SELIM08.doc Francisco Alonso, Selim 8 (1998): 171—191 ‘AS IT YS SEYDE TO FORE’. SOME LINGUISTIC EVIDENCE IN THE PROCESS OF COMPILING MIDDLE ENGLISH MEDICAL RECIPES1 1. INTRODUCTION The aim of this paper is to describe the compilation of fifteenth-century English medical recipes. My main interest is to unravel those devices used by medieval compilers and scribes to arrange recipes in the medical manuscript book. The material used for such a study consists of one of the receptaria contained in MS Hunter 185 (folios 17r-62r). The work is organised as follows: first, I describe medieval English med- ical books, then I present a description of MS Hunter 185. After that, I intro - duce the framework of discourse analysis (Brown & Yule 1996) which I in tend to apply in the data analysis which deals with the organisation of recipes. 2. MEDIEVAL ENGLISH MEDICAL BOOKS Medieval English medical books are indexed normally according to two main parameters: their contents, and the needs of the audience. The first parameter is adopted in Robbins (1970), in which the medical material is clas s ified into three different groups: prognosis, diagnosis and treatment. This classification does not completely satisfy our expectations, since, in the case of remedy books, for instance, these often contain recipes belonging to the three 1 I would like to thank Mercedes Cabrera and Alicia Rodríguez for the painstaking work of reading earlier drafts of this article. F rancisco A lonso 168 groups. Thus, this indexing procedure results in a continuous repetition of the same recipe books in the three cataloguing groups. The second parameter, the audience, seems to be a more plausible crite- rion for indexation, because the audience required specific type of books which, depending on what the reader expects from them, showed a particular use of language and a specific structure. In this sense, Voigts (1982) pro - poses the division of the Middle English medical material into two main groups. These groups are (a) remedybooks, and (b) learned treatises. The first group constit utes the bulk of medical books which contain handy information about the treatment of particular sicknesses in easy-to-follow steps. These steps present a fairly simple structure and plain language1, which sometimes included the use of Latin and Anglo -Norman. The audience of this group is larger than that of learned treatises, as Norri (1992: 33) points out: “Vernacular remedy books have often been associated with lay healers, but these books were also owned by university-trained physicians.” Learned treatises, however, were directed to a more limited type of audience such as university scholars. These books are based on classical and medieval medical tradition, and many are written in Latin. However, Latin was steadily being replaced by English by the fifteenth century 2, partly due to the growth of an incipient middle class bourgeoisie that demanded books of a more practical nature for their own instruction (Parkes 1973: 563). Thus, the need for translations and new written ma terial in English was urgent, if we consider that this middle class population owned very little knowledge, if any, of the Latin language. This insufficient knowl- edge of Latin was mainly due to the lack of proper instruction in that lan- guage, and whenever fifteenth-century people were instructed in Latin, the methodology was oriented towards a specific use, and hence the instruction was far from integral, as claimed by Rodríguez Álvarez in her description of fifteenth-century elementary schools (1997: 39): El latín que se daba era mínimo, simplemente se pretendía que los niños pudieran leerlo y pronunciarlo correctamente para poder leer 1 For a detailed study of the structure and the language of the Middle English recipe, see Alonso Almeida (forthcoming). 2 For the process of “Englishing” of medical material, see Voigts (1995: 184). L inguistic evidence in the compiling of Middle E nglish medical recipes 169 textos devocionales y cantar himnos, pero la gramática y el sig - nificado de los textos carecía de importancia. As said above, the remedybooks were aimed at a larger readership. Thus, the number of these books was quantitatively significant, as can be seen by the extant copies in British and American Libraries 1. This group has certainly enjoyed much editorial attention as pointed out by Voigts (1982: 47) if compared with learned treatises. However, research in this area mainly in - cludes studies on the description of the contents and the structure of these books and their indexing, but there is still much to be done on the writing of the medieval medical receptarium, at least, from a linguistic perspective. 3. MS HUNTER 185, FF. 17R-62R MS Hunter 185 is a Middle English medical recipe book which also contains a flora medica and an alphabetical list of simples. The manuscript is housed at the Glasgow University Library (Hunterian Collection, University of Glasgow, Glasgow). There is no evidence of authorship, or of ownership; although, according to Young and Aitken (1908: 132), on the vellum fly -leaf, there is what seems to be “probably the autograph of Henry Swinburne, B. C. L. (1560-1623).” MS Hunter 185 was written by different scribes, all of them using late fourteenth-century Anglicana, and Anglicana Formata. The folios object of this study are written by a single scribe using Anglicana Formata throughout. These folios present rubricated running titles and brown upper cases at the beginning of recipes in order to show the ordinatio. These visual aids are fundamental for the understanding of medical recipe compilations, since they constitute evidence of the scribe’s effort to show the thematic arrangement of recipes, as we will see later on in this paper. Folios 17r-62r contain a number of medical recipes written in English and very few in Latin. The nature of the recipes is varied, and they are organised generally following the traditional capitem ad pedem structure -type, that is, therapeutic remedies for the head are given at the beginning, and this is fol- 1 See Voigts (1995) and Schmidt (1994). F rancisco A lonso 170 lowed by remedies for the eyes, the nose, the mouth, and other parts of the body up to the toes. The remedies presented are mainly waters, oils, plasters, salves, ointments, drinks, laxatives, and powders among others. 4. FRAMEWORK OF DISCOURSE ANALYSIS In the analysis of the linguistic strategies used to compile recipes, I follow the discourse analysis model proposed in Brown & Yule (1996(1983)), henceforth B&Y (1996). Such model constitutes a significant contribution to text analysis for it provides points of view from previous discourse analysis theories, and offers clear descriptions of the strategies involved in the writing of discourse. Substitution, ellipsis and referential expressions are the devices which appear recursively in MS Hunter 185 as a way to establish textual cohesion. Let us move on to the description of the aforementioned devic es according to B&Y (1996). Substitution and ellipsis are described as those techniques whose relationship with other parts of the texts “forces the reader … back into the text to look for a previous expression to substitute, in the case of substitution, or to provide, in the case of ellipsis” (1996: 193). Thus, the use of those two strategies creates cohesive relations among different parts of the texts, since the reader needs the reference given elsewhere within the text so as to fully understand the in tended message. In order to illustrate these two concepts, let us concentrate on the following two examples taken from B&Y (1996: 193): (1a) Jules has a birthday next month. Elspeth has one too. (1b) Jules has a birthday next month. Jules has too. The examples above show instances of substitution and elision. The word one in (1a) is not understood by the reader unless he goes back to the previ- ous sentence for reference. Thus, one stands for birthday, establishing a relation of dependence of the second sentence on the first one. In the same vein, in (1b), the omission of any pro -form which stands for birthday makes the reader to go back into the text in order to interpret the message conveyed L inguistic evidence in the compiling of Middle E nglish medical recipes 171 in the second sentence, linking both sentences by means of a dependence relationship. The referential elements which may posit any interest for this study is anaphora. Anaphora describes the relationship of two elements within the text where the second refers to the first element back in the text for interpre - tation (B&Y1996: 192). The following exemplifies this point: (2) Look at the sun. It’s going down quickly. The previous example illustrates how the use of the anaphoric pronoun it links the second sentence to the one before. Thus, the reader must go back in the text to be able to interpret such anaphoric pronoun. This time the referent is the word sun which is included as said above in the previous sentence. 5. ORGANISATION OF RECIPES IN MS HUNTER 185, FF. 17R-62R After the introduction of the description of the manuscript and the linguistic theoretical framework to be used here, let us turn to the analysis of the com- pilation of recipes. I analyse the extralinguistic strategies which link recipes in the manuscript, then I focus on the linguistic strategies which have the same linking purpose. Chart 1 below illustrates graphically the manifestation of both extralinguistic and linguistic linking strategies. In chart 1, the extralinguistic strategies in MS Hunter 185 are represented by means of the dotted arrow, whic h corresponds to the marginalia, and the bold type-face, which corresponds to the rubrics. The linguistic strategies comprise the organisation of recipes from head-to-toe structure, which is here represented by means of an arrow, and the subgroupings of recipes with a same topic, here represented by boxes. In addition, linguistic strategies in - clude the use of referential expressions within the recipe boxes, which I de- scribe in the chart by means of curved two -headed arrows and parentheses; and the use of cross-references in the manuscript, represented in the chart by means of dotted two -headed arrows. F rancisco A lonso 172 The presence of these strategies in MS Hunter 185 are analysed below. I shall begin by presenting the extralinguistic elements, and finally, I will move on to the analysis of the linguistic elements. Chart 1 Key: RxA: Recipes Group A RxB: Recipes Group B RxC: Recipes Group C Linguistic strategies: from-head-to-toe organisation Ø ellipsis thematic boxes ( ) reference particles co-reference (anaphora, ellipsis , substitution) cross-referential expressions Extralinguistic strategies: marginalia bold type-face rubrics 5.1. EXTRALINGUISTIC ELEMENTS The organisation of the topic from head to toe requires visual techniques which help to reduce the time employed by the reader to locate the informa - tion needed. Those techniques are the use of rubricated running titles, and a marginal apparatus. The former facilitate the search of specific information. The entire title is written in red ink, including the key word, which indicates the recipe con- tents. It should be also noted that, in many cases, the title may contain up to four lines. Besides, occasionally the title does not offer any indication of the purpose of the recipe. In the manuscript studied here, we can see how this represented a serious problem for the reader, as shown by the fact that the reader has underlined the key word in a long title and in the recipe body text 1: (3) 1 Transcriptions mine. Abbreviations have been silently expanded throughout. Original spelling and punctuation have been retained in the transcription. Bold type- face indicates rubrics. L inguistic evidence in the compiling of Middle E nglish medical recipes 173 A good medicyne for to stanche blood whan ne a master veyne is cutte & pe wounde be la rge Tak a pese of salt beef pe leue & non of pe fatte a muche as wil in pe wounde & let hit be leyd in pe hote asschis in pe fuyre • & let roste it til it be pur¥ hoot / & also hoot pruste hit in to pe wounde • & bynde hit faste • & hit schal staunche a noon • & neuere steyne more vp warantise (MS Hunter 185, f.35r, ll. 2118-2126) (4) Here it tellip wharfore pis oyneme nt is good This plaster is good for olde brusoures & for akyng of lymmes hit destruyep an hard posteme & it is so - ueraynely good for broken bones pe surgerens of Salerne vsep it for alle (…) (MS Hunter 185, f. 49r, ll. 2784-2789) In (3), the reader makes clear what the recipe is for by means of underlin - ing in a different coloured ink the key word veyne in the title. In (4), the reader must read through the whole recipe if he wants to know the use of the plaster described. This is the reason why the reader underlines the string good for broken bones pe. The latter, to indicate the location of specific pieces of information, is also a useful way to identify recipes. This visual aid consists in the addition of keywords, numbers, signs, or even, drawings to the margins of the manuscript, in order to indicate the purposes of a particular recipe. In MS Hunter, marginal notes also seem to have been added by readers rather than by the scribe himself. This observation follows from my own preliminary palaeographical studies. The following examples illustrate this point1: (5) Also for pe / scabbes of a mannes body • Tak horho / ne cely- doyne & warmot & sorel / de boys & helena campana • ana & seep hem in faire / rennyng water • til half be wasted & 1 Heads in the manuscripts are here represented by the “face” symbol. F rancisco A lonso 174 perwith / wassch hem euery day wel ones til pe seke / be hool pourgh pe • grace of god Also for / scabbes in a mon or in a womman § / Tak pe rote of horshelue & seep it in / water til hit be nessche & tak olde sche / pes talwe & medle hem wel to gedre / in a mortere & do it in a lynnen clop / & anoynte pe sore a gayn pe fuyre & / he schal be hol Also anoper for pe same / Tak horhoue & seep it in rennynge / water til half be wasted & wessche pe / seke perwith pe water as hoot as he may /suffre • & anoynte hem with pe water oyne / ment pat is made of pese herbes Tak / pe rote of pe rede dokke and of celi- / doyne & wermot also & sorell de boys / ana & bray hem to gedre & tempre hem / to gedre with mayes boter & make an / oynement per of & pat wil hele eny maner / of scabbes certayn Also an other for / scabbes § Tak pe rededokke rotis & pym / purnell & scabiose & sorell de boys & / selydoyne & stampe hem & tempre hem to / gedre with mayes boter & fry¥e hem / wel to gedre & streyne hem in to a clene / vessell & perto a good quantite of powder / of brymstone & fry¥e • hit wel to gedre til / it be cold & perwith anoynte pe seke til he / be hol (MS Hunter 185, ff. 23v-24v) (6) For schorned heuedes and scattede / heuedes Tak poo dolyon ij handfules / ar he be floured pe while he is tendre & / sep hym wel in to a potel of strong ly¥e / til half be wasted & panne let wessche pe / schaled hed in strong pisse ry¥t hoot & schaue / clene of pe scalles & let no¥t for no bledyng / & panne make a plaster of pe poo delyon & / let hit on pe hed wel warme & so let it ly / a day & a ny¥t & panne tak it of & tak rye / mele & rennyng water & mak a pap ri¥th / pykke & spred it on a clop pat wil ouer sprede pe / sore & ley it on pe sore hed & let it lye iij day / es • & iij ny¥t or it be remeued & tak it panne / of & wessche pe hed in strong pisse a ¥eyne / & schaue it clene to pe flessch & tak rede / oynones as mony as wil suffice for a p laster / ouer pe sore & let boyle hem wel in water / & stampe hem & tempre hem vp with jus of Cala /mynte & tempre hem vp with olde barewes / grece faire j molte & j pured & vse pis last til / pe seke hed be hool for pys wil hele it sy / kyrly pourgh godes grace (MS Hunter 185, f. 27r) L inguistic evidence in the compiling of Middle E nglish medical recipes 175 (5) illustrates the remedies for the head highlighted by means of drawing a head in the margins, so that the reader interested in this specific bit of infor- mation can identify it in a very specific area of the text rather t han having to read through the whole manuscript. (6), which also shows remedies for the head, presents another drawing which could be identified as a head, though, in this case, its shape is less well outlined than in the previous example. These two drawings are attributed to a sixteenth-century owner of the book who presented it to one of his friends’ son who suffered from “a sore hedde”, possibly a scabby head, as indicated on the fly -leaf verso. Thus, the book’s owner could have drawn the heads to indicate where to identify these specific remedies. Having analysed the extralinguistic elements, let us move on to the fol- lowing where I analyse the linguistic strategies which assist in the compila - tion of recipes in MS Hunter 185. 5.2. LINGUISTIC ELEMENTS A first glance at the organisation of recipes in MS Hunter 185 reveals a the- matic organisation which arranges recipes from head to toe, in an attempt to establish some order to what apparently owns none. Thus, the scribe offers remedies for the head in the first place, followed by remedies for the eyes, the nose, etc., and groups them according to their therapeutic function. This grouping, in turn, facilitates the consultation of the material in a short amount of time, since the reader can follow those lexical items which refer to the parts of the body in a descendant order. The ‘from-head-to-toe’ technique shows the organisation of the informa - tion by contents, but sometimes the lack of tabula to indicate where to find specific information related to a part icular sickness rendered the searching task a difficult one. It should be noted at this point the practical nature of these books normally resorted to in order to find urgently therapeutic solu - tions to cast out states of sickness. However, this type of organisation is, to a certain extent, arbitrary, and scribes very often ignored it in their own books. This irregularity in the or- F rancisco A lonso 176 ganisation lies in the fact that scribes inserted recipes, say, for prognostica- tion in between remedies for scabbed hands and headache, as seen below: (7) For wertes in mannes hondes or wommanes hondes o per in o per places • Anoynte pe wertes whereuere pei be… For to wite ¥if a man schal lyue or dy¥e pat is jwounded Take pe jus of … … For pe hedache Tak & seep verueyne & beteyne & … (MS Hunter 185, ff. 24v-25v, ll. 1650-1683) In relation to the thematic organisation, thematic subgroups can be also found. These subgroups (boxes in chart 1) are characterised by the presence of referential devices, namely co-reference (represented in chart 1 by means of a double -sided arrow) and cross-reference (represented in chart 1 by means of a dotted double -sided arrow). The former applies to those devices which establish a connection among the different recipes of a same group. The latter refers to those elements which connect recipes which may even be- long to different groups. These elements are described below. 5.2.1. RECIPE GROUPS As indicated above, recipes are normally grouped thematically, so that remedies for a specific sickness are gathered together. These recipes are tightly linked by means of cohesive ties, mainly anaphora, substitution and ellipsis as can be seen below. Co-referential ties in recipe groups are given in the title of the recipes, and the understanding of the contents of the second and following recipes L inguistic evidence in the compiling of Middle E nglish medical recipes 177 depends on the first recipe in a given group, both in the case of anaphora and ellipsis. The following instance exemplifies this aspect1: (8) For pe wormes in a mannes body pat men clepyth lubrikes Tak beteyne & sauen & dry¥e hem & make poudere of hem & let pe seke drynke perof in hoot water & also sone as pe poudere is doune in to pe body it wil slee pe wormes & brynge hem oute sykerly For pe same Tak syngrene & fe perfoye * & myntes & stampe hem & ¥ef hem drynke For pe same Tak & drynke stancrop * with wormot (9) For to sle trenchis in a man nes body Tak fayre bakoun & diseit smale morcelles pe mountance of a litel messe & tak a good handful of beteyne & wassch it clene & grynde it smale & take pe potel of gotes melke & do it in a fayre vessel & do perto py bakoun & py beteyne • let hem s epe to gedre til py bakoun be j • now¥ & let pe seke ete pat mete warme & he schal be hool certeyne For pe same Tak a saucer * ful of jus of calamynte & as muche of Centorie & of hony & of vynegre ana & pe ferpe part of a sponeful of salt & medle al pys to gedre & loke pat pey be ri¥t salt & boyle hem a litel ouer pe fuyre & do hem in a clene vessel & let pe seke vse perof fastyng a saucerful at ones & pe same day at vnderne & an o per at none & he pat schal vse pis medi cyne let hym be fastyng til mydouere noon & he schal be hool (MS Hunter 185, ff. 40v-41r, ll. 2377-2405) 1 From now on, I will include the following symbol * in the examples in order to in- dicate where a new recipe within a group begins. F rancisco A lonso 178 In the previous recipes (examples in (8) and (9)), we can see how they are connected thematically by means of the string for pe same. In order for the reader to understand the purpose of the recipe, he must go back to the first recipe in the group for reference. Thus, in (8), pe same refers back to the previous title in full For pe wormes in a mannes body pat men clepyth lu- brikes, which exemplifies substitution and ellipsis. Substitution consists here in the replacement of the entire meaning of the title in the first recipe by the word same, which is semantically identical, though the structure of the title of the first recipe is rather more complex than the pronominal string pe same, which is formed by a simple noun phrase. Ellipsis is understood here in the sense that the head of the noun phrase pe same lacks an element of the nomi- nal category such as the word sickness, or even the hyperonym thing. (9) also exemplifies a case of substitution and of ellipsis. In this sense, pe same stands for For to sle trenchis in a mannes body. The word same refers back to the entire title, and also the head of the noun phrase pe same is filled here by a pronoun, exactly in the same fashion as in (8). The fact that the recipes whose titles are For pe same needs the presence of a first title in full may be interpreted as their being less effective than the first one, though there is little evidence that the scribe intended so. Sometimes, the title of the second and the following recipes in a group includes the Latin word Item ‘the same’. The interpretation of this Latin ele - ment, which manifests a new case of substitution, is found both in the first recipe of the group and in the very same title where it is included: (10) Item for hym pat is scoldet on his pyntel pat is y clepid pe potegal § Tak a lynnen clo p pat is clene and wassche & brenne it & make powder perof & tak oyle of egges & anoynte pe sore & put pe powder in pe holes when pei bep anoynted & pat schal helpe hem & make hem hool § (MS Hunter 185, ff. 42r-42v, ll. 2471-2477) In this example, the Latin word Item refers back to line 2449 in the manu- script For scoldyng of a mannes pyntel. In addition, in this case, the L inguistic evidence in the compiling of Middle E nglish medical recipes 179 reference for the word item is also found later in the same title, as the purpose of the recipe is repeated by means of a relative clause pat is scoldet on his pyntel. Thus, the reader does not have to go back into the text for reference. Many other recipes are introduced by means of the word also , which, in addition to representing an example of economy in language, it also stands as another case of substitution, and to some extent, of ellipsis, as can be seen in the following example: (11) For pe emeraudes so pat pei haue none grete pappes § Tak welle cresses & stampe hem & steue hem in • a pot with oute water & make a plaster & ley to pe emeraudes § Also for pe same * Tak litarge of golde & make powdere perof & tempre hit vp with oyle of violet & make a plaster & ley perto Also for pe same / * Tak moleyne • & styue hit in good red wyn & make a plaster & ley perto § Also * Tak Rewe & louache of eyper j lyke me che & grynde hem in a mortere & put perto hony & fry ¥e hem to gedre & mak a plaster & ley per to as hoot as he may suffre & alle pese medicynes bep gode for pe emeraudes (MS Hunter 185, ff. 46r, ll. 2639-2654) In the previous recipes, the word also might be well replaced by the string “another medicine” or “another remedy”. In the second and third recipes in (11), also is followed by for pe same which, as seen above, stands for information given in the title of the first recipe For pe emeraudes so pat pei haue none grete pappes. In the last recipe in this group, the title also is an example of substitution and ellipsis. Substitution here lies in the fact that it stands for the whole title in the first recipe, and the element which is elliptical is for pe same, whose meaning is included in also . The anaphoric pronoun another is also part of some of the titles of the recipes included in MS Hunter 185. The following example illustrates the anaphoric use of another in our recipes: F rancisco A lonso 180 (12) For to make a buyle drawe to gedre and brek § Tak galbanum & clense it at a fuyre & make per of a plaster & ley per to ij dayes or it be remoued Also * an other for pe same Tak a rosted ox non • & pe lylye rote & souredokkes & stampe hem to gedre with barewes grece & with oyle of olyue & make a plaster & ley to pe sore (MS Hunter 185, f. 23r, ll. 1594-1601) In this example, an other refers to a new recipe which is good for to make a buyle1 drawe to gedre and brek , thus linking thematically the second recipe in the group to the first one. There is also the elision of the word medicine which normally collocates with the word another in the occurrences of the former in MS Hunter 185, as seen in lines 1581-1582 Also an oper medicyne which constitutes the first instance of the words and in the manuscript. So far, the use of substitution, ellipsis, and anaphora have been described, especially those cases which serve the function of creating thematic groups of recipes within the receptarium, thus forming a clearer structure which gives shape to the remedybook. Below, I shall describe the use of other dis course particles which link recipes belonging to different groups, mainly by means of cross-references (represented in chart 1 by means of a dotted double-sided arrow), and which also favour cohesion within the recipe book. 5.2.2. LINKING THE RECIPE GROUPS We have seen previously how recipes are tightly linked in groups by means of anaphora, ellipsis, and substitution. These devices show a high degree of frequency. However, linking particles, which establish connection among the different groups of recipes, are less frequent; though, admittedly, they still 1 ‘swelling’. L inguistic evidence in the compiling of Middle E nglish medical recipes 181 play a significant role in the process of compilation of recipes. These particles constitute evidence of a serious attempt to perform editorial work by the scribe. In other words, the presence of these particles shows that the scribe knew, and could handle, the material he was compiling to such an extent that he was able to establish cross-references among recipes 1. Beforesaid (ll. 1338-1339) and its variants before j seid (l. 1413), jsayde aboue (l. 1593); and as it is seide be fore represent the majority of the internal cross-reference expressions employed by the scribe to make reference to re - cipes given earlier in the receptarium. The following instances exemplify this point2: (13) For pe sa me a plastyr Tak smalache wermot hony salt vynegre ana a saucerful & perto a quancyte of Ry ¥emele & boyle hem to gedre & make a plastyr on a clo p & ley to pe sore & vse pis plaster & pe drynke be foresaid & pou schalt be hool • (MS Hunter 185, ff. 17r-17v, ll. 1333-1339) (14) Also for pe felon pat makep the to swelle Take beteyne & camamylle heyhoue & egremoyne • ana an hanful & wessche hem & stampe hem wel • & take hertes grece & lyue hony of eyper a quantite & do perto barly mele half a quartron & frye hem wel to gedre • & let schaue pe hed & ley pe plaster a boue • as • hot as pe seke may suffre • & ¥ef hem to drynke plan 1 Traditionally, receptaria are considered to be characterised by adaptation and accre- tion (Voigts (1982: 44), Norri (1992: 34), Taavitsainen (1992: 329)); the latter im- plies some sort of pilling up of recipes as mere records of therapeutic knowledge. Thus, the presence of these internal cross-references support the idea of an incipient editorial work by the scribe who carefully connects recipes in our manuscript. Also, by means of these cross-references the scribe avoids the repetition of already -stated information such as how to mix the products in order to produce the medicine. 2 Italics highlights the examples I refer to in the text. F rancisco A lonso 182 tayne pe same drynke pat is jsayde a boue for pe hed ache (MS Hunter 185, f. 23r, ll. 1584-1594) (15) Also an other for pe same Tak pullyole de montayne pat is to say hillwort or brodewort a good handful & wessch hit clene & schere hit smal & do it in a mortere • & grynde hit smal & do perto half an vnce of poudere of peper & an vnce of poudere of comyn & med le hem to gedere haluendel be wasted & do perto also a potel of good wyn & penne • seep it as it ys seyde to fore & let pe pacient vse pis after mete & not before & but ones after none & at eue last & algate hoot & he schal be hool (MS Hunter 185, ff. 33v-34r, ll. 2062-2074) (16) Also for pe emaraudes a good medicyne / • Tak a stool with a sege as hit is saide before for pe fluxe & ri¥t so vse pis medicyne for pe emaraudes as of pe hote tiles per he take p frank encens he schal take poudere of myrre & of encense • ana & ¥ef he haue grete pappes bynde a boute pe pappes a rede selkys pred faste bi pe grounde & kerue a wey pe pappes a boue pe prede • & pus serue hym ry ¥t boldelich & tak an • handful of wermot & wassch hit & stampe it & fry¥e hit with oyle of olyue & make a plaster & ley perto as hoot as he may suffre • & a noon as he be on his bed & a morwe vse pis stool with pe sege & with pese powdres & with jnne iij dayes he schal be hool on warantise bote loke pat he haue pis plaster durynge iij ny¥tes or iiij at pe moste • (MS Hunter 185, ff. 45v-46r, ll. 2620-2639) L inguistic evidence in the compiling of Middle E nglish medical recipes 183 In (13), the scribe offers a remedy for the fistula; this remedy is the sec- ond in his group and consists in the making of a plaster to cure the sick. In this case, this treatment must be applied together with a drink described in the previous recipe in the manuscript. Instead of rewriting the making of the drink, the scribe uses the string pe drynke beforesaid which forces the reader to go back in the receptarium to find the information he needs. This time the location of such information represents an easier enterprise, since the reader looking for information on fistula might have read the previous recipe as well, since it is located just before it. However, two related recipes may sometimes be well aprt, as can be seen in (14) which contains information on swellings. Here, the scribe connects this recipe with another found in lines 1573 to 15751 by means of the string pe same drynke pat is sayde aboue for pe hed ache. As one may suppose, the expression is sayde aboue is rather unspecific for any reader to find success- fully the drink the scribe refers to. Thus, the scribe adds the complement for pe hed ache in order to make clear the drink that should be prepared. (15) seep as it ys seyde to fore finds its referent in the previous recipe in the manuscript. The reader is told to perform the same action with the ingre - dients as in the previous recipe, that is, the ingredients must be soaked in “good wyn from a potel” (l. 2057). Finally, (16) shows a similar use of cross-reference expressions, where the reader must look back into the text in order to find the referent. In this case, the reader is told to use the same type of chair with a hole (normally used to help pregnant women when they give birth) which is described in a previous recipe in lines 2247-2249: “Tak a stool with a sege pat is closed al a bowte & tak an hoot tyle pat is glowyng hoot & ley hit vnder pe sege & let pe seke sitte doun on pe sege & make pe sege on pis manere”. This recipe is found nearly 400 lines earlier in the receptarium, which indicates that the reader must locate the information without any other aid, save for the title in rubrics in line 2246 Also for pe fluxe. 1 Lines 1573-1575 reads: “Also here is for pe felon Tak & drynk plantayne & make powdere of pe same & ley per to.” F rancisco A lonso 184 Sometimes, the reader is given more precise indications of where to find specific information: (17) For to make a drynke for alle manere of feuer oper posteme & for alle sekenesse pat is in mannes body pat euer may be holpe with eny medicyne Tak euery day a quantite of pe powdere in pe recet be fore pat makep mencyoun of pe dropsie in pe wombe & tak jsope Rose maryne • violet verueyne • Beteyne herbe jon • Monsere Red planteyne pe whiche is a powdere anance Sauge ffeperfoye • & ache ana an • handful & was sche clene pyn herbes & do hem in to a mortere • & stampe hem a litel • & do hem in to an erpen pot … (MS Hunter 185, ff. 54r-54v, ll. 3029-3051) In this example, the scribe indicates the reader where to find the informa - tion to produce the medicine to help the sick to overcome a state of fever. As happened in some of the examples described above, the recipe the scribe refers to is found very early in the text. Thus, the scribe gives specific details to the reader so that he can find quickly the information required. In fact, the extra information given here pe dropsie in pe wombe coincides word -for-word with the rubricated title of the recipe the scribe mentions: “For pe dropsie in pe wombe & in pe feet…” (ll. 2559-2561). This may greatly benefit the reader in his search for this information. In MS Hunter 185, the reader must also look forward, rather than back- wards, into the text to find the information required. The only instance of this type is the following: (18) For to restore a¥eyn pe wombe whenne it is solible § Tak an handful of hennecresses seed anoper of weybrode & gryn de hem in a mortere penne frye hem wel to gedre with schepes talwe & frank encens L inguistic evidence in the compiling of Middle E nglish medical recipes 185 & make a plaster & ley to his nauel also hot as he may suffre • & let hym vse pis pat is jtolde next after (MS Hunter 185, f. 37v, ll. 2239-2246) In this case, the string pat is jtolde next after compels the reader to go forward in the receptarium to find the second part of the treatment for the womb which is presented in the following recipe: “Also for pe fluxe” (l. 2246). 6. CONCLUSIONS The present study illustrates the strategies used by the medieval scribe of MS Hunter 185 to connect medical recipes in a manuscript. Such strategies in - clude the use of specific visual elements, and the use of grammatical devices such as anaphora, substitution, ellipsis, and cross-references. As shown above, one of the aims of linking recipes is to create subgroups of recipes which cover particular information for specific sicknesses. Thus, the scribe employs systematically the expressions also , an other, item, for pe same in the titles so as to group recipes with a similar therapeutic function. Also, as seen earlier, by using such expressions in the second and following recipes in a given group, the scribe only needs to write in full the first title of the first recipe in that group; in this way, he avoids the repetition of the entire title in subsequent recipes. The use of internal cross-references, such as bifore saide, also sheds light to the process of compilation of recipes in the Middle Ages. These cross-references show that scribal work in the construction of the medical receptarium was far beyond the random compilation of recipes. Quite on the contrary, cross-references constitute an indication that the scribe knew his material well, and hence that he was able t o establish connections among the recipes. It should be also noted that the scribe supplies extra -information in strings like as it is seide be fore or before j seid by adding phrases such as for pe fluxe or for pe hed ache, which clearly define the recipe he refers to. The inclusion of such phrases depends on the relation of proximity between the recipes involved: the larger the distance between recipes, the longer the cross-reference. F rancisco A lonso 186 The information included in this paper represents just the starting point towards further research in the field of linking particles in Middle English recipe collections. The use of specific lexical items with linking function among recipes has been ignored here, but constitutes another area of research which might enlighten our understanding of medieval recipe compilations. Francisco Alonso Almeida Universidad de Las Palmas de Gran Canaria REFERENCES Alonso Almeida, F. forthcoming: ‘Gyf hyr pis medycyin’ Analysing the Middle-English Recipe Medical Discourse. Revista de lenguas para fines específicos 5/6. Brown, G. and G. Yule. 1996 (1983): Discourse Analysis. (Cambridge Textbooks in Linguistics). Cambridge University Press, Cambridge. Norri, J. 1992: Names of Sicknesses in English, 1400-1550: An Exploration of the Lexical Field. Suomalainen Tiedeakatemia, Helsinki. Parkes, M. 1973: The Literacy of the Laity. In David Daiches and A. Thorlby (eds.) The Mediaeval World. Aldus Book, London. 555-557. Robbins, R. H. 1970: Medical Manuscripts in Middle English. Speculum, 45: 393-415. Rodríguez Álvarez, A. 1997: Documentos notariales vernáculos del condado de Durham (Siglo XV): estudio y edición. Servicio de Publicaciones de la Universidad de Córdoba y Servicio de Publicaciones de la Universidad de Las Palmas de Gran Canaria, Córdoba. Schmidt, K. A. R. 1994: The Index of Middle English Prose and Late Medieval English Recipes. English Studies, 75.5: 423-429. Taavitsainen, I. 1994: On the Evolution of Scientific Writings from 1375 to 1675: Repertoire of Emotive Features. In F. Fernández et al. (eds.) Papers from the 7th International Conference of English Historical Linguistics. John Benjamins Publishing Co, Amsterdam and Philadelphia. 329-342. L inguistic evidence in the compiling of Middle E nglish medical recipes 187 Voigts, L. 1982: Editing Middle English Medical Texts: Needs and Issues. In Levere, Trevor Editing Texts in the History of Science and Medicine. Garland, New York and London, 39-68. Voigts, L. 1995: Multitudes of Middle English Medical Manuscripts, or the Englishing of Science and Medicine. In M. Schleissner (ed.) Manuscript Sources of Medieval Medicine. A Book of Essays. Garland Publishing, New York and London. 183-195. Young, J. and P. H. Aitken. 1908: Annotated Catalogue of Manuscripts in the Hunterian Library. James Maclehose and Sons, Glasgow. * † *