DIPLOMATIC E-TEXT EDITION of _BEOWULF_ , v.2.3 (C) Copyright Rochelle I. Altman (1994, 1992). Permission is explicitly granted for any non-commerical use of this material without restriction, and in any form required by the user, pro- vided that appropriate notice is given of its copyright status. Guide to the Diplomatic E-Text Edition of Beowulf CONTENTS Part 1: The Diplomatic Text 1 1.1 The Diplomatic E-Text Edition 1 1.1.1. The Base Text 1 1.1.2. The Editions 1 1.1.3. Folio Numeration 2 1.2 The Structure of the Diplomatic E-Text Edition 2 1.2.1. The Base Text 2 1.2.3. The Apparatus 3 1.3 Interpreting the Apparatus 4 1.3.1 Examples 4 1.4 Adding Editions to the Basic E-Text 6 Part 2: Working with the E-Text 8 2.1 Technical terms - computers 8 2.2 The Utility Set 8 2.2.1 Extractor 9 2.2.2 Decoder 9 2.2.3 Screen Font Handler 9 2.2.4 Printer Font Handler 9 2.2.5 Integrator 9 Part 3: The Manuscript Emulation Transcription 10 3.1 Technical terms - palaeography 10 3.2 "Suprasegmentals" - clumping and spacing in the manuscript 11 3.3 Letter assignments 12 3.3.1 Standard ASCII (ISO 8859-1) 12 3.3.2 ENTITY coded 13 1 PART 1 - The Diplomatic Text 1.1 The Diplomatic Edition Diplomatic editions have a standard form: a base text and an apparatus. The base text is whatever text has been chosen as canonical (e.g. the Leningrad text in biblical studies) or "best" thus far. Points of variance are marked within the base text, and each point is cross-referenced in same-page footnotes; this cross- referencing is called the apparatus. The apparatus normally has some means of identifying each edition used for comparison with the base text. In printed diplomatic editions, various editions are usually assigned symbols (sigla). Ideally, it should be possible to regenerate every source text used in a diplomatic edition from the base text and its apparatus. This diplomatic e-text has a base text and an apparatus, points of variance are marked, and source texts are identifed. It is possible to regenerate every text used in the diplomatic edition -- quickly, and easily -- with a computer. In addition, the electronic format allows the regenerated texts to be available, as many as needed at one time, for any form of analysis as may be deemed necessary -- at computer search speed. There are some major differences between the standard format of a diplomatic edition and this e-text edition: the apparatus is in a somewhat different form, and the base text is not an edition, but a transcription of the manuscript itself. 1.1.1. The Base Text: The base text is a transcription of the manuscript as it appears in the facsimile edition: K. Malone, ed. _The Nowell Codex_. Copenhagen: Rosenkilde and Bagger, 1963. (EEMF XII). As all editions are based on this one manuscript text, it is the logical text upon which to base a diplomatic edition. 1.1.2. The Editions This Diplomatic E-Text of _Beowulf_ (Beo.EDN) does not include any editorial commentary or glossary material whatsoever, and has been limited to the following editions: Dictionary of Old English. _Beowulf_. Toronto. K. Kiernan. _The Thorkelin Transcripts of Beowulf_. Copenhagen: Rosenkilde and Bagger, 1986. F. Klaeber. _Beowulf and the Fight at Finnsburg_. Lexington, MA: Heath & Co. 1950, (1922). K. Malone, ed. _The Thorkelin Transcripts of BEOWULF._ Transcript "A". Copenhagen: Rosenkilde and Bagger, 1951. (EEMF I). J. Zupitza, Norman Davis. _Beowulf_. London: OUP, 1959 (1882). 2nd Edition. (EETS 245). 2 While it may appear peculiar to classify all of the above as editions, that they in fact are is demonstrable. For a discussion of Zupitza and Thorkelin B as editions, see Kiernan (43-45). The Thorkelin A is a mechanical copy of the MS.; however, among other variations, Scribe A produces his own lineation; thus the A tran- script is treated as an edition. Edition abbreviations are as follows: Manu = Manuscript (EEMF) DOE = Dictionary of Old English Klaeber = Klaeber ThA = Thorkelin A Zup = Zupitza KK = Kevin Kiernan (used only for verification of ThA readings) 1.1.3. Folio Numeration: Folio numbers are listed at the beginning of each folio to the left of the text (e.g. 132rZ;134rM, 132vZ;134vM, etc.). The first number is the numeration found in Z and KK; the second number is the new foliation given in the EEMF Facsimile Edition and used in the British Museum "Electronic Beowulf" digital facsimile. 1.2 The Structure of the Diplomatic E-Text Edition 1.2.1 The base text: The base text is written, line by line, as it appears in the manuscript facsimile, letter forms, clumping, and spacing. The following example is l. 8, f. 132r (Z, KK), "th" = eth: Swatha driht guman drea mum lifd However, there are quite a few variants on this line in the editions, and fiber-optic light has revealed letters hidden under the paper holding the MS. leaves (KK xii, 46). Variants are marked by parentheses (). Each set of () indi- cates a variation in one edition or another. Sw(ath)a drih(t g)uman dre(a m)um lifd()() Hidden letters may not be visible on the facsimile, but as they are there; these letters are enclosed with plus (+) signs: Sw(ath)a drih(t g)uman dre(a m)um lifd(+o+)() There are also "lost" letters, letters that either were destroy- in the Cottonian fire, or letters that survived and have since crumbled away at the burnt edges. While only some of the missing letters can be recovered from the Thorkelin Transcripts, we can estimate how many letters were lost. Lost or missing letters are noted by the letter "j." This letter is not used in Old English texts, and therefore is available to indicate the lost letters. 3 The base text now appears as below: ()Sw(ath)a drih(t g)uman dre(a m)um lifd(+o+j)() A computer reads a space as a character. Thus, () is a null or empty character. In other words, () indicates a variant in at least one edition that is not in the base text. Every variant is enclosed between an open and a close parenthesis. The open parenthesis "(" tells the reader that a variation begins, the close parenthesis ")" that a variation ends. All variants are enclosed within a pair of parentheses. 1.2.2. The Apparatus: The apparatus is in a same-page inter-linear format. This design is readable by both computers and humans. The following discussion explains the apparatus and how it works. As seen in section 1.2.1., variants are entered into the base text marked by a pair of parentheses. In the apparatus, variants are marked by corresponding brackets: open bracket "[" tells the com- puter/reader that the following data cross-references material within parentheses, and to continue reading until reaching a closed bracket "]" or the end of variant marker. For every pair of parentheses, there is a corresponding pair of brackets. Editorial variants are separated by a vertical line "|." There is a line for each edition. On the screen, or on a printed copy, the text will now look like this: Base Text ()Sw(ath)a drih(t g)uman dre(a m)um lifd(+o+j)() Apparatus [$|$||$][a th|a th|ath|a*th][tg|tg|t g|t-g][am|am|a m|am] [on|on|on|on][,|,||] The open bracket "[" tells the reader to begin variant search. The close bracket "]" tells the reader that a variation ends. The four vertical bars "|" in the apparatus line above, represent the four editions. The editions are in alphabetical order: [DOE|K|ThA|Z] Order is critical. Once the order in which the editions are given has been determined, the order never varies. The position within the square brackets is the equivalent of the sigla in printed dip- lomatic editions. This strictly determined order allows the computer to regenerate on command every source text used in the diplomatic e-text. The techni- cal term for this type of format is "open-ended," which means that material may be added or subtracted without re-writing either the base text or auxiliary programs. 4 1.3 Interpreting the Apparatus: To sum up the previous discussion (1.2): Every pair of parentheses "()" in the base text indicates a variation. There is a pair of brackets "[ ]" for every pair of parentheses "( )." Every vertical bar "|" within a pair of brackets "[ ]" indicates an edition. There is a vertical bar "|" for each edition. Editions are in a predetermined order. Interpretation is straightforward. First we have the line of base text and its apparatus. Base Text ()Sw(ath)a drih(t g)uman dre(a m)um lifd(+o+j)() Apparatus [$|$||$][a th|a th|ath|a*th][tg|tg|t g|t-g][am|am|a m|am] [on|on|on|on][,|,||] This means, for example, that the letters "Sw(ath)a" appear in the base text as Swatha In the various editions, these letters appear as DOE Swa tha K Swa tha ThA Swatha Z Swa*tha The same format is repeated for every pair of parentheses and vertical bars in the base text and apparatus. When texts are extracted and regenerated, the parentheses, the brackets, and the vertical bars are removed. 1.3.1. Examples: The following discussion will step through, line by line, two lines of the Diplomatic E-text Edition of Beowulf. The editions are in the predetermined order as given above: DOE, K, ThA, Z. Symbols used in the e-text are defined as follows: +o+ = letter visible by fiber-optic light (KK) $ = Start of new line in an edition. j = lost letter (Note: "j" is not used in any of the texts and can be used to indicate these lost letters.) * = indicates word division in Zupitza and Thorkelin () = variation | = edition [ = begin variation ] = end variation 0 space = a ligature, 1/2 minim, or 1 minim space 1 space = 2 minims 2 spaces = 4 minims 5 In the examples given below, "th" stands for ð and "ae" for æ. The examples are from lines 8-9, folio 132r (Z, KK): Line 8: Base Text ()Sw(ath)a drih(t g)uman dre(a m)um lifd(+o+j)() Apparatus [$|$||$][a th|a th|ath|a*th][tg|tg|t g|t-g] [am|am|a m|am][on|on|on|on][,|,||] Base Text - Swatha driht guman drea mum lid+o+j Apparatus - [begin variant DOE - new line |K - new line |ThA - null n/a |Z - new line ]end variant [begin next variant DOE - a th Swa tha |K - a th Swa tha |ThA - ath Swatha |Z - a*th Swa*tha ] end variant [begin next variant DOE - tg drihtguman |K - tg drihtguman |ThA - t g driht guman |Z - t-g driht-guman ]end variant [begin next variant DOE - am dreamum |K - am dreamum |ThA - a m drea mum |Z - am dreamum ]end variant [begin next variant DOE - on lifdon |K - on lifdon |ThA - on lifdon |Z - on lifdon ]end variant [begin next variant DOE - , , |K - , , |ThA - null n/a |Z - null n/a ]end variant Line 9: Base Text ()eadi(g l)ice() o(thth)aet ()an o(n g)an ()fyr(e n)e fr(jj) Apparatus [|||$][gl|gl|g l|g-l][,|,||][th th|th th|thth|thth][||$|] [ng|ng|ng|ng][$|$||][en|en|en|en][em|em|e |em-] Base Text - eadig lice oththaet an on gan fyre ne frejj Apparatus - [begin variant DOE - null n/a |K - null n/a |ThA - null n/a |Z - new line ]end variant 6 [begin variant DOE - gl eadiglice |K - gl eadiglice |ThA - g l eadig lice |Z - g-l eadig-lice ] end variant [begin next variant DOE - , , |K - , , |ThA - null n/a |Z - null n/a ]end variant [begin next variant DOE - th th oth thaet |K - th th oth thaet |ThA - thth oththaet |Z - thth oththaet ]end variant [begin next variant DOE - null n/a |K - null n/a |ThA - $ new line |Z - null n/a ]end variant [begin next variant DOE - ng ongan |K - ng ongan |ThA - ng ongan |Z - ng ongan ]end variant [begin next variant DOE - $ new line |K - $ new line |ThA - null n/a |Z - null n/a ]end variant [begin next variant DOE - en fyrene |K - en fyrene |ThA - en fyre ne |Z - en fyrene ]end variant [begin next variant DOE - em frem |K - em frem |ThA - e fre |Z - em- frem- ]end variant 1.5 Adding Editions to the Basic E-Text The format is compact; both human and machine readable. It also is open-ended. That is, other editions may be added at any time merely by inserting another "|" before the close bracket "]" end variant sign. The only constraints are: 7 1. Edition order must be maintained -- It would be simplest to add the new entries at the end. 2. If another edition has a variant not previously entered, the new variant must be marked in the base text between parenthesis, and entries made in the appropriate places in the existing text. Example: This is line 9 on folio 132v as it is in the e-text. ()(+w+)iht u(n h)aelo() grim (&grae dig)() gearo sona [|||$][W|W|w|w][nh|nh|n h|n-h][,|,||] [ond graedig|ond graedig|&grae dig|ond graedig][,|,||] You wish to add Edition X; however, Edition X does not entirely agree with previous editions. Editor X prefers close punctuation and has added a comma after the word "grim." Now the text would look like this: ()(+w+)iht u(n h)aelo() grim() (&grae dig)() gearo sona [|||$|][W|W|w|w|W][nh|nh|n h|n-h|nh][,|,|||,][||||,] [ond graedig|ond graedig|&grae dig|ond graedig|ond graedig][,|,|||,] A null parenthesis has been added after (g)rim() in the base text. The number of vertical bars between begin and end variant brackets "[ ]" has been increased by one in all of the marked variants, and another variant set [||||,] has been added in the appropriate place. NOTE: Both the MS and ThA use a "yogh" instead of "g," also the 's' in "sona" is "long-s" in these examples. 8 Part 2 Working with the E-Text: 2.1. Technical terms - computers Items within quotes are cross-referenced. Application - A "program" or series of programs designed to perform specific procedures on a computer. Usually distributed as a "package." Architecture - The actual processor and operating system. ASCII - American Standard Code for Information Interchangte. Bit - A binary digit, either 0 or 1. Byte - A fixed length bit pattern. Today, reference to a byte usually means a bit pattern 8 bits in length. Configuration - The type, number, etc. of peripherals. Entity - A "Micro" instruction which associates one literal expression with another. Entities may contain a single letter or a repeated phrase. TEI mark-up uses & ; to identify an entity for replace- ment. Entity Replacement - The replacement of an encoded value with another according to a table of definitions. MICRO - A one-to-one transform which allows an expression to be exchanged for another expression. Entities are a form of Micro. Program - A named set of computer instructions designed to accomplish some specific purpose. Utility - A program which performs a general purpose function. A utility may be provided with the operating system, or may be written independently. 2.2 Utility Set: In order to get the most use out of a diplomatic edition of this sort, a certain number of utilities (programs) must be employed. A utility set that is designed for the following platform is available: PC Compatible (any model), at least 256K memory DOS 2 or higher EGA/VGA or better video HPII compatible laser printer (PCL IV) compatible. A hard disk is convenient but not required. This does not mean that other platforms/configurations can not be sup- ported with a minimum of programming effort. In some cases, e.g. UNIX, standard system utilities exist that can be used with little or no modification. 9 A specific user guide for the utilities is a part of the utility pack- age. It is useful, however, to describe the contents and function of these utilities to clarify for the potential developer what if any changes or developments will be needed for his or her configuration. The utilities are all stand-alone programs that are integrated with a menuing system (described under 5 below). Accordingly, a user, whose configuration differed from the one mentioned above in that instead of a laser printer he had an Epson FX compatible dot-matrix, would only have to modify the printer-specific utilities, leaving the others as- is. 2.2.1. Extractor This accepts the text as input and one or more edition numbers. It produces one output file for each edition requested. 2.2.2. Decoder This accepts a TEI marked-up file and an Entity replacement list as input. It produces a decoded (i.e. Encoded characters replaced by the corresponding value in the replacement list) file as output. 2.2.3. Screen font handler. This consists of a screen font and a small program to load the screen with the font (and to replace it with a normal font when done). 2.2.4. Printer font handler This consists of a printer font and associated loading software. 2.2.5. Integrator. This is a menu type program that executes the above utilities as requested and then allows display, comparison, and printing of selected items. For example, One might ask it to extract the Manuscript and the Thorkelin A and then display them (with scrolling and line-jump capability) on a split screen. One could then request a side-by-side partial (or complete) printout. It also includes a configuration option that allows the user to change his replacement lists, fonts, and other variables needed by the basic utilities. (NOTE: For the moment, only the stand-alone utilities are available. The complete menu driven Integrator will be distributed under separate cover some time during the summer of 1994.) The above capabilities are obviously the minimum. There is nothing to prevent enhancement such as adding hyper-text cross referencing, displaying variants in a different color than identics, etc. Nor is there anything in this structure to prevent the user from adding a new edition to the existing diplomatic text, or even creating a diplomatic edition of some other work in the same format. The architecture is deliberately simplified so as to allow any or all of these utilities to be written for another platform with- out any special language and/or programming technique requirements. 10 Part 3 The Manuscript Emulation Transcription: 3.1 Technical Terms - Palaeography: The following is a list of terms commonly used to describe features of a scribal hand. Ascender - That part of a letter which extends above the "headline," for example, "b" or "h" have ascenders. Aspect - The overall appearance of a script, e.g. rounded, square, pointed, etc. Baseline - The line on which the letters are written. Bilinear - The script is contained between the baseline and the headline. Examples are Roman Capitals, Rustic Capitals, and scripts generally written in majuscule letters. Bite/bitten- The letters are not spaced, but touch, or bite, an ad- jacent letter. Bow - The rounded closed curve of letters such as "b" and "q." Cursive - A rapidly written hand with more linked letters, fewer pen-lifts, and other such devices to increase speed of execution. Descender - That part of a letter which extends below the baseline, e.g. "p" or "y." Ductus - The direction of the pen strokes, speed of execution, and care in formation of letters. Hand - The writing of the scribe, which is concrete, as opposed to the script, which is the mental model. Hasta - The cross- or center stroke of a letter, e.g.the central bar in the letter "E" Headline - The line that marks the height of letters, e.g. a, e, o, and m are written between the baseline and the headline. Headstroke - The horizontal top stroke on letters such as "f." Ligature - Letters using common strokes, usually upright minims. Majuscule - All capital letters -- ABCDE, etc., written between the base and headlines. Minim - The basic upright stroke as in "i" or "l." Minuscule - A script which has both ascenders and descenders. It is also referred to as a quattrolinear script as opposed to a bilinear script. Pen-lift - The lifting of the scribe's pen to execute the next pen stroke, as in the letter "p" -- first the descending minim stroke is made, then the bow is added. Quattrolinear - Writing requiring four lines, as opposed to bilinear. See also minusucle. Tail - See tongue. Tongue - An extended hasta stroke, most common on 'e' and 'aesc'. 11 3.2 "Suprasegmentals" - clumping and spacing in the manuscript: NOTE: During the early medieval period, what is meant by a "word" is dif- ficult to define; still, it is convenient to use the term in this discussion. The use of the term "word" implies no judgement, merely the recognition of patterns that were later so termed. Two scribes wrote the text of _Beowulf_. Although both scribes write an insular hand, they are very different. The major difference between the two hands is perhaps most easily described by referring to levels of formality. Scribe 1 writes a clear, small, and legible hand; although he actually uses many triangular or squared forms, the overall aspect is rounded. He uses a relatively fine pen, and as few pen-lifts as are compatible with clarity; his writing is clear and quite legible. His hand appears to be modeled on a script designed for speed. Scribe 1's writing could be classified as an informal hand. Scribe 2 has a large, heavy hand. Because he also uses techniques to clearly demarcate the triangle in a beginning stroke, and tends to write a "square-a," the overall aspect is of a square hand. His pen is cut wider, and he uses many pen-lifts. He adds many little touches to his letters. For instance, all his e's appear to have a tongue to some degree because the bow of his standard "e" curves in slightly creating a small space between the bow and the hasta. Occasionally, older forms of letters, such as the open "alpha" a ("oc" l. 1, 179r Z, KK), or curved-tongue e (too many to list) appear. His writing appears to be modeled on a script which resembles those found in the formal hands of "deluxe" copies. Scribe 2 writes what could be called a formal hand. In spite of the very different overall appearance of the two hands, both scribes are quite consistent in following the same rules with re- gard to internal and external spacing of words. It is this noticeable and regular consistency which allows the manuscript to be transcribed as an e-text. The striking spatial regularity correspondends to the definition of a minim: A minim is the width of the vertical stroke of the letter "i." Letter width is defined in minims, where minims are proportional to the width of the scribe's pen. Normal spacing between words, or external spacing, is the width of the letter "o," or four minims. Normal spacing within "broken" words, is the width of two "i"'s, or two minims. Normal letter spacing within a word, or internal spacing, ranges from bitten letters to half a minim. However, if a number of straight minim letters are adjacent to each other, a full minim space is used -- apparently, but not certainly--for clarity, e.g. Scribe 1's "s u n u," l. 15, f.135v (Z,KK; f.137v M) and Scribe 2's "s u n u," l.3, f. 183v, l. 17, f.189v. 12 A full discussion and comparison of the spatial constant would require a book; an example must suffice. NOTE: ae = aesc; th = thorn; & = tironian and On the leaves written by Scribe 1, for instance, the letter "r" almost always is connected to the adjacent vowel or consonant by an extension of the final baseline stroke. Nonetheless, there is a distinct differ- ence in the spacing of individual words. The following examples are to be found on f. 132v. Z, K; 134v M: "gri m &grae dig gearo" (l.9) and "thritig" (l. 10); "fa ran" (l. 12), "gren dles," "craeft," "dyrne" (l.14), and "s o rge" (l.8; l.17). We also find "h r i ng" (l.5) and "hr e m ig" (l. 11). Another example in respect of "re" may be found on l. 10: "reo c &r ethe." This internal spacing is repeated on the leaves written by Scribe 2. On l. 4, f.180v (Z,KK; 183v M), we find "geo rne" and on l. 5 the word "h r u s a m"; on f. 179r (182r), "byr nend" (l. 18) and "dr aca" (l. 19), but "byrn an" (l.9) and "byrn" (l.18, f. 181r Z,KK; 184r M). "S o rga," on l.12, f.181v (184v) repeats the spacing used by Scribe 1 on lines 8 and 17 of f. 132v (134v), as "h r i n g" on l. 19, f. 189v repeats the spacing on l. 5, f. 132v. In other words, the apparent differences in the spacing used by the two scribes is an artifact of the width of their pens. The consistency displayed by both scribes allows the following conventions to be used in order to emulate manuscript spacing: 0 space = bitten, 1/2 or 1 minim 1 space = 2 minims ("i") 2 spaces = 4 minims ("o") Larger spaces are a combination of the latter two, and should be treat- ed as such. 3.3 Letter Assignments 3.3.1. Standard ASCII (ISO 8859-1): Although their hands are quite different, both S1 and S2 use the same conventions in their variant letters. Examples are not available of all forms, particularly capitals, as a variant which occurred in S1's por- tion of the MS. may either have been destroyed by fire, or there are no other occasions for the use of a particular letter on the leaves written by Scribe 2. There are many variant letter forms in the manuscript, as a quick glance at the ENTITY table demonstrates, and there are some differences between the letters forms used by S1 and S2. Nonetheless, while it is quite in- teresting palaeographically, for example, that the back stroke of S1's 'd' is nearly vertical and S2's is almost horizontal, there is no ambi- guity. The majority of letters have been transcribed in standard ASCII code. 13 Lowercase letters: The following have been transcribed as their modern counter-parts: a, b, c, d, f, h, i, k, l, m, n, o, p, q, r, t, u, v, x, z. Standard e has been transcribed as ASCII e (decimal 101). Wyn has been transcribed as 'w'. A sigmoid or round s is used when the letter in the manuscript is sigmoid, otherwise, 's' is transcribed long-s or low-s. according to its appearance. Capital Letters: Smaller forms appear within the lines of text, larger forms appear be- tween fit divisions. Capital E appears in the Roman and Uncial forms; the Uncial E has been assigned decimal code 229. Capital C, M, O, S (sigmoid), and T, are of the Roman Capital form, and do not change in appearance, only in size, and have not been assigned special codes. Except for being rather tri- angular, capital Thorn resembles lowercase thorn and has not been en- coded. Wyn has been transcribed as "w" throughout, and therefore has not been assigned. The large size capital "N" is found both in Roman and in Uncial capitals, the Uncial having a right-hand minim extending and curving below the baseline. Small capital N is of the uncial form, except on l.6, f.162r and l.19, f.166r, where Scribe 1 uses a quattro- linear lowercase letter, and on l. 10, f. 194r where Scribe 2 writes a baseline crossbar N. Because exceptions are uncommon, capital N has been transcribed as "N." The one example of small capital T on l. 9, f.176r, is a quattrolinear uncial "t." Two examples of large capital 'G' in the square form remain; small capital G is of the round form. As only these two large G capitals survive, no entity code has been assigned to this letter. Capital letters not mentioned above either do not appear in the MS, or have been assigned entity codes. 3.3.2. Entity encoded Letters: Entity definitions have been assigned following the Text Encoding Initiative (TEI) format (& ;): Lowercase letters: Entities have been assigned to: aesc (æ) eth (ð) yogh (&yogh;) thorn (þ) dotted y (&ydot;). Entities have also been assigned to: The crossed thorn abbreviation for that (&that;) i-macron (&imac;) u-macron (&umac;) accented vowels: virgule a (&avir;) virgule e (&evir;) virgule i (&ivir;) virgule o (&ovir;) virgule u (&uvir) 14 There are five different 'e' forms -- standard, short-tongued, long- tongued, high, and tongued high; four aesc -- standard, tongued, high, and tongued-high; and three 's' -- sigmoid, long, and low. These are described below. To present visual material without graphic imaging is quite difficult; one can never really describe the letter forms in words. There are a number of additional reasons for the inclusion of illustrations: flex- ibility and the needs of the end-user are foremost. 1. An end-user may wish to create his or her own fonts. 2. An end-user may decide to fold a number of variant forms into one, e.g. fold 158 & 159 into 159 (long-s). 3. An end-user may decide to remove all variants, e.g. fold 156, 157, 170, 216, 226, and 228, into 101 (e). 4. Illustrating the letter forms will help the reader visualize the manuscript itself. Therefore, in spite of the constraints of an electronic format, which limit the reproduction of the letters to approximations, illustrations of the letter forms have been included. S1's forms are shown to the left of the page, S2's to the right. Letter height markers appear on the left side of the page as shown below: Ascender oo ooo o Headline o oo oo o o o o o Baseline o o o o Descender o SCRIBE 1 SCRIBE 2 aesc: Standard form oOOOOo Headline ooOooOOOoo oOOOO OO OO OO OO oO OOOoooOoo Oo OOoooO OOO OOO oo OO OOo OOO Baseline oOOO Ooo OOooO OOo Tongued aesc oOOOOo Headline ooOooOOOoo oOOOO OO o OO OO OO oOO OOOoooOoooO Oo OOoooOOOooo OOO OOO oo OO OOo OOO Baseline oOOO Ooo OOooO OOo 15 High aesc Ascender oOOOOo oOOOo OOO Oo ooOO Oo OOO OO Headline Oo OOooooo oOOOOoooOoo OO OO oOO OOO Oo OO OO OOO oo OO OOo OOO Baseline oOOO Ooo OOooO OOoo Tongued High aesc Ascender oOOOOo oOOOo OOO Oo O ooOO Oo OOO Oo o Headline Oo OOoooooooooo oOOOOoooOoooooO OO OO oOO OOO Oo OO OO OOO oo OO OOo OOO Baseline oOOO Ooo OOooO OOoo e: Standard e oOOOOo Headline oOOo oOO OOo oO o OOOooOoo OOooOO OOO Oo OOO o Baseline Ooo OOooO Tongued e oOOOOo Headline oOOo oOO Oo oO o OOOooOooO OOooOoo OOO Oo OOO o Baseline Ooo OOooO Long-Tongued e oOOOOo Headline oOOo oOO o O oO o OOOooOooooO OOooOooooo OOO Oo OOO o Baseline Ooo OOooO 16 High e Ascender oOOOOOOo oOOOo oOO OO oO oO OOO oOO Headline OOooooO OOOooOoo OO OOO OO OOO Oo OOO o Baseline Ooo OOooO Tongued High e Ascender oOOOOOOo oOOOo oOO OO oO oO OOO oOO o Headline OOoooooooo OOOooOoooooO OO OOO OO OOO Oo OOO o Baseline Ooo OOooO s: long s Ascender oOOOOoo oOOOOoo oO Oo oOO oOOOO oO o OOO OOoo Headline OO OOOO ooO oOOOO OO OOO OO OOO Baseline OO OOo OO Oo Oo Descender low s Ascender oOOOOo oO Headline oO Oo oOOOO oOOo oO o O OOOoOOOOo oO OOOo Oo ooO OOO Baseline oO OOO OO OOO Oo OOo Descender O Oo 17 Capital letters: Not illustrated: AEesc (Æ) Eth (ð) Thorn (þ). SCRIBE 1 SCRIBE 2 A: Rustic Capital Ascender ooo ooo O OO Headline O OO O OO O OO Oo OOo o Baseline oOOoo oOoO Bowed Capital Uncial Capital Ascender ooo oOOOOo ooo oO OOo O oo O OOO Headline O oo OOO Oo oo ooooOOO O Oo oo oOO OOO O OoO oo oO oOOO o Baseline oOOo OOoo OOooO OOooO E: Uncial Ascender oOOOo oOOOOOo oO oo oOOo OOOO oO oOO Headline oO o oOO o OOOOOOOOOo OOOOOOOOOOOo oO oOO oO oo oOOo oOO Baseline OoooO OOoooOO H: Unlike most of the other capitals employed, 'H' does vary somewhat. While the basic form stays the same, the letter appears both solid and decorative. Small capital H is of the solid form; most of the large capitals resemble the illustration of the H on line 13 (Hroth) f.140r. The other illustrations are from l. 1 (HWAET) f.129r, l. 14 (Hwilum) f.133v, and line 15 (HVNferth), f. 141r. Capital 'H' within a line of text is one quattrolinear line in height; capital 'H' between fits is two quattrolinear lines in height, not including tail. 18 l. 1, f. 129r l. 14, f. 133v ( zzzzzz ) (= burnt) ( zzzz ) ( zz ) zz zz zzz zz z z z Z zz z z zz Z ooo zz zz zz z Z OO OOo zzz z z Zo OOo zz z z Z OO zz z z Z OO zz z z O zz z z o z zz z z z zzzz z l. 13, ff. 140r, 192v, 198v. l. 15, f. 141r zzzzzz zzzzzz zzzz zzzz zz zz zz zz zz zz zz zz z z z zz zZZZZz zz z z zz zz zz zzz zz zz zz z zzz zzz zzz z z zz zzz zz z z zz zzz zz z z zz zzz zz z z zz zz zz z z zzzz zz zzzz z z zz zz z z z z z z z z zz zz