Scepticism, Critique, and the Art of Writing: Preliminary Considerations on the Question of Textual Authority in Medieval Peshaṭ Exegesis

 ‘The glossed book’s lay-out [...]’ stems from Mary Carruthers, The Book of Memory: A Study of Memory in Medieval Culture (Cambridge: Cambridge University Press, 2008), 265; ‘Glosses not only use a variety of strategies [...]’ is quoted from Suzanne Reynolds, Medieval Reading: Grammar, Rhetoric, and the Classical Text (Cambridge: Cambridge University Press, 1996), 7.  Cf. the definition from the Maimonides Centre’s mission statement: ‘One of the most important research objectives is to examine whether the method of enquiry implied in the term ‘scepticism’ could be regarded an anthropological constant [...] across both Eastern and Western philosophy and culture. The assumption would be that people in every culture express doubts about the claims of authorities to truth, the reliability of texts, and their social or mystical relevance, doubts about the power and presence of divinities, or about the power of reason and the controlling power of social structures and their respective consequences [...].’ (https://www.maimonides-centre.uni-hamburg. de; accessed March 2018).  A term introduced by Carruthers, Book of Memory, 243.  This paper is based on my article “‘Daneben steht immer ein kluger Kopf ’: Die Glossenformationen im Codex Wien hebr. 220,” in Diligens scrutator sacri eloquii. Beiträge zur Exegeseund Theologiegeschichte des Mittelalters. Festgabe für Rainer Berndt SJ, eds. Hanns Peter Neuheuser, Ralf M. W. Stammberger, and Matthias M. Tischler (Münster: Aschendorff, 2016): 53–85, and was reworked under the respective topic during my stay as Senior Fellow at the Maimonides Centre for Advanced Studies—Jewish Scepticism, Universität Hamburg (October 2017–March 2018). My thanks to Jonas Leipziger who arranged, in particular, the edition of the gloss material in this paper, Bettina Burghardt who translated parts of this paper from German into English. Thanks also to all the staff mem-

In its specific philosophicals ense, sourced in Ancient Greek philosophy, scepticism and/or the sceptical approach seek to refrain from anyo fi ts own judgements regardingreligious beliefs or philosophicalviews. In that sense, the question is easily declined:There is no philosophical scepticism among the tosafists, for the tosafists wereneither philosophers, nor did their critical glosses on the Bible and the Talmud (more precisely: their critical notes and questions on Rashiʼsc ommentaries on the Bible and the Talmud) intend reachingastate of mind beyond anyaffirmation or denial of religious knowledge or beliefs.However,evenifwetake 'scepticismʼ in an extended sense, comprised of sceptical and critical strategies,c oncepts, and attitudes against elements of tradition,⁵ we would hardlyl abel the tosafistsʼ exegetical discourses as 'scepticism' or sceptical thought,although they challengetraditionalrabbinic exegesis, and in some cases utter harsh criticism of the Rishonim as theire xegetical forerunners.For instance, in Ms Oxford, Bodleian Library,Opp. 34 (Neubauer 186), fol. 116v,⁶ we find various exegetical dicta attributed to Rashbam⁷ describingthe new exegetical approach 'accordingtothe ways of the world',thus criticising the former exegetical explanations for being 'neither in accordancew ith the wayo ft he world, [based on] common knowledge,n or in line with the [meaning of]t he verse.'⁸ Moreover,i nt he samec ontext,w ef ind harsh criticism of his grandfather Rashi whose explanation of Deuteronomy2 0:19 is rejected, almosts coffed at,a s ' foolishness.'⁹ However,w hen dealing with medieval Jewishl iterature, one always has to wrestle with the question of textual authority,a nd so we mayaccede the dictum conveys an undermining of the exegetical authority.But Iwould like to propose the question: Is textual authority identical with exegetical authority?¹⁰ And if so, must it be labelled as ceptical mode or strategy, even in av ery broad sense?A re these exegetical comments am eans for conveying sceptical thought? Is ac ritical attitude towardsr abbinicu nderstanding tantamount to as ceptical approach?A nd  last,but not least: in which wayhave 'sceptical' strategies,concepts, and attitudes to be qualified in relation to the literaryg enre in which they appear?
In the following studyo nt he exegetical glosses in Ms Vienna,Ö sterreichische Nationalbibliothek, cod. hebr.2 20,Iwill present an edition of selected glosses¹¹ and examine, if, and in what manner, these gloss comments conveyn egative criticism of RashiʼsT orah commentary,a nd how and in which wayt hey challenget he authority of the great sage. As some of these glosses have alreadyb een attributed to Rashbam, my studyw ill comparet hem to the printed editions of Rashbamʼs Torahc ommentary (henceforth RTC).¹²

2T he Manuscript Tradition of Rashi and hisS chool 2.1 TextualW itnesses of RashiʼsT orahC ommentary
The so-called peshaṭ exegesisb yN orthern French Jews of the eleventh and twelfth centuries,and the question of how exactlyitistobeunderstood, has engagedscholars of Jewish Studies almost continuously for the last 150 years. The main focal point has been-and remains-the commentary by R. Shelomo Yiṣḥaqi (acronym Rashi;ca. 1040-1105, for even to this day, it is considered the fundamental text of Jewish Bible exegesis.¹³ His commentary is present in each and every traditionalBible edition. Regarding manuscript tradition and contemporary printeditions, Jewish Medieval studies are in am uch less comfortable position than their Christian counterparts.¹⁴ This has became even more so in recent years, when the relationship between the literary  An edition of the entiregloss corpus is currentlybeingprepared. All the glosses will be diplomaticallye dited, and presented as an annotated digital edition according to their mise-en-forme and arrangement on the manuscript'sfolio. Iamdeeplyindebted to Clemens Liedtke,M.A., Center for Jewish Studies Heidelberg(Project Corpus Masoreticum; accessed March2018) for providingmewith the electronic tools for the digital edition.  See below section 2.3.  Amongo thers see for instanceG ilbertD ahan, Gérard Nahon, and Elie Nicolas,e ds., Scepticism,C ritique, and the Arto fW riting heritageo fR ashi and the documentary evidence of his pupils came underc loser scrutiny. It was revealed that the Rashi commentary never reallye xisted,¹⁵ at least not before it made its wayi nto the typographic world. The first printing of Rashi's commentary on the Toraha lso marks the beginning of Hebrew book printing per se (Reggio di Calabria, 1475).¹⁶ Even the handwritten text-witnesses existing todayo riginatef rom an oticeably later period and are everything but uniform: the oldest manuscript,M sMunich, Bayerische Staatsbibliothek, cod. hebr.5(commentary only) was copied in 1233 in Würzburga nd also constitutes the oldest dated, illuminated Ashkenazim anuscript. The second important Rashi manuscript,M sL eipzig, Universitätsbibliothek, hebr.1 (B.H. fol. 1) originates from the thirteenth century (undated) and contains,a side from the Masoretic biblical text (with eclectic annotated Masoretic glosses), the Targumand acommentary attributedtoRashi and is alsoaccompanied by anumber of glosses by Rashi'smost eminent pupil, Shemaʽya, and the scribe Makhir.¹⁷ The third manuscript,c onsidered to be one of the most importanto bjects for research into Rashi, Ms Vienna,hebr.220 (Schwarz 23)isalso undated, but originates from aconsiderably later period (thirteenth/fourteenth century). Strictlys peaking,t he printed editions are unusable for academic purposes: aside from the traditionalB ible, editions in the so-called Rabbinic Bibles (Miqra'ot Gedolot)a re severelyf lawed. Today one resorts to either Rashi'sT orah commentary in the Abraham Berliner¹⁸ edition or the one contained in Miqra'ot Gedolot Haketer.¹⁹ There is as yet, no critical edition available, and of the existing Rashi commentaries,t he deviations are occasionally overlye xtreme. Such findingsr esulted in numerous discussions among scholars on whether to describet he Rashi corpus as 'author commentaryʼ or 'compilatory commentary'.The discussions, however,werelargely too narrowly focussed (providing basically no result Medieval Latin²¹ studies alreadydealing in detail with literary theory in medieval literature, gloss hermeneutics, and medieval readingt heories werei nsufficientlyi ntegratedi nto studies of Hebrew material. However,anumber of important published studies conclude that neither the romantic notion of the 'authorʼ as found in the nineteenth century nor the battle cry of the 'death of the author' based on ao nesided, exaggerated reception of Foucault'st heoriesa re the onlya lternatives.²² I have demonstrated elsewheret hat Rashi'sc ommentaries are better considered the first 'JewishG lossa Ordinaria',a nd in terms of their ambition,b ear comparison with Gilbert of Poitiers' (ca. 1080-1154) Media Glossatura or the Magna Glossatura compiled by Petrus Lombardus (ca. 1100-1160).²³ Rashi collects the most fitting Midrashim to highlightt he peshaṭ of av erse or word, while simultaneouslyi ntroducing new exegetical and grammatical approachesand insights by intertwining them with classicalm aterial well known from the rabbinic texts.

Tosafist, Copyist, Writer,o rA uthor
The problem of af luctuatingt radition, and the question of whether one single 'author' (if so, which one?)may have been the guiding spirit of the NorthernF rench exegetic tradition is not confined onlyt ot he Rashi tradition, but alsoi ncludes his school, i.e. the Bible commentators and tosafists he himself trained and bore connection to his Beit Midrash.A side from his acolyte and chronicler R. Shemaʻya (ca. 1060Shemaʻya (ca. -1130  ReconstructingR .S hema'ya'sb iography is beyond our ability,b ut he is known as the one who first disseminated Rashi'sliterary legacy.Rashi mentions him by name in his commentaries on Genesis 35:16and Ezekiel 42:11(source: Miqraʼot Gedolot Haketer). It is unclearifand to what degree they werer elated,byblood or by marriage.A ccording to Grossman, he probablyw as some sort of 'assistant' to Rashi: he edited his commentaries,augmented them with his own glosses (MsLeipzig,hebr.1 alone contains morethan 250ofShema'ya'sglosses) and probablyalso influenced Rashi to no small degree on halakhic matters; cf. Grossman, Early Sages,e sp. 43-45;132-133.
Abraham Geiger had alreadyreferredtoR.Yosef Qara as aglossator.²⁷ According to Geiger,R.Yosef commented less on the biblical text itself than on Rashi'scommentary.Geiger relied on handwritten text-witnesses referringt oR .Yosef Qara as ‫קיתעמ‬ ( ' copyist')o r ‫בתוכ‬ ( ' scribe').²⁸ Difficulties in the debate arise due to manyo ft he Hebrew manuscriptsnot labelling the various tosafists consistently, thus far no criteria have been established enablingaccurate attribution of text authorship and precisely its authorial intention. Terminological distinctions between scriptor, compilator, commentator and auctor,a sk nown from Latin medieval contexts,²⁹ cannot simplyb e transferred to Hebrew settings, and the relevant manuscripts are at least 130 years more recent thant he tosafists mentioned.
Attribution of exegetic commentaries to specific tosafists is also problematic as existing text-witnesses specify the glosses differentlyt oo ne another.T here exist glosses labelled with the originator'sn ame ('R. Yosef'; 'R. Shemu'el'), and those without attributiona nd glosses that are similar to others yetb ear differing names. With such the written evidence hasty attributions are ill-advised. The online resource AlhaTorah cites ac omment on Genesis 41:7 in Ms Vienna, hebr. 220a ttributed to R. Yosef Qara'sT orah commentary³⁰ and cross-references it to commentaries by Rashbam, Ḥizzequni, and Rashi (MsL eipzig, hebr.1), and they are, indeed, very similar to one another and very close to R. Yosef Bekhor Shor'sc omment (twelfth century; Orléans).
Before an exegetic attribution to one specific exegete is attempted and his approach towards former(rabbinic) commentaries outlined,the gloss-inventory should be assessed and evaluated for each individual manuscript.The debates over attribu-  tion thatu sed to focus on as ingle commentator and/or his school can now be extended to incorporate information on the geo-culturalb ackground of as pecific manuscript and its scribalt radition. In addition, one can prevent the prematurea ttribution of agloss on one single Bible verse. Not all exegetic glosses relate the comment unambiguously, via aprefixed lemma or other paratextual elements (e.g.circles or ornaments), to one, and onlyo ne, Bible verse. The reason for this mayb et hat some glosses do referd irectlyt ot he biblical text,while others relatet ot he already existing commentary as theirh ypotext.A ccordingt oS uzanne Reynolds, it is the teacher,the grammaticus,who 'reads' for the others, the pupils. In that,the glosses turn into 'written traces of am uch fuller readingp ractice. ' Scepticism, Critique, and the Art of Writing fortunately, the manuscript of RashbamʼsT orah commentary is the onlym anuscript in existence. It found its wayi nto the BreslauS eminary by wayoft he Mendelssohn and Fraenckel families and vanished as ar esultoft he Nazi persecutions. The Rosin edition was re-edited and annotatedi n2 009 by Martin Lockshin.³⁵ In 1985, SaraJ aphet and Robert Salters discussed the literaryform of Rashbamʼs commentary,a nd concluded with aq ualitative distinction between 'glossary' and 'well-structured, premeditated composition'³⁶ In their introduction to the Qohelet commentary,Japhet/Salterstook up aposition contrary to Rosin'sassumption, arguing that Rashbam'scommentary was 'by no means aglossary,' but a 'well-structured, premeditated composition, the writing of which is guided by al iterary insight into the book of Qoheleth.'³⁷ In distinguishingbetween agloss and 'acontinuous and fluent presentation, comprisingc ompletes entences and written in ab rief and concise idiom,'³⁸ Japhet/Salters (at least implicitly)c haracterised the 'act of glossing' as inferior to the 'act of writing/composing.' Japhet/Salters also attempted the same claim for Rashbam'sT orah commentary but had to acknowledge the dearth of available manuscripts-ad earth far worse todaythan that of Rosin'speriod, as the very manuscript he used to preparehis edition no longer exists. Similarly, Elazar Touitou rejects Rosin'sopinion with the argument that it is hardlyconceivable that the spacebetween the lines or in the margins would allow for ac ommentary of such proportions.³⁹ He added that one would also expectacommentary noted in the margins to have been intended for privateperusal and not publication.
The debate appears to have gone down ablind alley.Asaresult, this study proposes accessing the glosses differentlyand focussing on external parameters. To determine the function of ag loss text it seems reasonable to not onlyc omparet he glosses in am anuscript to othert exts in terms of their semantic content,b ut also to make the most thorough going investigations possibleo ft heir placement on the page, their shape and form and as ar esult, make theirc ommunicative function become apparent.A ccessing the exegetic glosses via the manuscriptʼsl ayout, mise-enpage, mise-en-texte,a nd the direction of the writing can also focus attentiont od etails, hitherto unnoticed, concerning the relationship between the gloss and the hypotext.
The manuscript is heavilyglossed by several hands, especiallyonthe first 25 folios (i.e. in the Book of Genesis/Bereshit;t he Book of Exodus/Shemot begins on fol. 26r).Atleast two groups of glosses, written in two hands, can be recognised: the first group GL 1 ,which, accordingtoSchwarz, originates, from the first scribe (upon which we will focus subsequently) and asecond,GL 2 ,which continues beyond fol. 22vand glosses intenselyb eginning on fol. 25.GL 2 in particular,o riginatesf rom ad istinctly later period; written in Italian cursive script indicates the late fifteenth and sixteenth century.The hypotext commented on is thus not the biblical text itself, but the recension of Rashi'scommentary.Asaresult, from the outset,the glosses can have several functions: They mayr e-explain the biblical text,l argely independentlyo fR ashi, or perhaps expand upon, modify,o rd isproveR ashi'sc ommentary.
GL 1 was written with adifferent pen, but the same ink as the main text,allowing one to assume that the manuscriptʼss cribe (Menaḥem) is also responsible for the glosses.The script indicates the time of the manuscript'scopying,and in amarginal gloss on fol. 152v,the scribe ponders the fact that,inhis Vorlage the tribe of Gad was missing in the listing of tribes: .The glosses from GL 1 can be attributed to different (groups of)a uthors:⁴⁴ )a s scribe, but leaveso pen whether other scribes contributed. However,d ifferent hands (in particular, Avraham over-elongates the Hebrewl etters Quf and Nun)a sw ella sachange in ink aree asilyo bserved (cf. the transit fromf ol. 22vt o2 3r). In addition,b eginningw ith fol. 30rv,f antastical animal-creatures, depicted in the same ink used for the main text,start showing up. Finally, the peculiar distribution of the glosses (cf. the followingt able) points to ac hange of scribe.  With the exception of the transitional pages, fol. 20v, 21r, which are arranged in one or two columns,r esp.  Onlyt he beginning of the weekly Torahp ortions is highlighted.  Unfortunately, some pages arecut off at the bottom, makingitimpossible to discern ascriptions. Furthermore, an entire array of gloss information in group GL 1 does not provide an ascription (wew ill see, however,thata tl east some implya na scription!). Moreover, one further group was prevented from giving us an ascription because the last and crucial line at the bottom or lateralm argin has been cuto ff.⁵² The Torahs egment, beginning on fol. 23r, shows onlyf aint traces of the later glossator'si nk (GL 2 ).
Elazar Touitou and myself have alreadyedited some of the glosses attributed to one 'R. Shemu'el' (this is all we have with which to identify him) and analysed their relationship to Rashbam'sc ommentary on the Torah.⁵³ While Touitou acknowledges an attribution to Rashbam as secure, Ifeel this leavessome serious questions unanswered raising fresh problems.T hey pertain, in particular,t ot he relationship between these glosses and the RTC.⁵⁴ It is important to state thatt he glosses have onlyb een evaluated thus far in terms of theire xegetic statements and not in light of their placement in the manuscript or outward appearance. Such an extension of the questions concerning the glosses,( bearing in mind that the commentaries could well have been used in as cholastic context), mayp rovide importantr esults, which, in turn, could assist in solving problems of the glosses' authorship or, more generally, how peshaṭ comments werea ctuallyi mplemented in the Middle Ages.
Indeed, it is notable that each gloss possesses aspecific mise-en-forme,evenifit is not quite consistentlyapplied. This finding has so farnot been taken into account. It is possiblethat it is nothing more than an arbitrary wayofembellishing the manuscript: The scribe did not wishing to write the glosses in the margins in a 'boring' manner.B ut the shape of the glosses mayb eac lue to the utilisation of the manuscript.T he glosses are certainlye ye-catching, since onlyt he smallest part appears as rectangles or simplyunarranged. Sometexts go around corners and are arranged at right angles;⁵⁵ 'balls' or 'heads',i .e.c irclesw ith or without directlya ttached texts;⁵⁶ sharplyp ointed triangles (mostly, but not always when the gloss is closed with ‫׳ת‬ tosefet/tosafot);⁵⁷ and combinations of circlesa nd triangles.⁵⁸ The glosses specificallya ttributedt oR .S hemu'el are mostlyi nt he shape of a 'head' (with or without 'shoulders'). Perhaps the glossator let the abbreviation ‫׳ר‬ ‫מש‬ ‫לאו‬ evolve via ‫ש״ר‬ into ‫שאר‬ ( ' head').⁵⁹ However,s ome glosses explicitlya ttributed to R. Shemu'el do not displayaheadʼsform, whereas some of the heads explicitlybearother signatures (e.g.R.Yosef).⁶⁰ Apagecut prevents some glosses to connect to aname, others remain anonymous. The subject of book 'forms' (including mise-en-texte and mise-en-page)has come in for renewed academic scrutinyrecently. Implementing text-anthropological means here requires more than simply identifying the commentaries' Sitzi mL eben but to consider the history of their materiality.B yu sing materiality in its broadest sense, we have alsot or egard the people who dealt with the artefacts/manuscripts as the original 'locus of the text'.⁶¹ Theg oal is not so much to write ah istory of the manuscript as to write the story of the people involved in it,byextrapolatingthe narrative contained in the materiality of the document.Sofar,this aspect has barelybeen considered in Judaic Medieval studies, in spite of its potential to bear fruitful results. The following sample descriptions of several glosses in Ms Vienna,h ebr.2 20,a re to be understood as the first step towardsabetter understanding of the exegetical glosses of the Northern French exegetics chool and their disputes with elder exegetical traditions. In order to relatea n' author' to these glosses, Iw illf ocus on some of the glosses explicitlya ttributed to R. Shemu'el, and present ad escription of the gloss, edition/translation, and as hortc omment.

Description
The first gloss explicitlyattributed to R. Shemu'el is found on fol. 8v,wherethe main text containingR ashi'sc ommentary on Genesis 20:16b egins. Thisg loss,written in the top margin abovet he middle column, is attributed in its first section to R. Shemu'el (  ‫׳ר‬  ‫מש‬  ‫ו‬ ). Furthermore, the gloss contains aF rench translation of the biblical lemma ‫למימ‬ ‫ל‬ ( who [would have] said)from Genesis 21:7.Asthis translation is located after the name ascription, it is not entirelyc lear whether it was part of the original comment.The glossw as centred abovet he second column, apparentlyc onsciously arranged into ac ircular form, a 'head' of sorts; this is clear from the abbreviation of the name ‫מש‬ ‫׳ו‬ (row 7) and the inclusion of as pace filler ‫יר‬ ‫׳‬ (row 8 ' Whos aid unto Abraham-Thisi sa ne xpression of praise and importance (in the sense of): "See, Who it is, and how great He is. He keeps (his) promise, he promises and performs!"' However,t he two comments-the glossa nd Rashi-are not connected with paratextual elements. This could be seen as ac lue thatt his comment was intended not so much as asuper-commentary on Rashi, but as an alternative comment on the biblical text.A nother clue pointing to this is the gloss being placed on top of the French translation of the biblical lemma ‫למימ‬ ‫ל‬ as aconcluding,almost categorically final, comment.

Edition and Translation
Whos aid to Abraham: This refers to [the previous verse] Everyone who hears will laugh with me (Genesis 21:6): Whobroughtmyfather the news and said, Aboy is born to you, and gavehim such joy (Jeremiah 20:15), YetIhaveb orne as on in his olda ge (Genesis 21:7

Explanation
The gloss (G_8v/1)t akes up the same lemma from Genesis 21:7 as in the Rashi commentary.I tbegins, using intertextual exegesis, by explaining the Hebrew expression ‫ימ‬ ‫למ‬ ‫לל‬ ‫םהרבא‬ as directlyconnected with the preceding text (Genesis 21:6b), according to which Sarah interprets the laughter in the sense of 'being laughed at.' This understandingisalso paraphrased in the last sentence of the gloss:Anyone who heard about the two old people'sbliss would make fun of such news. By doing so, the gloss stands in direct oppositiont oR ashi'sc ommentary in the main text body. Rashi justifies the usageofthe uncommon verb ‫למ‬ ‫ל‬ with the importance of the divine promise and Sarah'sa ppropriater eaction to this great deed of God.R TC argues similarly.⁶⁴ Therefore, bothR ashi and RTCunderstand the subject of the lemma ‫ימ‬ ‫למ‬ ‫ל‬ to be divine,⁶⁵ while our R. Shemu'el to whom the gloss is attributed assumesahuman subject who is unable to comprehend such amiracle and thereforejokes about it: Divine greatness (in the indicative mode) is in contrastwith the expression of human doubt (in the subjunctive mode). In its first part (lines 1-7) the gloss insistsonadecidedly different understandingasinthe Rashi commentary (and in the printed RTC) without explicitlyr efutingR ashiʼse xplanation.
In order to leave no doubt about the intended exegetic messageo ft he Hebrew expression, the gloss concludes with an Old French translation (pitron⁶⁶)that segues into another Hebrew summary.The fact that the Old French translation appears after the name ascription to R. Shemu'el leads one to conclude thatonlythe actual peshaṭ exegesis belongstoShemu'el'scommentary,and thatthe French gloss was added by the scribe. Interestingly,thereisanother Old French renderingfor ‫למ‬ ‫ל‬ in this context in the so-called Glossaire de Leipzig:⁶⁷ Drawingo nJ ob 8:2( ‫דע‬ ‫ןא‬ ‫מת‬ ‫לל‬ ), ‫למ‬ ‫ל‬ in this Glossaire it is translated as ‫ַפ‬ ‫ְר‬ ‫ַל‬ ‫א‬ ( parla;i nM odern French: 'affirma/déclara').⁶⁸ Irrespective of our scribe knowing this translation or not,a nd from wherever he obtained this translation, it certainlyintends to support the peshaṭ at hand. The explanatory gloss (pitron)i st he resulto fF rench culturalc ontacts,⁶⁹ indicating this manuscript wasstillused in aFrench-speaking and French-reading environment and auditorium. The French gloss does not simply support the peshaṭ exegesis, at the same time it demonstrates that here, the peshaṭ is guided by the protagonists' logic insinuated in the narrative.A st he gloss contrasts so vividlyw ith Rashi'sa nd RTC'se xplanations,o ne mayc onsider either the glossator (as an independently thinking 'head')p laced the explanation in its specific shape as as ign indicating he favours R. Shemu'el'sexplanation (not merelyfor ease of access), or thathewished, for the first time,top lace R. Shemu'el 'ahead' of his fellow scholars.
To sum up: by presentingacomment on the samel emmaa si nt he Rashi commentary yetn ot in the margins of aBible codex, but in the margins of aRashi commentary as its hypotext,the gloss not onlyadds asecond, alternative reading, but by means of the French pitron exposes Rashiʼsunderstanding of the verse as erroneous, and, thereby, undermines his exegetical authority.

Fol. 9v
3.2.1 Description On fol. 9v,three glosses are marked with the signature ‫׳ר‬ ‫מש‬ ‫לאו‬ .The first one (G_9v/1; 4lines)iswritten upside down in the topmargin. This one, too, displays avery specific form; its first two lines run preciselyabovethe middle and left columns, but its third and fourth onlya bove the left and,i na ccordancew ith the readingd irection, last columno nt his folio. The gloss is signed with the full name of R. Shemu'el. The biblical lemmaq uoted at the outset contains the beginningso ft he sentences of Genesis2 3:17-18. It is, therefore, related to the Rashi commentary thats tarts in the left column in row 30.A sinour latest example, there are no additional paratextual signs thatconnect the gloss to Rashi'scommentary.The commentary closes with as ummarisingr emark that encapsulates the whole explanation in one shorts entence.
G_9v/4 (starting with ‫יש‬ ‫נם‬ ‫א‬ )a nd G_9v/5 (starting with ‫יאוה‬ ‫חלש‬ ;d irectlyu nderneath G_9v/4)a re located in the bottom margin, below the left column, and display several peculiarities. Firstly, they are clearlylinked together.A tthe end of each line, G_9v/4 is shaped into al eft-a nd up-turning peak; they are joined by al ine from G_9v/5,m oving upwards and curving in the direction of the circle/head. Judging by the form, one would expect this to be as ingle gloss.H owever,the two lemmata are obviouslydistinct,having been takenfrom different verses and are introductions to twodifferent commentaries,each of which is expresslyascribed to ‫׳ר‬ ‫מש‬ ‫לאו‬ .Atany  [Wec annot say anything to you], either bad or good (Genesis 24:50). They were not [?] and they could not delay[ the matter], since the matter proceeded from YHWH (Genesis 24:50). We will call the girl (Genesis 24:57). Yous aid that your master was quites uret hat you would accomplish [your mission]. 'If she will go with that man, we will know that it is from YHWH (Genesis 24:50) and that all your words were right.' [Ane xplanation by] R. Shemu'el.
Scepticism,C ritique, and the Art of Writing

Explanation
G_9v/1 discusses abiblical turn of phrase that, at first glance, seems redundant: according to Genesis 23:17, apiece of land passed to Abraham 'as his possession' (in the presenceofthe Hittites), while Genesis 23:20states that it passed to him 'as aburial site.' The gloss explains that initially, the piece of land is bought by Abraham and passes into his possession (witnessed by the Hittites), but that it took another act 'from the Hittites,' i.e. their permission, to convert it into ab urial ground. This is also emphasised by the last sentence: onlyw ith the permission of surrounding real-estate owners can apiece of land be re-zoned as acemetery.Asaresult, R. Shemu'el'se xplanation agrees neither with Rashi's⁷¹ nor with RTC'sc omment.⁷² Rashi's commentary is not at issue in this gloss at all, as its primary concern, on the basis of Bereshit Rabbah 58:8,i st he use of the root ‫םוק‬ from Genesis 23:17t os ignify the 'elevation'⁷³ of apiece of land'sstatus when it passes into aking'shandsbyway of sale by ap rivatec itizen ( ‫דה‬ ‫וי‬ ‫ט‬ ). Rashi focusses on the changeover of the owner,a nd in addition exposes Genesis2 3:17 to be an incompletes entencet hat has to be supplemented by Genesis 23:18. RTCc laims, following the Babylonian Talmud, Qiddushin 27a, that Abraham'st ransaction was onlyb roughtt oafinal and irrevocable end after he put the piece of land to use by burying his wife. But this aspect is not at issue in the handwrittengloss,asindicated by the last sentence, which does not concern itself with the act of purchase but with the social context in which it occurred. G_9v/1 presents a peshaṭ explanation that harmonises the social conventions of the biblical narrative with contemporary regulations, i.e. an exegesis compliant with the sensus historicus,o rr ather,what the exegete believed it to be.The question still remains as to whyt he gloss was written upside down. Possibly, the writer wanted to make it obvious, even at first glance, thati nt his instance,apeshaṭ commentary is forwarded thatr uns contrary to the usual aggadic and halakhic explanation as givenb yR ashi and RTC. The argument is exposed rather indirectlya nd with the help of graphic means(mise-en-forme; mise-en-page). We maythereforecharacterise the glossatorʼs 'challengeo ft radition' af ight with closed visor. Contrary to Rashiʼsexplanation,⁷⁴ whereheargues that aperson enteringinto a covenant has to takeaholyobject into his hand,e.g., aT orah scroll or Tefillin, G_9v/ 4i nsists that 'placing under the thigh' has nothing to do with the bodyp art,b ut, rather,with the mannera nd degree of submission (son to father; servant to master). In addition,the gloss explains that biblical tradition knows of different ways to seal acovenant,and thatevenahandshake between parties was recognised. This line of argument fullya grees with the sensus historicus.
G_9v/5 expounds alemmathat is ignored by Rashi'scommentary.The relation to RTCisofparticularinterest in this instance, as G_9v/5 seems to elaborate on acommentary that RTCoffers as an explanation to Genesis 24:40-50,⁷⁵ accordingtowhich Abraham alreadyknew that they would succeed in their endeavour.G_9v/5 uses this literarycontext to ascribep rophetic qualities to Abraham,then extends this reasoning to the whole paragraph, Genesis2 4:7-57 (with the resultt hata ll its protagonists are considered to have prophetic ability!). Notably, this means G_9v/5 tells us that even the servant wase ndowed by God with prophetic powers,a sh ee stablished a sign that he alreadyk new Rebecca would enact.F urthermore, contrary to RTC'se xplanation, stating the familyagreed to the terms rather hesitantly, here they are portrayed taking am uch more positive stance for Abraham's( prophetic) confidence helps to guarantee the servant'ss uccess. The 'art of narration' displayedi nG _9v/5 is similar to RTCi na sm uch as direct speech is interwoven with the commentary, turning it into more of ar etelling thana ne xegesis, which is characteristic of the peshaṭ exegesis of the second generation afterR ashi.⁷⁶ Thisg loss simply ignores its hypotext (Rashi) and refers directlyt ot he biblical narrative. ' Under my thigh: Whoever takes an oath must take some sacredobject in his hand, such as aT orah scrollo rT efillin (cf. bShev 38b). As circumcision was the first commandgiven to Abraham, and it was very special to him because he suffered great pains whilecomplying, he chose this (bodily) 'object' (cf. ' He will send his angel. Ik now that they will allow you[ to take her]. AndIsaid: 'Oh, YHWH,t he God of my master Abraham,' etc. The reason for the extended speech is to let them know [for sure] that the matter proceeded from YHWH (Genesis 24:50).
[…] We cannot say anything to you, either bad or good.N either destroyingn or establishing[ the matter at hand] depends on us, since [it will happen] willy-nilly, whether we likei to rn ot,f or YHWH, whoi sa ll-powerful has arranged for it.[ … ]We will call the girl [in order to see] whether she would liket ow ait af ull yearo rt en months (Genesis 24:55) as we suggested, or to go immediately, as yousuggested.' On this paragraph, see in detail also Liss, Creating Fictional Worlds,6 1 -63.  See Liss, Creating Fictional Worlds,e sp. 120-135.
Scepticism, Critique, and the Art of Writing 3.3 Fol. 11r

Description
Fol. 11r is at extbook example of how,and with what purpose in mind, the writer integrated exegetic commentaries as glosses into his work. It alsoi llustrates whyi ti s importantt op ay proper attention to the glosses' mise-en-forme.
On fol. 11r,t here are three glosses sitting side-by-side (G_11r/1-3, from right to left), twoo fw hich are (discernibly) un-signed. Onlyt he left one bearst he signature of ‫׳ר‬ ‫מש‬ ‫לאו‬ (hereinafter:G _11r/3). The gloss in the centrei ss igned with the letter ‫׳ת‬ tosafot-the signaturea ppears in the middle of the gloss. Allt hreeg losses wered eliberatelys haped, with the use of abbreviations and/or space fillers, into R. Shemu'el'sc haracteristic layout of ah ead or circle form, and each was placed under one column of Rashi'sc ommentary.The first anonymous gloss (hereinafter G_11r/1) begins with the lemma to Genesis 25:21, ‫חכנל‬ ‫ותשא‬ ,and is found below the right column, whose first row contains Rashi'sc ommentary to Genesis 25:16.⁷⁷ The second gloss (hereinafter G_11r/2), signed with a ‫׳ת‬ in the third rowand thereforeidentifiable as agloss by an anonymoustosafist,islocated below the second column whose first row contains Rashi'sc ommentary to Genesis 25:20. The biblical lemma to G_11r/2, however,d oes not begin there, but in the last row of G_11r/1⁷⁸ (with ‫כבתא‬ ‫רו‬ ‫חקלית‬ from Genesis2 7:36) and is thereforel inked to this commentary.T he third gloss (G_11r/3)was placed below the leftmost column, which contains Rashi'scommentary to Genesis 25:22(starting alreadyinrow 31 of the middle column) and whose last row contains the commentary to Genesis 25:27, concluding on fol. 11v/row 10.Aparatextual link to Rashi'sc ommentary is absenti na ll threeg losses. This (verse) is meant to conveyt hat,j ust as Abraham moved to Egypt in the days of the (first) famine, Isaac intended to movetoEgypt,(that is) throughthe Philistines' land, because this was the shortest wayfromCanaan to Egypt.(But,) as it is written: God did not lead them by way of the (land of the) Philistines etc. (Exodus1 3:17), (so) it is also (meant) here, because the HolyO ne said:

Explanation
G_11r/1-3are glosses on the biblical text of Genesis25:21-26:2.First of all, the exegesis to Genesis 25:22f., which boththe Midrash and Rashi'sc ommentary conduct in a stronglya nti-Christian tone,⁷⁹ is here reduced to its peshaṭ:R ebecca'sw orries were unfounded, as the prophet informs her,b ecause her condition is the resulto f being pregnant with twins. This exegesisc loselyp arallels the commentary in RTC, though they are not verbatim copies.⁸⁰ The other comments,too, show close kinship  , and in it is because the people come to me to inquireofYHWH (Exodus 18:15).1( 23)Y HWH said to her through ap rophet.Two nations.D on ot be afraid! The discomfort of your pregnancyi sbecause youare carryingt wins in your womb, and the discomfort of apregnancy with two [fetuses] is greater than of ap regnancyw ith [only] one [child] […] "Behold, Ia ma bout to die:" Every dayIgo to hunt animals in the forest,where one can find bears and lions and other fe-Scepticism,C ritique, and the Art of Writing not with Rashiʼsc omments but with RTC, which is in part due to the fact that they also concern themselvesw ith the narrative'sl iterarya rc-as the cross-reference between Genesis 25:34a nd Genesis 27:36d emonstrates.
G_11r/1 deserves special attention; firstly, it appears to be an unsigned gloss and onlyw hen closely scrutinised does it reveal itself to be closelyt ied to G_11r/2.S econdly, it compresses am ajor biblical literary arc (Genesis 25:21-34) into af orm whose exegetic content parallels thato fR TC.I nG _11r/2,h owever,t his is labelled as at osafist'sa ddendum. At this point,i ts hould be clear thatt he circular form of R. Shemu'el'scommentaries is not chosen arbitrarily. There are two possible explanations for this: 1. The explanation to Genesis 25:34(includingthe referencetoGenesis 27:36) was available to the glossator as an anonymous tosafist'saddendum. He felt that it went well with other literary-theoretical peshaṭ explanations of R. Shemu'el, and therefore integratedi ti nto the triple configuration of glosses signed with R. Shemu'el. This being the case, he would have bestowed an 'author' upon an anonymous explanation, but alsol eft ah int for future scholars by adding the attribution ‫׳ת‬ ,e nabling us to reconstruct how medieval gloss collections turned into 'author'sc ollections', like RTC, that provide commentary withouta ny hypotext.A lternatively, 2. the commentary on Genesis 25:34( includingt he reference to Genesis 27:36) was available to the glossatorb ot as an anonymous tosafist'sa ddendum and as an explanation by R. Shemu'el. In this instance, he would have wished to credit both originators (byu sing signaturea nd head shape).
G_11r/2 and G_11r/3 explain whyg oing to Egypt would not have been in Isaac's best interest,while pointedlydisagreeing with Rashi'scommentary on Genesis 26:2.⁸¹ In particular, drawingo nt he Midrash, Rashi explains that Isaac is comparable to a burnt offering (ʽolah)w hich one is not allowed to present on the wrongs ide of the curtain, for it would be rendered void. G_11r/3,h owever,i nsists (indirectly) on the peshaṭ thatfollows from the immediate literary context.There is no need for the Midrash. Strictlyspeaking, G_11r/3 is alreadypart of fol. 11v,because Rashi'scommentary ad loc. begins there (second column,second row ff.). That it was nevertheless placed on fol. 11r can be explained by the desire to combine G_11r/3 with G_11r/1 and G_11r/ 2, therebyidentifying the three heads as aconnected commentary by R. Shemu'el. It is apparent thatt he glossator focussed much more on the biblical text as his hypotext than on Rashi's: the triple-form appears on the pageonwhich, with Parashat Toldot (Genesis 25:19-28:9), the specific narrative arc, to which these commentaries pertain, arises.

Description
There is onlyone gloss (G_15v/1)o nf ol. 15v that was written by the first scribe (Menaḥem).⁸² It is attributed to R. Shemu'el, and it,t oo, is instantly recognisable as a Shemu'el 'head',asitwas giventhis form by filling in blank space wherenecessary. This gloss was placedbelow the first (right) column, in which the last rowofRashi's commentary on Genesis 32:25 begins. The lemma ‫יו‬ ‫קבא‬ ( ' and […] wrestled')c omes with aparatextual reference( circle), which, however,i sn ot repeatedi nt he gloss itself, and is probablym eantt orefer explicitlyt othe biblical lemmat olink the gloss to the biblical hypotexta nd not to Rashi'sc ommentary.

Edition and Translation
And therewrestled aman with him (Genesis 32:25). The Holy One, Blessed be He,promised him I will surely do you good (Genesis 32:13), but he did not believehim, and Jacob was greatlyafraid  Therei s, however,a na dditional gloss on fol. 15v,i ntroduced with ‫ך״ד‬ and written in the later hand (below the left column).

Explanation
The commentary begins with the wrestlingm atch, which it interprets as ap unishment for Jacob'sl ack of trust.I ta ttemptst oe stablish ac ause-and-effect chain between Jacob'sf ear (Genesis 32:8) and the wrestling match at the Jabbok (Genesis 32:25ff.): Jacobd id not react properlyt oG od'sp romise and wasp unished for it.I n contrast to Rashi⁸³ who not onlyf ocusses on the morphologyo ft he verbal form ‫יו‬ ‫קבא‬ ,b ut also devotes significant attention to the 'man' (  ' And there wrestled am an with him:M enaḥem [ibn Saruq]e xplained it as "am an coveredh imself with dust," as an expression [belongingt othe semantic field of] "dust," since they werer aisingt he dust with their feet through their movements.H owever,i ta ppears to me that it means "he became bound up" (2 Kings 9:14)a nd this is an Aramaic expression, as in "after they became bound up [with it]" (cf. bSan 63b) and "and he would bound the [four threads unto the loop to form] as lipknot" (cf. bMen 42a)-an expression of entanglement,f or such is the wayo ft wo people whoare struggling to throweach other-that one hugs and twines himself round [the other] with his arms.Our Masters of blessed memory explained that he was Esau'sministeringangel (cf. Hebrew quotation unvocalised]: 'Buta na ngel wrestled with him,s oa st on ot allow him to flee in order that he might see the fulfilment of God'sp romise that Esauw ould not harm him. When he saw that he could not prevail,i.e., the angel saw, and that [Jacob]was tryingt ocross and flee against the angel's will […]But the reasonthat Jacobwas punished and lamed was because the HolyOne, Blessed be He, promised him, but he still [repeatedly] attemptedtoflee. Similarly, we find that anyone who attempts its 'retelling',which deals with the motif of Jacob'sflight on an almost epic scale, reflects another developmental stageo fpeshaṭ exegesis, bothw ith respect to literary technique and the psychological characterisation of the biblical protagonist.⁸⁵ If the gloss at hand in Ms Vienna,h ebr.2 20,d id indeed originate with Rashbam, it would be compellinge vidence thatt he commentary in RTC should be attributed to a( later)s cholar: mi-de-ve-Rashbam (Ps.-Rashbam) rather than Rashbam himself.⁸⁶

4C onclusion
Alreadyt his initial glimpse on the cases tudies presented here shows that these tosafists' glosses do not comment on their hypotext,R ashi, but expound the biblical text by focussing on the plot of the biblical narrative and its story line, the psychologyofthe biblical characters,oroncontemporary profane lore and knowledge.However,with regardtotheir literary shape, it is important to underline that they do not constituteacontinuous and fluent presentation, but agloss commentary that along with its external mise-en-texte represents the consensus patrum, in this case: the consensus magistri,i .e.R ashi. By adding explanations on the biblical text into (a recension of)t he Rashi commentary,the tosafists not onlys how that in the Christian environment the Bible had become an important tool for their intellectual discourse at ajourney or refuses ajourney against God'swill, is punished: In the [story of the call of]Moses it is written [first]: "Send someone else, whomever you want!" (Exodus 4:13), [and then the textg oes on]: And YHWH'sa nger blazed up against Moses […]H owever,a ccordingt ot he peshaṭ [therei sam ore explicit perceivable effect]: SinceM oses was reluctantt og o, it came to pass on the way at the lodging-place, YHWH met him and sought to kill him (Exodus4 :24).L ikewise, in [the case of]J onah, who was swallowed up into the bellyofthe fish [as aresultofhis refusal to go](cf. "And his bones were dislocated" (Job 33:21).'  Ihavedemonstrated elsewherethat RTCdisplays aquitefascinatingpsychological sensitivity.Jacob'sfear of Esauisthe main reason for his attempt to flee. However,RTC does not refertothis emotional stateo fm ind explicitly, but rather indirectlyt hrought he motif of Jacob'sa ttempt to flee, i.e., the depiction of his preparations for escape and the events occurringtohim. Icompared this literary technique, in which acharacter'sactivities are indicators of his internal state of mind, to literary featuresi nt he romanceso fC hrétien de Troyes (c. 1140-c. 1190) which often makeu se of this specific literary practice.O nt he whole subject,c f., in detail, Liss, Creating Fictional Worlds, 66-67,1 54-161; eadem, "Kommentieren als Erzählen: Narrativität und Literarizität im Tora-Kommentar des Rashbam," Frankfurter Judaistische Beiträge 34-35 (2009): esp. 103-110 and 118-121.  This would also solve some of the problems that occur when comparingthe RTCwith the roman-cesofChrétien de Troyes and their literary features, sinceChrétien wrote his major poems and romances( Erec and Enide; Cligès; Yvain, the Knight of the Lion; Lancelot,the Knight of the Cart,a nd Perceval, the Story of the Grail) in the last thirdofthe twelfth century,not earlier than 1170,adate in time that Rashbam did not experience anymore, as he died probablynolater than 1158. Iwill address this question in mored etail in the edition of the entireg loss material to follow. Scepticism, Critique, and the Arto fW riting tinctive grammatical, exegetical,o rp hilosophical approach in their attempt to demarcatet hemselvesf rom other authors. In philosophical as well as in most of the grammatical and exegetical treatiseso ne findsa ni ntroduction (haqdama), in which the authorʼsm ethoda sw ella sh is critique is explicitlys tated, or,a si np hilosophical as well as (later)kabbalistic exegesis, forward the claim that anyexegetical endeavour should lead to the discovery of some deeper meaning 'behind' the literal surface.⁸⁹ A 'doubting author' sets up acounter authority about asubject,abelief, in medievalphilosophical terms: a res.Byconstrast,agloss does not necessarilyaim to explicitlydoubt anything,but to explain the issue at hand or append new ideas on a text:averbum.⁹⁰ Scepticism/sceptical thoughta lways expresses doubts about a res. We might,therefore, conclude at this point thatthe investigation of glosses in amedieval manuscript as aform of test case could provethatwhen dealing with medieval Jewisht osafist literature an extension of the term scepticism/sceptical thought beyond its epistemological meaning actuallyc auses more problems than it solves. In addition,f utures tudies on the question of the different patterns of critical thought in tosafist literature should be carried out far more in relation to the literaryf orm in which it is expressed than has been done up to the present.