CODICOLOGICAL DESCRIPTION OF

'THE HAGUE MISCELLANY'

The Hague, Koninklijke Bibliotheek, MS 128 E 2

(olim 721 and AA 64)


by

Jos Biemans (Amsterdam)

in collaboration with Margriet Hülsmann (Amstelveen)

and Brigitte Schludermann (Hull)

Brief Description

Parchment, 67 leaves, c. 251 x 185mm; modern (probably eighteenth-century) ink foliation in the upper right-hand corners of the rectos (J-67); collation: 1⁸, 2⁸ lacking 7, 3⁸ (3 and 6 singletons), 4–6⁸, 7⁶, 8⁸, 9⁶, with quire signatures in red crayon and horizontal catchwords; pricking and ruling in lead point for double columns and top and bottom horizontal lines; written space c. 185/190 x 140mm, 38–42 lines; written in brown ink in a littera gothica cursiva by at least two scribes, some headings in a littera gothica textualis; underlining in red; the capital letters to the verse lines washed in yellow (quires 1–7 and some pages of quires 8–9), or rubricated in one continuous vertical line (the remaining pages of quires 8–9), usually in red and on two corresponding pages in yellow (fols 57v and 58r); paragraph marks touched in red; 2-line initials in red throughout (with guide-letters in brown ink), a 7-line decorated initial on a cusped gold ground with partial border decoration (fol. 1r); ownership mark of John Count of Nassau and his wife Maria van Loon (fol. 67v). Eighteenth-century binding, brown calf over pasteboards, blind-tooled and gilt, armorial shield of stadtholder William V, Prince of Orange-Nassau, blocked in gold on both covers, elaborately gold-tooled spine; paper endleaves, speckled edges in red.

Contents: 165 texts of varying lengths (longest 672 lines, shortest 2 lines), comprising poems (Minnereden, sproken, riddles, sayings), songs (minne-songs, dancing songs), with comments in between the texts and in the margins, in Middle Dutch (with German elements), in Middle High German (adapted to a Northern variety), and in varieties from the Middle German and Lower Rhenish regions, the latter in a mixed language from the Dutch-German border region, two French texts (a poem and a double ballad) and a few Latin sayings (see A: Index of Poems).

1.1  Condition of the Writing Material

The text is written on parchment of medium quality. Flesh and hair sides are easily distinguishable from one another, the latter being darker in colour. Some leaves show irregularities at the edges, which must have been present from the beginning.[1] Several leaves have one or more small holes dating from when the parchment was being prepared.[2] Some leaves have relatively large tears, which originally had sewing repairs; most of the thread has now disappeared.[3] One corner contracted at some point.[4] Obviously later damages can be found, such as wormholes, torn off corners, and a large tear;[5] one leaf has a sharp V-cut.[6] In many places the parchment has become transparent or glassy.[7]

1.2  Reconstructing the Original Format

An older binding preceded the present (see Binding). The three edges of the text block were trimmed at least once. This is obvious in the upper margin from the mutilated border decoration (fol. 1r), the two mutilated marginalia (fols 34r and 60r), and the cut top of an elongated ascender (fol. 12v); in the lower margin from two mutilated marginalia (fols 28r and 58r); and at the edges of all three margins from small pieces of the parchment folded inwards.

These foldings give an indication of the original dimensions of the vellum leaves.[8] After unfolding, small parts of the margin that survived trimming become visible, thus showing the width of the cut strips, which varies from about two to four millimetres. The unfolded outer top corner of fol. 21 is particularly interesting: its slightly curving vertical edge seems to be the original edge of the leaf before it was trimmed. This may suggest that the trimming of the text block took place only once. Therefore the original folios may have measured at least c. 258 x 188mm.[9]

The inner bottom corners folded inwards indicate that the trimming was executed at a later date, after frequent usage, as can also be seen from fol. 1 recto where a diagonal fold crosses the upper part of the leaf. At the point where it meets the outer vertical edge the trimmed edge is not straight, but has a small protruding corner, which indicates that the fold already existed at the time of the trimming (for evidence of trimming in the upper margin see Decoration; for the period in which the leaves were trimmed see Binding).

1.3  Quiring

The 67 leaves are collated into nine quires, all quires following the so-called Rule of Gregory (flesh side facing flesh side and hair side facing hair side), the flesh side always facing outward, following the formula FHFH (F standing for flesh side, and H for hair side.[10] Five quires are regular quaternions (quires of four sheets folded to form eight leaves: quires 1, 4 to 6, and 8; Figure 1a). Quire 2 may originally also have been a regular quaternion, the seventh leaf (between fols 14 and 15; Figure 1b) having been cut out – possibly during the writing process – without any loss of text. Quire 3 is not a regular quaternion: instead of being made up of four double leaves, it consists of three bifolia (sheets folded to form two leaves) and two singletons forming a 'composite bifolium' (fols 18 and 21; Figure 1c). It has to be emphasised that such a quire consisting of a mixture of bifolia and singletons is not unusual. Quires 7 and 9 are ternions (quires of three sheets folded to form six leaves). The fact that the ninth, i.e. the last, quire is a ternion is not uncommon: the number of leaves depends on the length of the text. In the case of the seventh quire, however, we have an unusual situation, which will be discussed below (see Why is the Seventh Quire a Ternion?).

The collation is as follows: 1⁸ (fols 1–8), 2⁸ lacking 7 (fols 9–15), 3⁸ – 3 and 6 singletons (fols 16–23), 4–6⁸ (fols 24–47), 7⁶ (fols 48–53), 8⁸ (fols 54–61), 9⁶ (fols 62–67).

Figure 1. Composition of Quires in KB 128 E 2
Figure 1a: Regular Quaternions: Quires 1, 4 to 6, and 8
Figure 1b: Quire 2 (Quaternion with Leaf 7 Cut Out)
Figure 1c: Quire 3 (Quaternion: 3 Bifolia + 2 Singletons)
Figure 1d: Quires 7 and 9

1.3.1  Why is the Seventh Quire a Ternion?

A point of discussion is why quire 7 is a ternion, interrupting the regular pattern of quaternions. There are two possibilities: The quire could have been a ternion from the beginning or a quaternion. If the quire was originally a ternion, one plausible explanation could be that it was, at one point, planned as the last quire of the manuscript. However, there are no clues for such a supposition from the sequence of the texts which continue without any obvious end point from fols 51 recto to 53 recto. If it was originally a quaternion, then, at a certain point, the innermost bifolium must have been removed. This conclusion is based on the following evidence:

• The leaf signatures, which can still be seen in the lower margins of the first three rectos, are c, cc, and ccc (fols 48, 49, and 50; see Quire Signatures).
   
• The outer side of the outer bifolium of quire 7 (fol. 48r) shows its flesh side, forming the pattern FHF; this pattern is the normal situation in this codex.

In 1940, at a time when generally speaking codicological evidence was not yet taken into account, Kossmann suggested that the outermost bifolium had been removed. In that case, however, the first folio of quire 7 would have had to carry the signature cc with the hair side facing outward. Then the quire would have followed the pattern HFH instead of the present FHF.

In his argument Kossmann focuses on the irregular transition of the text from quire 6 to quire 7 which, according to him, is caused by a problem encountered in the writing process.[11]

He points to the following evidence:

• The long Poem 120 (K 84) begins in quire 6 (fol. 44rb) and breaks off at the very end of that quire. It breaks off after the first three lines of an eight-line stanza (fol. 47vb) with two abbreviation signs: et and cetera. Presumably the scribe had noticed that, for the last nineteen stanzas, he had been repeating about half of the text of Poem 1 (K 1).
   
• The et and cetera abbreviations have been executed by a later hand.
   
• The next poem on quire 7 ("Owe owe elende") does not begin at the top of column a, but after a space equivalent to five lines (fol. 48ra).
   
• The catchwords "Owe owe elende" at the end of quire 6 refer to this poem (fol. 47vb).
   
• At the beginning of fol. 48ra two guide-letters 'o' occur, one on line 2 and the second on line 6, where Poem 85 actually begins ("Owe owe elende"). The scribe obviously intended to start Poem 85 at the top of the column, but then changed his mind and began copying Poem 85 on line 6.
   
• On line 2, where the first guide-letter 'o' appears, the scribe wrote a one-line Latin saying ("Probatio..."), followed by a blank space equivalent to three lines.[12]
   
• Furthermore, Kossmann considered the thin area of the parchment in the lower margin (below column b where Poem 84 breaks off, fol. 47vb) to be an erasure. In his opinion, the scribe had erased the previous catchword which had referred to the fourth verse line of the broken off stanza and replaced it by the present catchword.

On the basis of this evidence Kossmann suggested that the scribe continued copying Poem 84 into quire 7 as far as the beginning of the second bifolium (now fol. 48ra), then broke off when he noticed he was repeating a text, removed the first bifolium, and replaced the catchword at the end of quire 6. However, this is probably not what happened, considering the codicological evidence as outlined above, namely the leaf signatures and the FHF pattern of the parchment leaves.[13] Also there is no evidence for some of Kossmann's arguments.[14]

Nevertheless, the transition from quire 6 to quire 7 remains irregular. It is peculiar that the scribe broke off Poem 84 at the end of quire 6 after the third line of an eight-line stanza (fol. 47vb) and that he began the next poem on line 6 of quire 7 (fol. 48ra, leaving – in the first instance – enough space to complete the previous stanza of Poem 84?). It is also peculiar that the catchwords at the end of quire 6 refer to "Owe owe elende" on line 6 of the facing folio of quire 7 instead of to the Latin saying "Probatio.." on line 2. And it is even more peculiar that the first guide-letter 'o' (fol. 48ra, line 2) suggests that the scribe had initially intended to start with Poem 85 at the top of the column. One can therefore conclude that there is no satisfactory explanation.

1.3.2  Quire Signatures

Quite a few of the original quire signatures in red crayon can still be seen, some, however, only with great difficulty. They appear in the lower margins of the first four rectos, to the right of the centre of the page or in the centre (see Table1). The signatures in quire 6 are nearly illegible.

The first four rectos in each quire are listed with their folio numbers and their leaf signatures (if applicable). The signatures which are hardly perceptible are in brackets.

The quire signatures are only partially present and sometimes hardly legible. They follow the model 'a, aa, aaa, aaaa'. In quires 3, 4, and 5 no signatures can be found. Between quire 7, the unexplained ternion (see above), and quire 8 the sequence skips the letter 'd'. Strangely enough, the alphabetical sequence of the letter symbols in consecutive quires does not proceed from the beginning of the alphabet, as one would expect from their function of facilitating the arrangement of the quires.

1.3.3  Catchwords

Catchwords appear on the last page of each quire – except the last one – in the form of complete horizontal verse lines written below the right text column (see Table 2 below). Those written by Hand A (see Scripts and Scribes) are marked either by a decorative penwork frame on all four sides (fol. 8v), or by an extended paragraph sign underlining the whole 'catch-line' (fols 15v and 23v). From the point of view of the copying process and the linguistic mixture of the texts, it is interesting to compare the catchwords with their corresponding verse lines: see Table 2.

When comparing the catchwords and the corresponding verse lines, there are two cases where the catchwords have abbreviations and the verse lines do not (fols 23vb/24ra, 31vb/32ra). In four other instances the catchwords and the corresponding verse lines vary substantially:

• In the first instance (fols 8vb/9ra) the word 'here' is used in the catchword instead of the word 'vrou' which appears in the verse line and is needed by the context – an obvious mistake.
   
• In the second instance (fols 23vb/24ra) the catchword again uses a more common Middle Dutch form than is used in the text, namely 'vrqntlich' in the catchword for 'vrontlich' in the verse line: in the manuscript forms of 'vrunt-' occur 33 times, forms with 'vrqt-' 3 times, and forms with 'vront-' occur 5 times.
   
• As far as differences in spelling are concerned (39vb/40ra), the scribe uses a less common word form: the catchword has 'zach' (5 times in the manuscript), whereas the verse line has 'sach' (71 times in the manuscript). It is difficult to explain logically the differences in spelling. The spelling differences occur mainly with Hand B (fols 31vb, etc., see Scripts and Scribes).
   
• In the last instance (fols 53vb/54ra) the catchword tends to use a more common Middle Dutch word form than is used in the text, namely 'du' in the catchword instead of 'dou' in the verse line: in the manuscript 'du' occurs 126 times, 'do' 11 times, 'dou' 17 times, and 'doo' once.

1.4.1  Pricking

Prickings can still be seen in most of the upper and lower margins for the ruling of the vertical lines, and in the outer margins for the top and bottom horizontal lines.[16] The method of pricking is not systematic for all quires, quire 1 and quires 3 to 8 showing the same overall pattern, and quires 2 and 9 showing different patterns.

In quires 1 to 8 the pricks look like small incisions, sharply cut and therefore often difficult to find, cut vertically in the upper and lower margins and horizontally in the outer margins (see below). In quire 2 additional horizontal pricks are found at the outer edge of the upper margin, together with the vertical incisions (fols 9–15). In quire 9 only vertical pricks occur (or incisions?). At least in some cases the pricking of various bifolia was done simultaneously: in the quaternions by two bifolia together,[17] and in the last ternion by three bifolia at a time.[18]

The number of prickings varies. In the upper and lower margins of the rectos of quire 1 and quires 3 to 8 five small incisions can be seen near both edges: two appear close to each other at the left, two close together near the centre, and one to the right (see Figure 2, for a more exact description). In quire 2, only in the upper margin, there are also five vertical incisions, but only one at the left (and not two as above), two in the centre, and two to the right above each other (see Figure 2, for a more exact description); in quire 9, four vertical pricks are found only in the middle of the lower margin (see Figure 2, for a more exact description).[19] The prickings in the outer margins consist sometimes of two small incisions for the top horizontal as well as the bottom horizontal.[20] Sometimes a group of three or even more incisions is found per line.[21]

Figure 2. Pricking Patterns in KB 128 E 2
Figure 2a. Quire 1, and quires 3 to 8, example: fol. 66 recto
Figure 2b. Quire 2, example: fol. 28 recto
Figure 2c. Quire 9, example: fol. 9 recto and fol. 11 recto combined

1.4.2  Ruling

The ruling is done in lead point or lead pencil and consists of frame ruling (providing a frame for the writing space) for double columns, with – generally – six vertical lines and a top and bottom horizontal one per page (see Figure 3. Ruling Pattern in KB 128 E 2). At the left side of the columns, usually, a double bounding line is drawn, guiding the capital letters and the second letters of the verse lines. All vertical lines extend to the edges of the page, and the top and bottom horizontal lines to the edges of the outer margin and into the inner margin or gutter or spine fold (many lines have become faded and are no longer visible). The ruling of the double bounding lines was apparently guided only by two corresponding pricks, appearing at the far left for columns 'ra' and 'vb', one in the upper and one in the lower margin. The twin lines for columns 'rb' and 'va' were drawn parallel to the bounding lines without the aid of pricks (referring to the double bounding lines of the left columns on the rectos). Both pricks found at short distances to the right apparently functioned as guides to the vertical line on the verso, justifying the right side of the second column of text. On fol. 29v a double line is drawn at the right side of the second column of text.[22]

Figure 3. Ruling Pattern in KB 128 E 2

1.5  Mise-en-page

1.5.1  The Definition of Text Sections

Various methods of marking the beginning of a new text section are used: blank line(s), two-line initials, heading, explicit (to close the previous text), and less obvious methods such as capital letters and paragraph signs. Quite often a combination of these methods is used.

For a complete overview see Table 7. (Appendix 1) Overview of the Textual Structure: Text Sections and the Distribution of Comments and Structural Signs

1.5.2  Textual Structure

The entire text, consisting of various text sections (poems, songs, and comments, from 2 to 672 verse lines in length), is written in two columns. As usual in manuscripts with rhymed text, every verse line is written on a new line.[24] The numbers of lines per column vary from 31 to 42. This is connected to the fact that the writing space has no ruling per line, but only frame ruling. With a few exceptions blank lines separate the various text sections, although, because of the lack of ruling per line, it is sometimes difficult to define a blank line (sometimes it is more half a blank line).[25] The longer text sections begin, in general, with a two-line initial in red.[26] Most of the guide-letters for these initials can still be seen inside or beside the initial. A seven-line decorated initial introduces the text of the manuscript as a whole (fol. 1ra; see Decoration).

The first letters of the verse lines, or 'capital letters', are guided by the left bounding line, and the second letters by the second bounding line of the column. As a result there is, in general, a slight space between the first and the second letters. This is especially so from fol. 44r onwards, where the first letters are written on the first bounding line. Quite often, however, there is no space between the first letter and the rest of the verse line.[27] Because of this the first letters give the impression of forming a small column of their own, the so-called capital-letter column. These first letters receive additional emphasis, either by being washed in yellow or rubricated with a vertical line in red or yellow (see Rubrication; see also Ruling and Capital Letters).

For an overview of the textual structure, see Table 7, which lists poem by poem the various structural elements. The elements are grouped under 'Comments' (headings, comments within text, explicits, closing and other comments) and 'Structural Signs' (beginning of text sections, subdivisions).

1.5.3  Headings, Explicits, and Comments

The individual text sections are sometimes preceded by a heading or other comment, and/or followed by an explicit or other comment. It is sometimes difficult to decide whether such a line is a comment preceding a poem, or following the previous poem, or a loose comment not relating to either poem. In a few instances comments or other instructions for the reader occur within poems, including marginal comments relating to the text.

For a list of the headings and explicits and other comments see B: Index of Occasional Lines.

1.5.4  Paragraph Signs, Nota-Signs, and Pen Strokes

Paragraph signs regularly indicate subdivisions of text, either to mark separate strophes or significant text sections. Four of about 500 paragraph signs in total are accompanied by nota-signs, which also occur separately or function as introductory comments or as 'explicits'.[29]

Fairly frequently small vertical, slightly slanted, strokes appear to the left of a column, and sometimes to the right.[30] They are found mostly at the beginning of text sections, often near initials, but also within a poem (at the beginning of subsections?). These strokes could have had a guiding function for the application of initials (besides the actual guide-letters) and the execution of paragraph signs. If this is so, quite a number of paragraph signs have not been executed. They could also have been an indication for the rubricator.

1.5.5  Underlining

Headings, explicits, comments, and nota-signs, as well as catchwords, are frequently emphasised through underlining in brown or red. In some cases they are partially or completely boxed in.[31]

1.5.6  Rubrication

Rubrication in red is used for emphasising text structure. Structural signs, such as paragraph signs and nota-signs are touched in red. Red is also used for underlining or (partially) boxing in words or word groups, such as headings, comments, and explicits, and some quire signatures. Only two-line initials and some paragraph signs are entirely written in red. One verse line is crossed in red.[32] There are no headings in red and it is, therefore, not possible to compare the hand of the rubricator with the hand of the copyist. There is, consequently, no direct indication as to who executed the rubrication.

Special emphasis is placed on the capital letter column by means of rubrication. Two methods are used: (1) yellow wash of individual letters and (2) a continuous line in red or yellow from top to bottom. The yellow wash is found throughout the first seven quires and on one page of quire 8 and two bifolia in quire 9.[33] The continuous line is found on most pages of quire 8 and on the second bifolium of quire 9, covering the first vertical guiding lines on which the capital letters are written.[34] On the final verso of quire 8 the capital letters are not washed or rubricated with a continuous line.[35] Considering the distribution of the different ways of rubrication and particularly the continuous lines in red or yellow, it is noteworthy that two corresponding pages are always rubricated in the same way. This obviously indicates that the rubricator worked by separate sheets, handling two corresponding pages at the same time (i.e. one side of a bifolium).

The phenomenon of the continuous lines covering each first guiding line, emphasizes the fact that the capital letters are not written against or after the first guide line, but are centred on the vertical line. In addition one notices that this way of writing the capital letters results in a blank space of greater width between the capital letters and the rest of the text column than is normally the case when these capitals are written between the guiding lines: the latter method was followed in the first part of the manuscript, up to fol. 44 recto. It appears that from fol. 44 verso onwards the capital letters are centred on the first guiding line. Such a sudden change in writing habit may indicate that a different scribe continued writing (see Scripts and Scribes).

1.5.7  Decoration

A seven-line decorated initial V (in the form of a U), painted in blue on a cusped burnished gold ground, introduces the text on fol. 1 recto, covering the space in the upper left-hand corner of the text area. It contains foliate decoration of tripetals in blue, sprouting from a red curling vine. The gold ground extends into the upper margin in the form of a gold cusped bar, accompanied by a vine in red and blue. In the middle of the upper margin the vine splits up, its side-shoot turning upwards and splitting up to the left and to the right, filling a large part of the upper margin with tripetals in red, blue, and gold. The vine following the gold cusped bar extends to the right-hand corner of the page where it curls downwards and up again, giving seat to a bird which is characterized by its nice crest. Also from the cusped ground of the initial sprouts a foliate twig, drawn with the pen, extending into the left margin. Only part of this decoration is visible because the remaining paper, resulting from a former binding, covers the greater part of it (see Reconstructing the Original Format). The decoration dates from around 1400 (and is therefore probably contemporary with the writing of the text, see Date) and has most likely been supplied by a Northern Netherlandish illuminator.

For grotesque faces on the ascenders see Scripts and Scribes.

1.6  Scripts and Scribes

The text is written in a gothic cursive bookhand, a littera gothica cursiva. Four headings, two explicits, and one comment are written in a littera gothica textualis, used as a display script.[36] One French text, together with two subsequent German sayings, is written in a different cursive script, using a broader quill.[37] Many first lines of the text columns are enlivened by the elongation of ascenders, some of which are, in addition, decorated by the scribe with grotesque faces, i.e. human faces drawn with the pen (in some cases drawn with a finer quill?).[38]

Initially, three hands can be distinguished: A, B, and C. It is obvious that Hand A and Hand B belong to two different scribes, Scribe A and Scribe B. On the other hand, there is evidence that Hand B and Hand C could belong to the same scribe writing two different scripts (see below). Hand A wrote the first part until fol. 21, where Hand B took over for part of a folio, and then Hand A continued until fol. 28. On the verso of this folio Hand B took over again until the end of the manuscript, except for part of fol. 42 recto where a different cursive script is used, i.e. Hand C (see Table 3).[39]

Looking at the transition of Hand A to Hand B, and vice versa, one notices that the respective scribes did not take over from each other between poems, but always within a poem. Two of the three transitions occur in the middle of one column (fols 21rb22 and 22ra22) and the third occurs from recto to verso (fol. 28). Consequently, there are no changes between Hand A and Hand B from one quire to the next.

Hand A (i.e. Scribe A) is characterized by a careful style of writing which expresses itself through a regular and firm ductus of the pen. This is especially apparent in the distinctive forms of the relatively large capital letters introducing the verse lines, when compared to Hand B (see Capital Letters and Table 4).

Hand B (i.e. Scribe B) stands in sharp contrast to Hand A; this is especially noticeable at the point where Hand B takes over from Hand A for the first time (fol. 21rb22). Scribe B writes a more current, pointed script, and the capital letters are lower, smaller, and spikier, in comparison to Hand A.

Hand C attracts attention through its broader cursive script, presumably written with a broader pen. A more regular and upright aspect characterizes this script in comparison to the cursive script of Hand B on the same page (fol. 42r). The change of script coincides with a change of language: the relevant text is a double ballad in French (with two headings in Dutch, followed by two short sayings in German). The scribe was not very familiar with the French language.[40] It is possible that this scribe was the same as the one responsible for Hand B, trying to imitate a French way of writing.[41] This idea receives further support from the fact that the first two words – "Bilde milde ..." (fol. 42rb26) – of the longer poem written in Hand B, and following the texts written in Hand C, are apparently written with the same broader pen used for Hand C. This makes it very likely that one and the same scribe was responsible for both Hand B and Hand C.[42]

1.6.1  Hand B: A Closer Examination

A problem arises as one continues studying Hand B. In the course of the remainder of the manuscript the general aspect of Hand B changes. It is not clear whether it is the scribe identified with Hand B (Scribe B) who finishes the codex or whether other scribes writing in an identical style have participated in the copying process. This means that Hand B could turn out to be written by more than one scribe. For instance, in the vicinity of fol. 33 the writing seems to become coarser and in the course of the manuscript the letterforms are often less carefully written. Either the same scribe gradually began writing more quickly, or more scribes were involved. A closer examination of the script is hampered by the fact that the script on the flesh sides is slightly different from that on the hair sides. Obviously, the smoother surface of the flesh sides generally led to a little smaller and more carefully written text than could be produced on the other side of the vellum. The differences in the aspect of the script could also be caused by the use of a new quill or a newly cut quill. These variations in the writing process make it difficult to ascertain whether one scribe finished the codex or several scribes were involved. In considering whether Hand B should be split up into more than one hand, some help may come from other features, such as the placing of capital letters at the beginning of the verse lines.

1.6.2  Scribal Practice

Besides differences in ductus, Hand A and Hand B are also characterized by some differences in scribal practice:

• Hand A writes the first verse line of each column carefully on the top line. Hand B, on the other hand, does not write as carefully above the top line, especially in the second part of the codex where the line of text often crosses the writing line.
   
• In contrast to Hand A, who uses consistently majuscule letter forms when writing capital letters, Hand B uses both majuscule and minuscule letter forms – see especially the letters B, G, H, S, V, and W. Moreover, this practice is inconsistent throughout the manuscript, especially in the case of the letter S (see Table 5, showing a different practice in the latter part of the codex compared to the earlier part; see also Capital Letters.
   
• Another change in scribal practice of Hand B appears on fol. 44. Up to fol. 44 recto the capital letters are placed between the double bounding lines, but from fol. 44 verso onwards the capitals are written on top of the left line (see Capital Letters). As a result, the blank spaces between the capitals and the columns of text vary in width. Up to fol. 44 recto quite often no blank spaces are left between the capital letters and the rest of the verse lines; from fol. 44 verso onwards, generally more blank space occurs between the capitals and the rest of the text.

The changes of habit in the latter part of the codex, as observed above, may point to a change of hands. Consequently, this evidence supports the supposition that 'Hand B' represents more than one scribe.

1.6.3  Capital Letters

As mentioned above, the general text structure is dominated by the first letters of the verse lines which form a small capital letter column at the left hand side of each column (see Mise-en-page). These first letters – emphasized with yellow wash or a vertical line in red or yellow – have the function of capital letters introducing the verse lines (see Rubrication). In general, these capital letters are detached from the rest of the column of text (see Scripts and Scribes, Scribal Practice). In three text sections the verse lines are written continuously and no capital letter column appears. The verse lines in these sections are also introduced by capital letters.

1.6.3.1  Definition of Capital Letter Forms

A closer investigation of the letter forms in the capital letter columns reveals a relatively great variety of letterforms which can be categorized as follows:

1. majuscule letterforms: the letters are clearly distinct from the corresponding cursive minuscule letters in the running text. They are either built up of various pen strokes, or are written with fewer pen-lifts, or even without lifting the pen;

2. enlarged minuscule letterforms: the letters are not (very) distinct in comparison with the corresponding cursive minuscule letters in the running text. The enlarged form gives these letters the aspect of capital letters; the most striking examples are: D, H, M, V, and W;

3. minuscule letterforms, not enlarged: the letters are (nearly) identical in form and size to the corresponding letters in the running text.

The distinction between category 1 and categories 2 and 3 is clear, whereas the distinction between categories 2 and 3 is difficult to establish.[44] Table 4 therefore only expresses the basic distinction between majuscule and minuscule letterforms.

Some additional remarks on the categorization of the capital letter forms have to be made:

• most of the letters D, H, V, and W have been put in category 2 (enlarged minuscule); although these letters have a strong capital letter aspect, their actual forms do not differ much from the minuscule letter forms in the running text.[45] (especially the letters written by Hand A have a stronger capital letter aspect than those written by Hand B; see below);
   
• the most frequent form of the capital letter M has also been put in category 2; although its form differs in size from the minuscule form in the running text, it does not have a true capital letter aspect.[46]
   
• the letter O has, in general, been put in category 1 (majuscule letter forms), although its form is not quite distinct from the minuscule letter form in the running text; only the very small o's are put in category 3 (minuscule letter forms).

1.6.3.2  Differences in Capital Letter Forms between Hands

The capital letters of Hand A are relatively large and have a solid aspect, which is clearly visible on fol. 21 recto where Hand B first takes over. The capitals of Hand B are smaller and spikier in character. The differences between the two hands can be traced in a number of letterforms, especially in the following cases:

• Hand A consistently writes a majuscule form of the letter B; Hand B only writes a majuscule B twice (fols 21vb36 and 30ra23).
   
• the following letters have the same pattern in Hand A and Hand B and are predominantly majuscule: A, C, E, J, (for O, see definition above), R, and T;
   
• there is a distinct difference between the forms of the letters E and S.
   
• of special interest are those letters which have a distinct majuscule forms and are used frequently, namely the letters B, G, N, and S. In those cases Hand A uses predominantly majuscule forms and Hand B not enlarged minuscule forms (category 3);
   
• in Hand A most of the letters D, H, V, and W (category 2: enlarged minuscule letter forms) have a strong capital letter aspect; in Hand B, on the other hand, their actual forms do not differ much from the minuscule letter forms in the running text;
   
• the following letters have the same pattern in Hand A and Hand B and are predominantly minuscule: D, H, K, L, M (see definition above), V, W, Y, and Z;

In general, it can be observed that Hand A is much more distinct and consistent in its use of capital letterforms than Hand B (see Scripts and Scribes, Scribal Practice).

Figure 4. Letter Shapes For Hands A and B gives an overview of the different letterforms, as used by Hand A and Hand B.[47]

Table 4 gives a quantitative overview of the differences in the use of majuscule and minuscule letterforms. Most striking are the letters B, E, G, and S. They confirm the conclusion that Hand B, compared to Hand A, uses many more minuscule letterforms in the function of capital letters. (Hand A is responsible for 42.32% of the text, Hand B for 57.30%, and Hand C for 0.38%).

• Note: No distinction was made between the capital forms I and J. However, the capital forms are listed separately to demonstrate that the minuscule u is used several times in the capital letter column (in contrast to the normally used v).[48] In general it should be noted that in the concordance the word forms have been alphabetised according to the modern alphabet: thus i and j and u and v are listed separately.

1.6.3.3  Distribution of Capital Letters in Manuscript Sections

As a test case for the distribution of the capital letter forms in Hand B and an illustration of how the quantitative information in the various lists can be used, the distribution of majuscule versus minuscule of four letter forms, namely B, E, G, and S, which by definition only fall into categories 1 or 3, were counted for consecutive sections of ten folios. The study of this distribution is meant to test any changes in practice that may occur in Hand B in the course of the manuscript.

In general the distribution remains comparable, except for the complete reversal in the distribution of the majuscule versus minuscule S/s in the section fols 50–59.

1.6.3.4  Distribution of Capital Letters after Two-Line Initials

The preference of Hand A for majuscule forms (category 1) and Hand B for minuscule forms (category 3) can also be observed in the two verse lines next to the painted two-line initials at the beginning of a new text. After each two-line initial, which begins the first verse line but also precedes the second verse line, the letters following the initial in both lines (i.e. the second letter of the first verse line and the first letter of the second verse line) are usually capital letters.

Table 6 gives insight into this practice of Hand A, Hand B, and Hand C after two-line initials. Because the most striking difference lies in the contrast of true minuscules (category 3) with letters that have a majuscule aspect (categories 1 and 2) this contrast has been studied by comparing, this time, categories 1 and 2 with category 3. The frequencies in this table confirm once again the difference in practice between Hand A and Hand B, the former preferring majuscules and the latter minuscules.

1.6.3.5  Capital Letters in the Middle of Verse Lines

A few letters appear in a majuscule form in the middle of verse lines. These are the A and the J, and in the odd single case a C and an S. Here also a difference can be noted between the usage by Hands A and B. Both hands are comparable in the use of the J, whereas the A is used much more by Hand B. The odd cases of a C and an S occur in Hand A.

See Figure 4. Letter Shapes for Hands A and B

See Index of Codicological Signs

1.6.4  Abbreviations

See Coding List

See Palaeographical Finding List

1.6.5  Corrections

See Index of Codicological Signs

1.7  Binding

Before the manuscript received its present binding, it must have had an older one. The description of the codex in the auction catalogue of 1749 (see History), is not really worthy of being called a description, but it includes the indication 'Lig-Lign.', i.e. ligatus ligno or ligatura lignea, revealing that the manuscript had a binding on wooden (probably oak) boards. A few years later, some time between 1755 and 1759, the manuscript was re-bound together with approximately forty other manuscripts, presumably on the instigation of Princess Anna / Anne of Hanover (1709–1759), the widow of William IV and from 1751 regent for their under-aged son, William V (1748–1806).[49]

The manuscript is bound in brown calf over pasteboards, of the type called 'russia'. At the outer edges both covers have a frame consisting of a gold-tooled ornamental roll between two blind-tooled triple lines. In the centre of each cover appears the armorial shield of stadtholder William V, Prince of Orange-Nassau, blocked in gold. The Garter, carrying the device "HONI.SOIT.QUI.MAL.Y.PENSE", surrounds his arms. Two crowned lions hold up the Garter and the armorial shield; the whole is also crowned on top.[50]

The spine has five raised bands and is elaborately tooled in gold. Between the first and the second raised band of the spine is a black leather label with gold lettering indicating the contents of the manuscript "POESIE HOLLAND" (Holland certainly is an abbreviation of the French Hollandoise). Edges of the binding and turn-ins were left unornamented. On both sides of the spine, head and tail, single endbands are placed, sewn with thread in white and light brown (originally red?), to form a decorative pattern.

The codex contains four endpapers at the beginning as well as four endpapers at the end, i.e. eight in total, the first and the last being pastedowns. The three outer edges of the text block are speckled in red. Apart from two eye-catching inkblots and a few smaller ones on the lower cover, the binding is still in excellent condition.

1.8  Date and Localisation

1.8.1  Date

To determine the date of the manuscript we have to rely first of all on the information provided by the texts themselves. Aleida Nijland has pointed out that the poem on fol. 22rb27–26va10 (Poem 62 = K 42), "ein jammerliche clage" contains a terminus post quem, by referring to the death in 1345 of William IV, Count of Holland.[51] Another terminus might be found in the poem on fol. 5vb11 (Poem 17 = K 12), incipit "Die ene stede bezeten heeft", where at the end a reference occurs that the Pope is living at Avignon. The Popes spent their exile in Avignon between 1308 and 1377, so the latter date might be taken as a terminus ante quem. However, Aleida Nijland discussed this possibility and she rightly pointed out that the manuscript must be younger – from about 1400 (pp. 134–38). In his introduction to the facsimile edition of the manuscript Kossmann too, lists the facts:

Apart from obtaining information from the content of the manuscript, one can also study the form and the shape of the script. Several datings are suggested. The Dutch palaeographer Willem de Vreese (1869–1938) dated the codex twice: c. 1340 and c. 1345.[52] It is surprising that he neglected the results published by Aleida Nijland as early as 1896.[53] In his own documentation on Medieval Dutch manuscripts, the Bibliotheca Neerlandica Manuscripta (BNM, now at Leiden University Library), some notes on MS 128 E 2 are kept. On one of these notes De Vreese originally wrote as a dating "einde 15de eeuw" (end of the fifteenth century), but he crossed these words out and wrote above "geschreven ca. 1340" (written c. 1340). Presumably he adopted the first date he found in the literature on the manuscript, for instance in Aleida Nijland's thesis, and then changed his mind later on, possibly after he had studied the codex in autopsy. In any case, in his Bouwstoffen of the Middelnederlandsche woordenboek, his oldest date is 1340, later 1345.[54]

In his edition of 1940 Kossmann studied the language, spelling and script of the manuscript:

Und endlich die Sprach- und Schreibgebilde der Handschrift, auch sie weisen auf die Wende des Jahrhunderts, auf die bairische Periode der Grafschaft Holland (1358–1433) und erinnern überdies persönlich an den Vater des Eigentümers, den Grafen Engelbert von Nassau, der im Jahre 1404 durch seine Heirat mit Johanna von Polanen das hochdeutsche Nassau mit dem niederländischen Sprachgebiet zusammenfügte ("Einleitung," E.F. Kossmann, Die Haager Liederhandschrift: Faksimile und Transskription (Den Haag: Martinus Nijhoff, 1940), p. 1; see Provenance).

His date is the end of the fourteenth century, or as Aleida Nijland suggested, c. 1400. The first date was confirmed, that is to say not contradicted or denied, by G.J. Lieftinck, and the latter date is now generally accepted.[55]

1.8.2  Localisation

Unknown (see below).

1.9  Provenance

The manuscript KB 128 E 2 once belonged to Johan IV (1410–1475), Count of Nassau, Lord of Breda etc. and his wife Maria van Loon (1424–1502), who had married in 1440. Their library constituted the earliest Nassau library.[56] Nine manuscripts from their library are still known.[57]

Six of these manuscripts were written during their own lifetimes, three are older. One of these older manuscripts was produced in Germany during the fourteenth century and contains Wilhelm von Orlens, the German translation by Rudolf von Ems of an originally French chivalric romance. Johan could have received this manuscript from one of his German ancestors. The second of the older manuscripts, dating from the second half of the fourteenth century, is a Psalter written both in Latin and in a dialect from one of the border regions of the 'German' and the 'Dutch' languages (probably Westphalia).[58] Nothing is known about its earlier provenance.

The third of the older manuscripts is KB 128 E 2. Johan and Maria are the first known owners of the codex, but the manuscript itself must be older, i.e. c. 1400. We know nothing about any previous owners and can only speculate about its provenance. One of the salient features of the collection of texts is the variety of German / Dutch language mixtures, and over the last century a multitude of theories have been put forward to explain this phenomenon. Who was responsible for putting together this collection? Were these texts gathered ca. 1400 especially for producing this book, or is it a copy of an older manuscript with the same contents? Because of the presence of originally Middle Dutch texts (e.g. Noydekin and Augustijnken van Dordt), originally Middle High German texts (Walter von der Vogelweide and Reinmar der Alte), and genuinely mixed texts (Dutch-German and German-Dutch border along the Rhine and north of it), it is unlikely that the manuscript was part of the library of one of Johan's German ancestors and equally unlikely that it came from Maria's ancestors in Holland and Brabant, the Counts of Loon. The 'public' must have accepted both the 'Dutch' and the 'German' literary tradition and this may have been part of a regional tradition.

Johan's parents, Engelbert I of Nassau and Johanna van Polanen are, as members of the House of Nassau, serious candidates for being interested in such a collection of texts. Engelbert I, Count of Nassau Dillenburg (c. 1370–1442), was the first Nassau of the Otthonian line, who came to the Low Countries. In 1403 he married Johanna van Polanen (1392–1445), heiress of Breda, etc. Through this marriage the Nassaus obtained the rich possessions of the House of Polanen, a noble family from the County of Holland (Polanen near Monster in Delfland). Engelbert and Johanna used to stay at the castle in Breda, built around 1350 by Jan II van Polanen (d. 1378), Johanna's grandfather. Engelbert settled down in Breda and contributed much to the prosperity and the flowering of the city. He became the founder of the House of Nassau in the Low Countries, a House that was to play a prominent part in the history of the Netherlands. It is attractive to speculate about the possibility – see Nijland (1896) and Kossmann (1940) – that the texts of KB 128 E 2 were gathered for Engelbert and his wife.[59] Perhaps the manuscript was produced as a present on the occasion for their wedding in 1403. That would conform perfectly to the date that the manuscript was supposed to have been produced. The manuscript, however, is not a very sumptuous book, such as one could expect for such an occasion. But even if it was not made for Engelbert and Johanna themselves, someone from the cultural environment of Breda or its surroundings could still have ordered it.

Nijland, Kossmann, and Van Oostrom also point out that the collection was put together during the period when the County of Holland was ruled by House of Bavaria (1358–1433). During that time the Court in The Hague came under strong 'German' influence.

As mentioned earlier, we know from the ownership mark on fol. 67vb 20–23, that the manuscript was once the property of Johan IV, Count of Nassau, and his wife, Maria van Loon: "Dit boech hue:t zo Jonche: Johan g:eue zo nossou zo vij$d8 vnd ma:ien van loen sijn:e huijsv:auwen." The hand responsible for these lines also wrote the ownership marks in five other manuscripts of Johan IV and Maria.[60] They are almost identical in wording and must be dated between 1440 (year of marriage) and 1475 (death of Johan IV). After their deaths (Maria died in 1502) the codex remained in the library of the Nassau family in Breda castle, together with the other books in their possession. When William the Silent (1533–1584), Count of Nassau and Prince of Orange, came from Dillenburg to the Low Countries in 1544, he found, among other treasures, the Nassau library in Breda. From this time onwards, this library formed the basis of the famous library of the House of Orange-Nassau.

In the course of time the library suffered several disasters. Books were taken away or mutilated, but KB 128 E 2 survived unscathed. About 1636 the collection of manuscripts and printed books was moved to the old court at The Hague. Fifty years later the library was catalogued by the well known Dutch poet and influential secretary of three Princes of Orange-Nassau, stadtholders of Holland, Zeeland, et cetera, Constantijn Huygens, Lord of Zuylighem (1596–1687). Although the catalogue bears Huygens' name, Anthony Smets, Court Librarian, did most of the work.[61] On fol. 223 verso of this handwritten catalogue KB 128 E 2 is briefly mentioned: "1277. Oude Verssen int hoochduyts geschreven op parquement. toebehoort hebbende Graaf Jan van Nassau".[62] In 1702 William III (1650–1702), stadtholder of Holland and Zeeland and King of England, died. He was the last male descendent of the House of Orange-Nassau and once more the survival of the library was in danger. Finally, in 1734, William's properties devolved upon Frederick William I (1688–1740), King of Prussia. His successor, Frederick II, the Great (1712–1786), decided in 1749 to sell the entire library by auction.[63] Meanwhile Karel Hendrik Friso (1711–1751), Prince of Orange Nassau, the later Willem IV, succeeded William III as stadtholder of the Seven United Provinces. Thanks to his personal interference the old library of Orange-Nassau did not disappear completely from the Netherlands. His court-librarian, Jean Royer, managed to buy approximately 60 per cent of the manuscripts (but only 5 per cent of the printed books, which were apparently thought to be less important). KB 128 E 2 was one of the manuscripts that were saved in this way. They were placed in the library of the stadtholders at The Hague.

Between 1755 and 1760 more than forty volumes from the old library of Orange-Nassau were re-bound. They were bound simply in natural brown leather, as can still be seen in the case of KB 128 E 2 (see Binding). In 1795 the French invaders took a considerable number of manuscripts, printed books, maps, etc. as spoils of war to Paris, but luckily the manuscript described here was not among them. The book remained in The Hague and was incorporated in the National, later Royal, Library that was officially founded in 1798.[64]

1.9.1  Marginalia

On a few pages marginalia occur in the margins, possibly added by a later hand (a fifteenth-century user of the manuscript?).[65]

General acknowledgements: also of great help in establishing this description were the following publications: Bernhard Bischoff, Latin Palaeography. Antiquity and the Middle Ages, translated by Dáibhí ó Cróinín and David Ganz (Cambridge, Cambridge University Press, 1989); Michelle P. Brown, A Guide to Western Historical Scripts from Antiquity to 1600 (London, The British Library, 1990); the same, Understanding Illuminated Manuscripts. A Guide to Technical Terms (London, The British Library, in association with the J. Paul Getty Museum, 1994).

APPENDIX 1