressions are considered part of phraseology. (42) In the Miller's Tale, Absolon is not consistent in his choice of pronouns for addressing Alison, as we saw in example (4). After that collocationally-phraseologically triggered change, he uses Y in 3698-3707. In 3726, however, he changes back to T, which is then also continued in the later passage (3794-3797):
9) Lemman, thy grace, and sweete bryd, thyn oore!" [Darling, your favour, and sweetheart, your mercy!"]
Miller's Tale (3726) reflects the popular poetry of the time and also includes an allusion to sponsa [mea] in the Song of Songs. (43) Chaucer used ore in the sense of 'mercy' only at this place (cf. Donaldson 1963: 48). A corresponding formula (thyn ore!) can be found in Harley Lyric no. 32 (verses 16-21; here taken from Brook 1968: 72, emphasis added): (44)
10) Adoun y fel to hire anon ant crie, 'Ledy, pyn ore! Ledy, ha mercy of py mon! Lef pou no false lore! Jef pou dost, hit wol me reowe sore. Loue drecchep me pat y ne may lyue namoore! [Immediately I fell down in front of her and cried, "Lady, your mercy! Lady, have mercy on your servant! Do not believe in wrong advice! If you do, it will grieve me deeply. Love afflicts me so that I cannot live any more!]
The change to T in Miller's Tale (3726) may thus be attributed to a quotation with regressive influence on thy grace 'your favour' in the same line. Another allusion to the Song of Songs itself (1.15) is thyn eyen columbyn 'your dovelike eyes' in Januarie's address to May (Merchant's Tale 2141). (45)
And lastly, the expression Bileveth this as siker as your crede '(literally) believe this as sure as your creed', i.e. 'as guaranteed by divine degree or promise', spoken by the canon to the priest in Canon's Yeoman's Tale (1047) and prompting the use of Y for the rest of this conversation, belongs to a group of phraseological units known as institutionalized or stereotyped comparisons. (46) According to Whiting--Whiting (1968: C541) it occurs in the forms As sooth (sicker, true) as the Creed from the beginning of the fourteenth century on. But it seems that whenever the second person pronoun is used in these forms (two instances in Whiting--Whiting, none in MED apart from the Chaucer quotation), it is Y. (47)
The following cases with potential collocational-phraseological interest were excluded. One of them concerns the phraseological character of a particular line. This is the pronoun change in Monk's Tale (2451) from Y to T. Up until this line, Chaucer changed the pronouns from Dante's T to Y in the address of the starving children to their father Ugolino (Divina commedia, Inferno 33.1-75; cf. the Italian text with English translation printed in Bestul 2002: 429-432). But when they offer their father their own flesh to eat, Chaucer changes to T: Oure flessh thou yaf us, take oure flessh us fro 'You gave us our flesh, take our flesh from us' (cf. Dante's tune vestisti queste misere carni, e tu le spoglia 'you did clothe us with this wretched flesh, and do you strip us of it!', 62-63). At this point, the children address their desperate father with a devotion usually given to the divine Creator. Although Dante is the immediate source, allusions to the bible, in particular Job 1.21, 8.11 and 19.26, as well as the eucharist (John 17.53-58), might have been responsible for the change. (48) In two further cases it is the first occurrence in a series of pronoun switches which might have been triggered by the collocational-phraseological force. This is Reeve's Tale (4033) (John to Symkyn: "I pray yow") and Summoner's Tale (1772) (friar to Thomas: "God yelde yow"). However, as we in general started counting from the first deviation, these cases do not appear in our list. For reasons discussed in sections 4 and 5.2., above, passages with general T, especially in Prudence's speech in the Tale of Melibee, could also be considered as cases in point. In particular, for the allusions to the commandments ("thou shalt") and formulae structuring the text containing a form of conseillen and Y (including the first occurrence Tale of Melibee 1026: advocate to Melibee: "we conseille yow") collocational-phraseological influence could have been postulated. All these cases, however, are excluded from our account.
6. Evaluation of the evidence and outlook
In the discussion of the pronoun changes to which we attribute a collocational-phraseological influence we adopted a careful rather than an overly inclusive strategy. Still, about one third of the changes that we did include (31 of 90) can be explained by our hypothesis. As far as the proportion of insertions (1A, 1B) and conversions (2A, 2B) as the effect of these changes is concerned, there is no discernible tendency. Table 2 summarizes the absolute and relative numbers according to types.
Even if a more inclusive approach had been chosen, the numbers would not allow for an analysis by statistical methods since they derive from a limited textual basis, namely the Canterbury Tales. The explanatory power of each single case with its relevance for this text will have to be checked for Middle English in general, as some of the brief references to the entries in the MED show. Further studies along the suggested lines based on a wider textual basis are therefore very desirable.
In particular, our results may be considered in the context of historical English phraseology--an area of research which is still in need of extended and systematic investigation. Thus, developments in the collocational-phraseological store of the language are still largely undetected. Regarding the data that we discussed in this paper it is interesting, for instance, that a collocation which was dispreferred in Chaucer, namely to pray + T, but apparently preferred by a majority of speakers, developed the lexicalized form prithee in the sixteenth century. Ulrich Busse has analysed the use of (I) pray you and (I) pray thee/prithee in the Shakespeare Corpus (2002: ch. 7) and found, among other things, that the use of Y with to pray is more frequent than to pray and T (2002: oh. 7.3). Prithee, however, was gaining ground in Shakespeare's time, also in the context of an address with Y (2002: 203-204). (49) Another question in connection with language change is the influence of French collocations and phraseology on the English language, both on collocations and on the development of the lexicon and the phraseological store of English. (50) These and related questions will be left for future studies to answer.
In conclusion, we hope to have shown that pronoun changes in Chaucer's Canterbury Tales can be triggered by the collocational-phraseological force of the language. This force is so strong that it may even override the pragmatically preferred choice of the pronouns of address in the particular situation--after all, this is a rather sensitive area of communicative interaction. We are thus challenging the view that every deviation in the use of second person pronouns has an expressive function. However, our hypothesis does not attempt to exclude the possibility that both pragmatic and collocational-phraseological forces may have an effect at the same time--including the reinforcement of a pragmatic effect through a counter-collocational-phraseological usage. On the contrary, both explanatory paradigms are generally valid and may be employed in a complementary way in many cases. As has become apparent in our discussion of the data, though, pragmatic explanations fail in some cases and leave the collocational-phraseological hypothesis as the only explanation for a significant number of changes which have so far appeared unmotivated and arbitrary.
Benson, Larry Dean (ed.) 1988 The Riverside Chaucer. (3rd edition.) Oxford: Oxford University Press.
Brook, George L. (ed.) 1968 The Harley lyrics: The Middle English lyrics of ms. Harley 2253. (4th edition.) Manchester: Manchester University Press.
Littlehales, Henry (ed.) 1895 The prymer or Lay folks prayer book. (Early English Text Society, original series 105.) London: Kegan Paul, Trench, Trubner & Co.
Pearsall, Derek (ed.) 1999 Chaucer to Spenser. An anthology of writings in English: 1375-1575. Oxford: Blackwell.
Robinson, Fred Noris (ed.) 1957 The works of Geoffrey Chaucer. (2nd edition.) Boston, Mass.: Houghton Mifflin.
Bartsch, Sabine 2004 Structural and functional properties of collocations in English: A corpus study of lexical and pragmatic constraints on lexical co-occurrence. Tubingen: Narr.
Becker, Monika 2003 Familiar Dialogues in Englissh and Frenche: Sprachliche Interaktion und ihre Vermittlung in der fruhen Neuzeit. (Fokus: Linguistisch-Philologische Studien 26.) Trier: Wissenschaftlicher Verlag.
Benson, Morton--Evelyn Benson--Robert Ilson 1997 The BBI dictionary of English word combinations. (Revised edition.) Amsterdam--Philadelphia: Benjamins.
Bestul, Thomas H. 2002 "The Monk's Tale", in: Robert M. Correale--Mary Hamel (eds.), 409-447.
Blake, Norman 1992 "The literary language", in: Norman Blake (ed.), 500-541.
Blake, Norman (ed.) 1992 The Cambridge history of the English language. Vol. 2: 1066-1476. Cambridge: Cambridge University Press.
Brown, Roger--Albert Gilman 1964 "The pronouns of power and solidarity", in: Thomas A. Sebeok (ed.), 253-276.
Burnley, David 1983 A guide to Chaucer's language. Norman: University of Oklahoma Press.
2003 "The T/V pronouns in later Middle English literature", in: Irma Taavitsainen--Andreas H. Jucker (eds.), 27-45.
Busse, Ulrich 2002 Linguistic variation in the Shakespeare corpus: Morpho-syntactic variability of second person pronouns. (Pragmatics & Beyond: New series 106.) Amsterdam--Philadelphia: Benjamins.
Campbell, Robin 1981 "Language acquisition, psychological dualism and the definition of pragmatics", in: Herman Parret--Marina Sbisa--Jef Verschueren (eds.), 93-103.
Correale, Robert M.--Mary Hamel (eds.) 2002 Sources and analogues of the Canterbury Tales. Vol. 1. Cambridge: Brewer.
Cowie, Anthony P. 1998 "Introduction", in: Anthony P. Cowie (ed.), 1-20.
Cowie, Anthony P. (ed.) 1998 Phraseology: Theory, analysis, and applications. Oxford: Clarendon Press.
Cowie, Anthony P.--Ronald Mackin (eds.) 1993 Oxford dictionary of phrasal verbs. (Second edition of Oxford dictionary of current idiomatic English. Vol. I.) Oxford: Oxford University Press.
Cowie, Anthony P.--Ronald Mackin--Isabel R. McCaig (eds.) 1983 Oxford dictionary of current idiomatic English. Vol. 2: Phrase, clause & sentence idioms. Oxford: Oxford University Press.
 [Reprinted under the title: Oxford dictionary of English idioms.]
Donaldson, Ethelbert Talbot 1963 "Idiom of popular poetry in the Miller's Tale", in: William K. Wimsatt, Jr. (ed.), 27-51.
Elliott, Ralph W.V. 1974 Chaucer's English. London: Deutsch.
Finkenstaedt, Thomas 1963 You und thou: Studien zur Anrede im Englischen (mit einem Exkurs fiber die Anrede im Deutschen). Berlin: De Gruyter.
Glaser, Rosemarie 1986 Phraseologie der englischen Sprache. Tubingen: Niemeyer.
1998 "The stylistic potential of phraseological units in the light of genre analysis", in: Anthony P. Cowie (ed.), 125-143.
Gray, Douglas 1988 "Explanatory notes to The Miller's Tale", in: Larry Dean Benson (ed.), 842-848.
Harley, Martha Powell 1988 "Chaucer's FRIAR'S TALE", Explicator 46/2: 4-5.
Hausmann, Franz Josef 2004 "Was sind eigentlich Kollokationen?", in: Kathrin Steyer (ed.), 309-334.
Honegger, Thomas 2003 "And if ye wol nat so, my lady sweete, thanne preye I thee, [...]: Forms of address in Chaucer's Knight's Tale", in: Irma Taavitsainen--Andreas H. Jucker (eds.), 61-84.
Hope, Jonathan 1993 "Second person singular pronouns in records of Early Modern 'spoken' English", Neuphilologische Mitteilungen 94: 83-100.
Johnson, Judith A. 1977 "Ye and thou among the Canterbury pilgrims", Michigan Academician 10: 71-76.
Johnston, Everett C. 1962 "The pronoun of address in Chaucer's Troilus", The Language Quarterly 1: 17-20.
Jucker, Andreas H. 2006 "Thou art loothly and so oold also': The use of ye and thou in Chaucer's Canterbury Tales", Anglistik 17/2: 57-72.
Jucker, Andreas H.--Irma Taaavitsainen 2003 "Diachronic perspectives on address term systems: Introduction", in: Irma Taavitsainen--Andreas H. Jucker (eds.), 1-25.
Karpf, Fritz 1930 Studien zur Syntax in den Werken Geoffrey Chaucers. Part I. (Wiener Beitrage zur englischen Philologie 55.) Wien--Leipzig: Braumuller.
Kaske, Robert E. 1962 "The Canticum Canticorum in the Miller's Tale", Studies in Philology 59: 479-500.
Kastovsky, Dieter--Arthur Mettinger (eds.) 2000 The history of English in a social context: A contribution to historical sociolinguistics. (Trends in Linguistics; Studies and Monographs 129.) Berlin--New York: Gruyter.
Kennedy, Arthur Garfield 1915 The pronoun of address in English literature of the thirteenth century. Stanford, Calif.: Stanford University Press.
Kerkhof, Jelle 1982 Studies in the language of Geoffrey Chaucer. (2nd edition.) Leiden: Brill.
Koziol, Herbert 1943 "Die Anredeform bei Chaucer", Englische Studien 75:170-174.
Kurath, Hans et al. (eds.) 1952-2001 Middle English dictionary (MED). 13 vols. Ann Arbor: University of Michigan Press.
Lass, Roger 1999 "Phonology and morphology", in: Roger Lass (ed.), 56-168.
Lass, Roger (ed.) 1999 The Cambridge history of the English language. Vol. 3: 1476-1776. Cambridge: Cambridge University Press.
Lumiansky, Robert M. 1955 Of sondry folk: The dramatie principle in the Canterbury Tales. Austin: University of Texas Press.
 [First paperback print; Austin: University of Texas Press.]
Makkai, Adam 1972 Idiom structure in English. (Ianua linguamm, Series maior 48.) The Hague--Paris: Mouton.
Mazzon, Gabriella 2000 "Social relations and forms of address in the Canterbury Tales", in: Dieter Kastovsky--Arthur Mettinger (eds.), 135-168.
Mel'cuk, Igor 1998 "Collocations and lexical functions", in: Anthony P. Cowie (ed.), 23-53.
Miller, Clarence H. 1992 "Three phrases in the Canterbury Tales: 'your dyvyntee' (FrT III-D 1512), 'your gentillesse' (SqT V-F 695), 'Goode lief' (Pro WBT III-D 431)", Notes and Queries 237: 152-155.
Moon, Rosamund 1998 Fixed expressions and idioms in English: A corpus-based approach. Oxford: Clarendon Press.
Nathan, Norman 1956 "Pronouns of address in the 'Friar's Tale'", Modern Language Quarterly 17: 39-42.
1959 "Pronouns of address in the 'Canterbury Tales'", Mediaeval Studies 21 : 193-201.
Nevala, Minna 2004 Address in early English correspondence: Its forms and socio-pragmatic functions. (Memoires de la Societe Neophilologique de Helsinki 64.) Helsinki: Societe Neophilologique.
Oizumi, Akio (ed.) 2003 A lexical concordance to the works of Geoffrey Chaucer. 5 vols. Hildesheim: OlmsWeidmann.
Pakkala-Weckstrom, Mari 2004 "Discourse strategies in the marriage dialogue of Chaucer's Canterbury Tales", Neuphilologische Mitteilungen 105: 153-175.
Parret, Herman--Marina Sbisa--Jef Verschueren (eds.) 1981 Possibilities and limitations of pragmatics. Proceedings of the Conference on Pragmatics, Urbino, July 8-14, 1979. Amsterdam--Philadelphia: Benjamins.
Prins, Anton Adriaan 1952 French influence in English phrasing. Leiden: Universitaire Pers.
1959 "French influence in English phrasing (a supplement)", English Studies 40: 27-32.
1960 "French influence in English phrasing (concluded)", English Studies 41: 1-17.
Sebeok, Thomas A. (ed.) 1964 Style in language. (2nd edition.) Cambridge: M.I.T. Press.
Shimonomoto, Keiko 2000 The language of politeness in Chaucer: An analysis of the use of linguistic features reflecting social and interpersonal relationships. Tokyo: Waseda University Enterprise.
2001 The use of ye and thou in the Canterbury Tales and its correlation with the terms of address and forms of the imperative. Tokyo: Waseda University Enterprise.
Sinclair, John 1991 Corpus, concordance, collocation. Oxford: Oxford University Press.
Skeat, Walter W. (ed.) 1894 The complete works of Geoffrey Chaucer. Vol. 5. Oxford: Clarendon Press.
Steyer, Kathrin (ed.) 2004 Wortverbindungen--mehr oder wenigerfest. (Institut fur Deutsche Sprache, Jahrbuch 2003.) Berlin--New York: Walter de Gruyter.
Taavitsainen, Irma--Andreas H. Jucker (eds.) 2003 Diachronic perspectives on address term systems. (Pragmatics & Beyond: New series 107.) Amsterdam--Philadelphia: Benjamins.
Walcutt, Charles Child 1935 "The pronoun of address in Troilus and Criseyde", Philological Quarterly 14/3: 282-287.
Whiting, Bartlett Jere--Helen Wescott Whiting 1968 Proverbs, sentences, and proverbial phrases from English writings mainly before 1500. Cambridge, Mass.: Belknap Press of Harvard University Press.
Wilcockson, Colin 1980 "'Thou' and 'ye' in Chaucer's Clerk's Tale", Use of English 31/3: 37-43.
2003 "Thou and tears: The advice of Arveragus to Dorigen in Chaucer's Franklin's Tale", Review of English Studies 54: 308-312.
Wimsatt, William K., Jr. (ed.) 1963 Explication as criticism. Selected papers from the English Institute 1941-1952. New York--London: Columbia University Press.
GABRIELE KNAPPE--MICHAEL SCHUMANN
Otto Friedrich University, Bamberg--University of Berne
(1) For pronoun shifts in Troilus and Criseyde, cf. Walcutt (1935); Johnston (1962); Finkenstaedt (1963: 77-84); Shimonomoto (2001: 7-8).
(2) All quotations in this paper are taken from Benson (1988); emphases added.
(3) We concentrate on changes in the pronoun use in the speech of the same communication partners. Hence we do not take account of (supposedly) "incorrect" but consistent pronoun choice. For our purposes, we consider the Canterbury Tales as a homogeneous piece of art despite its complicated genesis and related problems (cf., e.g., Benson 1988: xxv).
(4) Cf. Shimonomoto (2001), based on her 1986 Sheffield M.A. thesis; Honegger (2003); Mazzon (2000); Jucker (2006).
(5) Cf. Brown--Gilman (1964), first published in 1960. Our use of the terms macro- and micropragmatics is not congruent with Campbell's (1981: 101).
(6) Wie du mir, so ich dir 'an eye for an eye; tit for tat' (literally: 'As you [do] to me, so I [do] to you'); Du ahnst es nicht! 'Would you believe it?' (literally: 'You do not guess this'); Du sollst nicht toten 'You shall not kill' (cf. Thou shalt not kill); Das kannst du dir abschminken 'You can forget about that' (literally: 'You can remove this [like make-up]'); wenn Sic gestatten 'if you permit'.
(7) In addition, "idioms" in one or the other definition--a concept with a partial overlap with collocation and phraseology--have also been studied in their own right.
(8) Cf. also Moon (1998: 26) and the minimal consensus that Bartsch found for the conceptions of collocations: "frequently recurrent co-occurrences of lexical items" (2004: 65).
(9) Other classes of collocations which have been suggested are lexical versus grammatical ("Grammatical collocations consist of a dominant word--noun, adjective/participle, verb--and a preposition or a grammatical construction"; Benson--Benson--Ilson 1997: ix), syntactic (e.g. had been, one of many--of), or paradigmatically related co-occurrences such as jam and marmalade, apricot (for a short summary, cf. Moon 1998: 26-27).
(10) Collocation studies have repeatedly stressed the factor of "automatic" language use which may run counter to rules applying to the free combination of elements.
(11) All definitions of present-day English phraseological units are taken from Cowie--Mackin--McCaig (1983 ) or Cowie--Mackin (1993).
(12) Cf., e.g., the publications by Hausmann, recently the discussion in Hausmann (2004). From his point of view of lexical functions, Mel'cuk (1998:29-31) deals with these kinds of units under the term "collocation", a subclass of semantic phrasemes.
(13) Cf. the overview in Cowie (1998: 7).
(14) To trip/dance/tread the light fantastic '(facetious) to dance' from Milton's lines Come, and trip it as ye go/ On the light fantastic toe. In this phraseological unit, a limited choice of collocates is possible for the verb.
(15) The notions of semantic unity and idiomaticity are reflected in Bartsch's (2004: 77) inclusion of an obligatory "element of semantic opacity such that the meaning of the collocation cannot be said to be deducible as a function of the meanings of the constituents" in her definition.
(16) Cf., e.g., Moon (1998: 61), Cowie (1998: 14-15) about pure idioms.
(17) Cf., e.g., Mel'cuk's (1998: 29) notion of "pragmateme", and Makkai's group of "sememic or cultural-pragmemic idioms" (1972: ch. 1.3.5).
(18) Paraphrases of the more idiomatic and perhaps not frequently used of these examples: One swallow does not make a summer 'one fortunate incident etc. should not be taken to mean that the general situation has improved or is about to'; a new broom (sweeps clean) 'somebody recently appointed to office or a responsible post (starts with an energetic programme of reform and change, sometimes not welcomed by those already there)'.
(19) Strangely, in (711) the Wife even uses the T-form to address the pilgrims (cf. Shimonomoto 2001: 44: "demands of rhyme"; see also Nathan 1959: 199).
(20) Difficulties in interpreting these forms occur in passages such as (1002-1003) or (1410-1427), where the "tone of sermon" with universal T is predominant, but where some pronouns unequivocally refer to Melibee: "Thy name is Melibee" (1410) and "thy three enemys been entred into thyn house by the wyndowes/and han ywounded thy doghter in the forseyde manere" (1425-1427).
(21) Cf. also the change in Canon's Yeoman's Tale (1290). On a brief discussion of the collocation of shal with T, cf. section 5.2, below.
(22) Lenvoy de Chaucer at the end of the Clerk's Tale (Chaucer to all women): (1200) Y [right arrow]) (1202) T; Merchant's Tale (Justinus to Placebo): (1530) Y [right arrow]) (1535) T, Pardoner's Tale (Pardoner to the pilgrims): (590) Y [right arrow]) (648) T, Parson's Tale (590) Y [right arrow]) (591) T (thou shalt; maxims/biblical references).
(23) Bibliographical references to pragmatically based suggestions in earlier studies and information from the Middle English dictionary (MED) are provided for each case. Entry forms in the MED and the selection of the quotations have to be treated with care, however, as the compilers did not always pay heed to the form of the second person pronoun.
(24) Burnley (2003: 31) mentions for your curteisie as a phrase in Chaucer. Cf. also the attestations in MED, s.v. leve (n. (2)) (d): T before Chaucer, Y/T after Chaucer; MED, s.v. courteisie (n.) 3. (b.): only Y.
(25) Cf. MED, s.v. preien (v. (1)) 5: T before (1390), then Y/T.
(26) See on this passage especially Honegger (2003: 68-69, 75-78); furthermore Shimonomoto (2001: 43); Mazzon (2000:139); Karpf (1930: 42); Kerkhof (1982:231).
(27) Lady does co-occur with T, though; cf. Knight's Tale (2260) (as opposed to Shipman's Tale 491), Man of Law's Tale (850-851), Merchant's Tale (2367) and other cases.
(28) Cf. Lumiansky (1955 : 103). Karpf (1930: 45), Johnson (1977: 74), Shimonomoto (2001: 9) and Kerkhof (1982: 228) comment on the relationship between the Host and the Monk. Kerkhof (1982: 228) wrongly attributes 2768 (speech of the Knight) to the Host. Cf. also MED, s.v. God(n. (1)) 9. (c): T/Y.
(29) Cf. MED, s.v. liken (v.(1)) 1b.-1c.
(30) Miller (1992: 152, n. 3) has if that yow leste as a "standard phrase"; MED, s.v. listen (v. (1)) 1a-1c. Cf. also Prins (1952: 160) on the influence of French s'il vous plait on if you please in Early Modern English.
(31) On these passages, cf. also Kerkhof (1982: 229); Shimonomoto (2001: 36, 30-31); Mazzon (2000: 151-152). On the collocation in the MED, s.v. tellen (v.) 4. (e.).
(32) Collocates are armes, bak, body, bosom, brestes, cheke, eres, ers, eyen, face, feet, hand, heed, herte, lymes, mouth, tayl, tonge, tooth and visage. Cf. also MED, s.v. thin (pron.) 1.d., your (pron.) 4a-4b).
(33) Cf. Wilcockson (1980: 40-42; 2003: 310); Kerkhof(1982: 229); Shimonomoto (2001: 33).
(34) The numbers for the Tale of Melibee are: ye shal (3), thou shalt/shaltow (34), ye shul(en) (43), thou shul (0) andye sholde (7).
(35) On the first passage, cf. Karpf (1930: 43); Kerkhof (1982: 231); Shimonomoto (2001: 43); Honegger (2003: 70-72). On the second passage, cf. Kerkhof (1982: 230).
(36) On the first passage, cf. Karpf (1930: 42); Mazzon (2000: 139); Shimonomoto (2001: 43); Honegger (2003: 77). On the second passage, cf. Kerkhof (1982: 228).
(37) This means we can allocate a double collocational-phraseological motivation to Miller's Tale (3362), Wife of Bath's Tale (1012), Friar's Tale (1567), and Merchant's Tale (2169).
(38) On these passages, cf. also Kerkhof (1982: 227); Lumiansky (1955 : 108, 110); Shimonomoto (2001: 9). Cf. MED, s.v. bifallen (v.) 3. (a): Y/T.
(39) For another phraseologically-conditioned change in this pair of interlocutors, cf. example (1) and section 5.1., above. On this passage, cf. also Nathan (1956: 42); Miller (1992:152).
(40) For a different interpretation, cf. Walcutt (1935: 283-284).
(41) Cf., e.g., Robert Henryson, The Wolf and the wether (2595), in: Pearsall (1999: 498).
(42) Cf. Glaser (1986: ch. 5) and also, for instance, Makkai (1972: 169-172, 177-178), who acknowledges the (pseudo-)idiomatic character of this group of "cultural-pragmemic idioms" in his account of the idiomatic structure of English: "the quotation has to be essentially institutionalized, i.e. familiar enough to be fairly sure of being recognized by most speakers, whether in its original form or in a varied form as an allusion to the original quotation" (Makkai 1972:177).
(43) Cf. Donaldson (1963: 47-48); Gray (1988: 848); Kaske (1962: 482).
(44) MED, s.v. or(e (n. (2)) 1 (d) suggests that "thyn ore" is always used with T, and also in the address of Jesus. Prins (1952:280-281) lists Thine ore! as a formula deriving from French Vostre mercy!--apparently with a change from Y to T.
(45) cf. Shimonomoto (2001: 30); Mazzon (2000: 151-152) for pragmatically-based explanations.
(46) On the multiple switchings between the canon and the priest, cf. Mazzon (2000:163-165).
(47) Cf. MED, s.v. crede (n. (2)) (b), siker (adj.) 4b (b).
(48) Job 8.11 and 19.26 are also part of the Office for the Dead, cf. Littlehales (1895: 60, 69). Mazzon (2000: 141-142) attributes this change to T to an increase in dramaticity and emotional tone alone.
(49) Busse regards the development of prithee as a case of grammaticalization. Pronoun changes were still rather common in Early Modern English; cf. Busse (2002: ch. 2.7). On prithee, cf. also Nevala (2004: 171-172).
(50) The fullest book-length study on French influence in English phraseology is still Prins (1952) with supplements (1959, 1960), who claims French influence for some phraseological units with second person pronoun such as Thine ore! and if you please (cf. fun. 44 and 30, above). A detailed study of the exact forms of this influence is still missing, but small-scale comparisons have shown that Chaucer seems to be independent of the pronoun use of his sources (cf. Nathan 1959: 198).
Table 1. Instances of second person pronoun change Tale Speaker [right arrow] Addressee: lines T/Y-form Type KnT widow [right arrow] Theseus: 920 T, 927 Y, 930 T 3 Palamon [right arrow] goddess Venus: 1105 Y, 2237 Y, 1B 2249-50 Y, 2254 Y Emilye [right arrow] goddess Dyane: 2312 Y 1A MIT Alison [right arrow] Nicholas: 3287 Y, 3722 T 2B Nicholas [right arrow] Alison: 3297 Y 2A Absolon [right arrow] Alison: 3362 Y, 3726 T 2B Gerveys [right arrow] Absolon: 3781 T 1A RvT Aleyn [right arrow] John: 4044-45 Y (rhyme position) 1A John [right arrow] Symkyn: 4128 T 2A MLT Custance [right arrow] her father: 1105 Y (after 2A Custance [right arrow] her mother Y, 276-ff.) WBTP Wife of Bath [right arrow] Pardoner: 188 Y 2A Wife of Bath [right arrow] husband: 241 Y, 331 Y, 1B 369 Y, 434 Y Husband [right arrow] Wife of Bath: 319 Y 2A WBT olde wyf [right arrow] knight: 1009 T, 1012 Y, 3 1015 T, 1088 Y (rhyme position) knight [right arrow] olde wyf: 1100 T 1A FrT yeoman (= devil) [right arrow] summoner: 1399 Y, 1B 1567 Y summoner [right arrow] yeoman (= devil): 1444 T, 1B 1526 T old widow [right arrow] summoner: 1618 T, 1623 T 1B SumT friar [right arrow] Thomas: 1785 T, 1832 Y, 1944 T, 3 1955 Y, 1970 T, 1974 Y, 1985 T, 1999 Y, 2089 T, 2112 Y (rhyme position), 2154 T Thomas [right arrow] friar: 2131 T 2A CITP Host [right arrow] Clerk: 14 T 1A ClT Walter [right arrow] Grisilde: 483 T, 492 Y, 890 T 3 MerT Januarie [right arrow] May: 2141 T, 2169 Y, 2367 T 3 SqT Franklin [right arrow] Squire: 686 Y 2A FranT Dorigen [right arrow] God: 867 Y, 879 T, 881 Y 3 Aurelius [right arrow] god Phebus: 1041 Y, 1077 T 3 Arveragus [right arrow] Dorigen: 1482 T 1A ShipT merchant [right arrow] his wife: 384 Y, 395 T 2B PrTP Prioress [right arrow] Mother of God: 486 Y 2A Mel an advocat [right arrow] Melibee: 1026 T 2A Melibee [right arrow] Prudence: 1233 Y, 1262 T, 3 1427 Y MkT son/children [right arrow] father: 2451 T 2A MkTP/ NPTP Host [right arrow] Monk: 1932 T, 2 788 Y 2B NPT Chauntecleer [right arrow] Pertelote: 3106 T, 3121 Y 2B narrator [right arrow] Gaufred (Geoffrey of 2A Vinsauf): 3351 Y (rhyme position) Chauntecleer [right arrow] fox: 3428 T 2A NPTE Host [right arrow] Nun's Priest: 3460 Y 2A SNT Urban [right arrow] God (Christ): 199 Y (rhyme 2A position) Cecilie [right arrow] Almachius: 463 T 2A CYT canon [right arrow] priest: 1047 Y, 1119-20 T, 3 1125 Y, 1153-54 T, 1181 Y, 1236 T, 1250 Y, 1290 T, 1327 Y, 1360 T, 1361 Y MancTP Manciple [right arrow] Cook: 42 Y 2A ParsTP Host [right arrow] Parson: 68 Y 2A Table 2. Numbers and percentages of second person pronoun changes according to types Type 1A Type 1B Type 2A number of changes 6 14 17 of these: 1 7 6 collocational- phraseological Type 2B Type 3 Total number of changes 10 43 90 of these: 5 12 31 collocational- phraseological