Editions and Printings: Matching Editions
Part I. D.
(See also Part II.B.2, on editing dates)
Dates are perhaps the most complex area to analyse when considering whether a record
matches the piece. Differences in the actual date of publication are significant. The
problem is that dates may appear as publication dates, copyright dates, or printing dates,
and that, depending on the situation, any of these types of dates may or may not represent
the actual publication date. Therefore, it may be a challenge to determine whether a
given date is significant. Complications result from ambiguous presentation in the piece,
the fact that these types of dates often occur in combination, and changes in cataloguing
rules over the years.
If there is a question about matching dates: For new titles, OCLC and TULANET should
be searched (or re-searched), while for adds, at least TULANET should be searched. All
records that are possible candidates for a match should be carefully reviewed. Other
factors being equal, the record with the earliest acceptable date(s) is preferred. Please
note that quality of cataloguing copy is not a factor.
An important caveat is that the significant date(s) in the 260 $c subfield cannot be later
than the date(s) in the piece; in other words, you can move backward in time from a
date in the piece that is not considered significant, but you cannot move forward. For
example, if your piece has a U.S. copyright date of 1935 and a later printing date of 1942,
you may use a record with a copyright date of 1935. On the other hand, you could not
use a record with a copyright date of 1946.
One exception involving copyright dates: If you have only a copyright date in your piece,
but you have reason to believe that the actual date of publication was later than that
copyright date, you may move forward in time to use a record with a later bracketed
publication date. (See "Copyright dates: Exceptions" for examples.)
Since simple printing dates are not considered significant, a 260 $g printing date does not
affect matching even if it is later than any date in your piece.
Cataloguers (including adds cataloguers) who are trained to detect printing dates:
If you can determine that a date in the 260 $c subfield of a record is simply a
printing date, you may disregard that date for matching purposes and instead match based
on any significant date (such as a copyright date).
(Cf. Part II.B.2, section on editing printing dates.)
- E.g. T.p.: 1903
- T.p. verso: c1898
- 260 $c 1905, c1898
- We already have copy 1 on this "1905" record. Comparing the pieces, you can tell
that they are simply
- different printings of the same edition.
- The ideal matching record for both copies would be an 1898 record. If there is no
1898 record, you
- may use this "1905" record, deleting the 1905 date.
However, when in doubt, it is safer to consider dates significant and not disregard them.
In the examples below, the "ideal" record in terms of matching dates is indicated together
with acceptable alternatives, provided that everything else matches. Again, if you have
any questions, it is safest to check with your trainer or with the principal cataloguer.
Return to top of this document
-
The following types of dates are significant and must match for the record to be
a match (unless you suspect an error):
Return to top of this document
Possibly significant dates
The following types of dates may or may not be significant, depending on the
context:
- Title-page dates: Combined with other distinctive elements (differences
between a record with an earlier date and the piece), title-page dates are
likely to represent actual publication dates. Clues could include different
dates accompanied by distinct place of publication, publisher, pagination,
or series elements, as defined elsewhere in this document.
- E.g. T.p.: 1982
- T.p. verso: c1980 ... Reprinted with corrections 1982
- A 1982 record should be used.
- On the other hand, if there are no additional differences between a record
- with an earlier date and the piece, a title-page date is now generally
- considered just a printing date.
-
(Cf. Part II.B.2, section on editing printing dates.)
- E.g. T.p.: 1965
- T.p. verso: Copyright 1952; Reprinted 1958, 1962, 1965
- The ideal matching record, if everything else matches, would be a
1952 record. Matching
- records might also be found with dates
between 1952 and 1965, in which case the earliest
- of these records
should be chosen. A record with 260 c c1967 could not be used.
- Printing dates when no other dates are provided:
(Cf. Part II.B.2, section on editing printing dates.)
Printing
dates may appear in association with a statement explicitly indicating
printing, such as "Third printing 1967" or "Reprinted 1958" or may be
associated with phrases that are more ambiguous, such as "First paper-back
edition" or "3e édition." They may be represented in the record either
explicitly as a printing date (e.g., "1980 printing") in the 260 $c or $g
subfields or as an inferred date of publication, in brackets.
Combined with other distinctive elements (differences between a record
with an earlier date and the piece), printing dates found by themselves are
likely to represent actual publication dates. Clues could include different
dates accompanied by distinct place of publication, publisher, pagination,
or series elements, as defined elsewhere in this document.
- E.g. T.p. verso: Copyright 1946 by Alehouse Editions
- Reprinted by Grog Press 1974
- Since a new publisher is involved, the 1974 date here represents an
actual publication date;
- a 1974 record would be a match; a 1946
record would not be a match.
- On the other hand, if there are no additional differences between a record
with an
- earlier date and the piece, a printing date is not significant.
- E.g. 1ère édition Editions éléphantines 1986
-
3e édition 1988
- There is a record for the 1. éd., with 260 $c 1986; everything else
matches the piece.
- This record can be used, with the 1988 date considered a printing
date and disregarded.
Dates accompanying a first printing are generally significant.
- Dépôt légal/depósito legal dates:
(Cf. Part II.B.2, section on editing dépôt légal dates.)
Generally, when found in
combination with other dates, a legal deposit date is not significant.
However, if it is the only date in the piece, it may be used as an inferred
date of publication if there are other distinctive elements in the record.
- E.g. T.p. verso: c1960
- Colophon: Dépôt légal 1re trimestre 1973
- There is a record with 260 $c [1974, c1960]; everything else
matches.
- This record may be used. Note that one might interpret the
bracketed 1974 date as another dépôt légal date, as a printing date,
or as an inferred date concocted by a cataloguer. Whatever the
case, the fact that the dépôt légal date in the piece does not match
the bracketed date in the record does not prevent a match.
- E.g. Colophon: D.l. 1986
- There are no other dates in the piece.
- 260 $c [1986]
- If everything else matches, this record is a match.
- Corrected dates: If the 260 $c subfield has one date, followed by an actual
publication date represented in brackets with "i.e.," you may use this
record if either date is present in your piece, there are no other conflicts,
and no better record is available.
- E.g. T.p.: 1878
- The only record that matches otherwise has 260 $c 1878 [i.e.
1887].
- This record may be used. (A cataloguer may still need to decide
whether the 1887 correction
- is plausible or whether it should be
deleted.)
- E.g. T.p.: 1977
- The preface is dated April 1978
- The only record that matches otherwise has 260 $c 1976 [i.e.
1978].
- This record may be used.
If the date of the first printing is later than the stated date of publication,
the printing date should appear as a bracketed correction to the stated date
of publication.
(Cf. Part II.B.2, section on editing incorrect dates.)
- E.g.: T.p.: 1983
- T.p. verso: First printing 1984
- The ideal matching record would contain 260 $c 1983 [i.e. 1984].
A record that either
- contained only the date 1983 or the date 1984
would be acceptable. If both of these latter
- records are available,
prefer the 1984 record.
Return to top of this document
Dates that do not matter
The following categories generally do not matter for matching purposes:
Return to top of this document
Nature vs appearance of date
The nature of the date -- not its appearance -- is a factor in matching.
Wherever and however it appears, if the piece has an actual publication date, it
should match the date in the record. The way in which the date is presented in
the record does not matter; it may be represented with or without brackets, as an
explicit publication date, a copyright date, or a bracketed inferred publication date
based on a printing date.
(Cf. Part II.B.2, on editing dates.)
- E.g. A piece that states "First Canadian edition 1992" can be matched with a
record with 260 $c c1992.
- E.g. A piece with "New and expanded edition 1954" can be matched with a
record with 260 $c [1954]
- E.g. A piece with "c1983 ... Primera edición 1983; 2a edición 1983 ... depósito
legal 1983" can be
- matched with a record with 260 $c 1983, 260 $c c1983,
or 260 $c [1983].
- E.g. A piece with "Zweite, vermehrte und verbesserte Auflage 1977" can be
matched with a record
- with 260 $c 1977 or 260 $c [1977].
- E.g. A piece with no publication-related date, no preface date, and no other
visible date can be matched
- with a record with 260 $c [1873], 260 $c 1874,
or 260 $c [1878]. You would probably need
- to weigh other factors to
decide which of these records to prefer.
HTML document last reviewed: 24 January 2000