Fingerprint

Main page

GENERAL

The fingerprint is a tool to identify and distinguish editions. For this purpose, a formula is created following fixed rules. It contains the following sections:

• Year and format
• the position of some clearly defined quire signatures in reference to the line of text above. Each position contains three elements: indicator, signature and text above the signature

General rules

1. For multi-volume works, a separate fingerprint is taken for each volume.
2. The fingerprint is based on the ideal copy of the book. Any binding or imposition mistakes are ignored and silently corrected. Should the correct binding structure of the ideal copy be unclear, multiple fingerprints may be added.
3. The fingerprint has a fixed punctuation between the different parts. To increase the searchability in Brocade, the fixed punctuation of the STCN fingerprint is supplemented with a '#' between the different parts.
   • The year-and-format-code is followed by 'space hyphen space # space'
   • The group(s) of prelims, body matter and end matter are always separated by 'space hyphen space # space'
   • The first and second position within one group are always separated by 'space colon space # space'
   • At the end of the fingerprint (after the last position) nothing further is added

Example 167808 - # a1 *2 ai : # a2 *5 oei - # b1 A1 amaai : # b2 D5 tisniewaar

YEAR AND FORMAT

The year is given in the way it is presented in the imprint on the title page.

• Years are always transcribed to Arabic numerals. Example: M.DC.Lij becomes '1652'; αζζα becomes '1661'.
• Years in other calendars are not converted to the Christian calendar. Example: 'An 7' becomes '0007'.
• Misprints or other mistakes are not corrected. Example: 1693 incorrectly composed as '1963' remains '1963'. Exception: if the year is defective (e.g. '16.9', '0000' is noted.
• If there is no explicit year of publication mentioned in the imprint on the title page, '0000' is noted. Since single-sheet prints have no title page, their year-format code will always start with '0000'.
• Years in spurious (copy-) imprints are retained. Years from colophons etc are not retained.
• Misprints or other mistakes in Roman numbers resulting in a non-existent number (e.g. M.D.Mij) are replaced by '0000'.
• In the case of misprints or other mistakes in the year, when it is clear from other copies that they were corrected on the press, two fingerprints are taken.

The format is given as 02, 04, 08, 12 etc.; oblong is not indicated. If the format cannot be defined with certainty, it is given as '00'. For single-sheet prints, it is therefore possible to have a year-format code reading '000000'.

INDICATORS

The indicators are named as follows:

	FIRST POSITION	LAST POSITION
PRELIMS	a1	a2
BODY MATTER	b1	b2
END MATTER	c1	c2

Any subdivisions within a work, expressed by a new beginning of the signatures, are named separately in the fingerprint. The relevant indicators then receive a serial number (e.g. 1b1, 1b2, 2b1, 2b2). These serials numbers are given independently for the prelims, body matter and end matter indicators.

Positions that are not present, are not displayed. If only one position can be taken within a work, meaning that first and last position coincide, this is then given as 'a1==a2', 'b1==b2', etc. There are no spaces between the the indicators and the equality sign. For single-sheet prints, only one position can be recorded and it will therefore always be 'b1==b2'.

At the transition between prelims and body matter, it could happen that one gathering contains multiple signatures. If in such a case the transition between the two different signatures coincides with a clear content division between prelims and body matter, the prelims are considered a separate bibliographical entity and they are therefore represented in the fingerprint by the indicators 'a1' and 'a2'.

Example: gathering in 8 leaves signed as follows:

[*1] *2 *3 A4 A5 [A6] [A7] [A8]

quire structure: */A8 B-D8

fingerprint: # a1 *2 A : # a2 *3 en - # b1 A4 zo : # b2 D5 voort

However, if there is no clear content division at the transition between the two different signatures, the first (divergent) signature is considered an irregularity and the fingerprint only contains the indicators for the body matter 'b1' and 'b2'.

Example: gathering in 8 leaves signed as follows:

[*1] *2 *3 A4 A5 [*6] [*7] [*8]

[*1] *2 *3 *4 A1 [*6] [*7] [*8]

quire structure: A-D8

fingerprint: # b1 *2 A : # b2 D5 voort

In both cases, the actual situation is explained in the cataloguing remarks.

SIGNATURES

☞ Main article: Signatures for the Fingerprint

How to reproduce and choose signatures for each part of the publication, what to do with multi-volume works, unsigned publications and unusable signatures.

General

• The signatures are given in simplified form (e.g. 'Aaa iij' becomes '3A3'). Attention: the first folio in a gathering, for example 'A', is usually found without a number and therefore noted as 'A' and not as 'A1'.
• In the exceptional case that there is more than one signature per page, either the bottom signature or the signature that is most to the right is used.
• By signature, the whole signature is meant, including any surrounding brackets, full stops, ornaments, etc. are included. A signature such as '(Aaa iij.)' is abbreviated to '(3A3.)'.
• Misprints in the signature are not corrected. They are explained in the cataloguing remarks.
• When different states result in different fingerprints, a second fingerprint is added. This can happen when a work is published with two different dedications, or when a misprint was corrected during the printing process.

Choice of signatures

The chosen signatures are the first and last of each primary or secondary bibliographical (and not as regards content) section.

• The prelims are only included in the fingerprint when they are signed separately. In theory, it is considered as a whole. However, each content division that coincides with one signature (* of § or + etc., whether it is in a series or not) received its section in the fingerprint (1a1 ... 1a2 ... 2a1 ... 2a2 ... etc.). Because different states (different dedication, extra liminary poems) are often found in the prelims and because one binder might place an index at the front and another rather at the end, it is better to have too many than too little sections in the fingerprint.
• The text of the body matter is considered as a whole when one or more alphabets are signed in a continuous and regular manner. The procedure is repeated each time a subpart of the text (as regards content) starts with a new alphabet, even when the previous gathering happens to end with a Z.
• If a new subpart of the text (as regards content) nevertheless starts within a gathering, but is signed from that point with a new alphabet, the procedure is also repeated.
• Pieces of text that are added in but signed separately also receive a separate mention in the fingerprint. A collation formula such as A-2D4 a-g4 2E-2F4 is represented in the fingerprint in three sections as:

1b1 A2 ... : # 1b2 2D3 ... - # 2b1 a1 ... : # 2b2 g3 ... - # 3b1 2E1 ... : # 3b2 2F3 ...

→ See also: Bibliographically semi-independent publications.

• If a regular series of signatures is interrupted by signed prelims of a second part (as regards content), these prelims are entered separately in the fingerprint and the otherwise regularly and continuously signed body is considered as one whole. A collation such as A-2C8 *8 2D-3P8 is thus split into prelims and body matter in the fingerprint as:

# a1 *2 ... : # a2 *5 ... - # b1 A2 ... : # b2 3P5 ...

• The end matter is only included in a fingerprint when it is signed separately. The same rules apply as for the prelims.

Multi-volume works

Each volume of a multi-volume work receives its own fingerprint. A new field with a new year-and-format code is used for each volume. Each fingerprint is preceded by 'volume number # space'.

Example

1# 162508 - # a1 ...

2# 162608 - # a1 ...

3# 162608 - # b1 ...

For a variant in the fingerprint, it is enough to repeat the fingerprint of the relevant volume in a new field.

Example

2# 162508 - # b1 A2 onze : # b2 P3 v

2# 162508 - # b1 A2 nze$ : # b2 P3 v

Unsigned

• When a book (or its body matter) is wholly unsigned or if there are only unusable signatures, an alternative fingerprint is taken. A choice is made for the first (but not the title page!) and the last printed recto page of each subdivision (both bibliographically and as regards content) of the book (prelims, body matter and end matter, with possibly clearly divided parts). On those pages, the third word of the final line of text is taken instead of the absent signature. If the last line of text does not contain a third word, we move to the penultimate line. Then, we note the bit of text of the preceding line, which is directly above that third word of the final line.

In many cases, single-sheet prints are unsigned and an alternative fingerprint needs to be taken. The following points require attention:

• a colophon (if present) is never used to take an alternative fingerprint. The use of 'standing type' for this colophon might result in a great number of editions (e.g. series of ordinances) to have identical fingerprints. The fingerprint is therefore always taken from the third word of the last line of the actual text
• if this third word is not suitable for taking a fingerprint, the penultimate line of text is used instead, and we keep moving one line up until a suitable line/word is found for taking a fingerprint
• text on single-sheet prints is often divided in columns. The fingerprint is taken in the first column on the left of the page, even when one or more other columns continue lower down the page. When the last line of this first column is unsuitable for taking a fingerprint, we move to the penultimate line of that first column, not to the other columns

Example

In a collation formula [A]-[F]4, in which [A]1, [B]1 and [F]1 are respectively the start of the prelims, the text of the body matter and the index, the following choices are made:

f. [A]2 recto and [A]4 recto for the prelims (positions a1 and a2)

f. [B]1 recto and [E]4 recto for the body matter (positions b1 and b2)

f. [F]1 recto and [F]4 recto for the end matter (positions c1 and c2).

• Words that are attached to the next word without spaces (e.g. l'hiver, qu'ils) are considered as one word. Therefore, if the third word to be used for the fingerprint is attached without spaces to the fourth word, this entire combination is used for the fingerprint.
• The ampersand '&' is considered a word when counting words for an (alternative) fingerprint.
• A punctuation mark following a 'third word' does not belong to that word and is therefore not included in the fingerprint. This holds true for any punctuation mark, including the abbreviation mark (e.g. Nov.).
• When an unsigned body matter is found in combination with signed prelims or signed end matter, the normal rules apply for these prelims and end matter.
• If the preceding lines are unusable (e.g. atlases, books of plates, sheet music, tables), the fingerprint contains nothing further than the year-and-format code.

Example Fingerprint: 163208

Unusable signatures

When a signature is unusable, the following (for the positions a1, b1 and c1) respectively the preceding (for the positions a2, b2 and c2) usable signatures are taken instead of the unusable signature.

A signature is unusable:

• if there is no text above it
• if the last line of text ends precisely above the quire signature. If the last line of text ends before the start of the quire signature, then the text from the penultimate line of text is used
• if the bottom of the quire signature is more than 25 mm from the bottom of the last usable line of text
• if the quire signature falls below an illustration, graph, table, line, staff (music), map, etc. or below the white or the vertical line separating two columns
• if the symbols are in a non-Latin or non-Greek alphabet (Arabic, Hebrew, cyrillic, ...)
• if there is white space between two letters of a deliberately expanded word above the quire signature

Sometimes, it can be useful or necessary to use a signature after all, even if it is unusable. In this case, this is marked by a * (asterisk) before the indicators (e.g. *b1 A2 ...). This can be the case if it is the only or only semi-usable signature in a section of a book with otherwise unusable signatures. Such a fingerprint is also taken if the signatures in an entire work are separated by a line from a framed text. If there is but one fingerprint position, the star is not repeated (e.g. *b1=b2 ...).

TEXT ABOVE THE SIGNATURE

The piece of text that falls precisely above the necessary signature, is noted. In actual reality, this means: the characters that are wholly within the signature. In the case of continuing doubt regarding the situation, two fingerprints can be taken. It can happen that a heavier use of ink leads to differences/

If a signature only falls below a part of one character or parts of two consecutive characters, these parts count as whole characters.

Example

M    MA    nu wel

A    A2    A3

Here, respectively M, MA and $ are noted in the fingerprint.

By characters, we mean: typographical entities such as letters, numerals, punctuation marks, symbols, etc. A space counts as one character, regardless of its length, and is displayed as $ (dollar sign). The characters are transcribed according to the general STCV-transcription rules. The 'ij' (regardless of the language or meaning) is considered as one or two characters, depending on the typography used within the book.

For the capital letter Q with a long tail that continues under one or more of the following letters, this tail is neglected for the purposes of the fingerprint. For example, if in the words 'urbs. Quae est', the tail of the Q continues underneath the letters 'uae es', then the following counts as text above the signature:

Qu if the circle of the Q and the u are above the signature

bs.$Q is the letters bs. and the circle of the Q are above the signature

ae$e if the letters ae and the letter e are above the signature

This is all despite the continuation of the tail of the Q. Similar cases (where a curl continues underneath or above the following letters) are treated the same way. However, for the italic long ſ connected to the following letter in a ligature, for example 'st' (ſt), the full letter needs to fall above the signature in order to be included in the fingerprint.

The ligatures Æ, æ, Œ, œ, & (always transcribed as a ligature) and ß (transcribed as 'ss') are considered as one character. All other ligatures are considered to be multiple characters. Abbreviations with curls are ignored; if there is an abbreviation mark, this is noted.

Small caps are also considered capital letters: Iesv Christi thus becomes 'IESV CHRISTI' in the fingerprint.

If a fingerprint in Greek characters can be avoided by using the respectively following or preceding signature, this is to be preferred. It means any problems with transcription of ligatures and accents can be avoided. If this is not possible, the Greek letters are transcribed to Latin letters.

As much as possible, other alphabets (Arabic, Hebrew, Cyrillic, Old Church Slavonic, Coptic, Demotic, Arameic, Linear A, ...) are transcribed for the fingerprint. In the worst case, the fingerprint will be limited to the year-and-format code.

In order to avoid interpretation differences regarding punctuation marks (especially comma's), the rule is that there is never a space between a punctuation mark and the preceding word. Between the punctuation mark and the following word, however, a space is always noted.

INPUTTING THE FINGERPRINT IN BROCADE

Taking the fingerprint