GEN-MEDIEVAL-L Archives

Archiver > GEN-MEDIEVAL > 2000-08 > 0966889836


From: Stewart Baldwin< >
Subject: Primary and Secondary Sources
Date: Mon, 21 Aug 2000 20:30:36 GMT


Recent postings about the difference between primary and secondary
sources have made me decide to put in my two cents worth. One problem
is that you often see no actual definition of the difference between
the two, so that one must decide from context what people consider the
words to mean. Another problem is that the definitions themselves are
fuzzy. However, I think that amother source of confusion is that
historians and genealogists seem to have different definitions about
what primary and secondary sources are (based on the way I have seen
these terms used). Furthermore, I think that the historians are
right, and that genealogists would benefit by trying to use these
terms in a way more typical of historians.

To give an example, I have seen some genealogists refer to published
parish register transcripts (even very high quality ones) as secondary
sources. On the other hand, if a historian cites a highly regarded
critical edition of a contemporary medieval chronicle whose manuscript
is unavailable for examination (which I think is an analogous
situation), I do not believe that that historian's colleagues would
accuse the historian of using secondary sources, simply because the
original manuscript was not consulted.

Thus, I think it would be more appropriate to think of transcriptions,
abstracts, and translations of such sources as being primary, but with
an additional qualifier, such as an "edited/transcribed primary
source" (in the case of a published critical edition which claims to
give an exact reading of the manuscript), an "abstracted primary
source" (for example, will and deed abstracts), or a "translated
primary source" (for example, a translation into English of a
contemporary medieval chronicle originally written in Latin).
Although one must take the possibility of transcription or translation
errors into account in such cases, I think that the genealogical
public would be much better served by calling such source primary with
an appropriate qualification, rather than just indiscriminantly
calling anything secondary if it doesn't happen to be THE original
manuscript.

In additon to this, there are some other factors that should be
mentioned. One is that some sources can be regarded as partly primary
and partly secondary. I have seen examples of poorly written
genealogies which happened to contain a transcription of a letter
written by a contemporary that gives genealogical relationships.
Thus, even though most of such a work would be regarded as a secondary
source, I would be inclined to consider the small piece of the work
that contained the actual transcript of the letter as being a
transcribed primary source.

For earlier medieval sources (and some modern ones as well), the
picture is sometimes even more complicated. Rafal Prinke has already
mentioned the Anglo Saxon Chronicle [ASC], which is a good example of
this complication. While I would agree with Rafal in general that ASC
is a secondary source, there are some important caveats. One is that
ASC is not a single work, but is in fact a collection of closely
related chronicles (usually labelled with letters A, B, C, etc., to
indicate which rescension is being considered). Second is that the
most recent part of these chronicles is often a contemporary
chronicle, so that at least part of these works might be regarded as a
primary source (or a transcribed primary source, with the
qualifications suggested above). Another important factor that
distinguishes the versions of ASC from many of the records typically
used by genealogists for later periods is that the earlier portions of
ASC are based on sources which are now lost, so that even though this
earlier portion of ASC is technically a secondary source, the
historian does not have the option of referring to the primary sources
to see what they say, and is therefore put in the position of using
the sources in much the same way that primary sources are used.

Finally, I should mention that I am somewhat dubious about the idea of
trying to distinguish between "secondary" sources and the so-called
"tertiary" sources. The distinction seems awfully vague to me. For
example, if someone finds a new royal line by providing very detailed
primary documentation for one crucial, but previously undocumented
link, and then gives the rest of the royal descent by reference to
high quality scholarly articles, do you really want to call something
like that tertiary? (After all, most of the documentation would only
be from secondary sources.) It also brings up the question of "why
stop there?" Why not have "quaternary" sources, and so forth. [Would
sources that provide no documentation at all be "infinitary" sources
in this terminology? :-)] It seems to me that if you want to try and
judge secondary sources on the quality of their documentation (which
seems to be the motive of distinguishing "secondary" and "tertiary"),
then a more useful measure would be whether or not the trail of
documentation is actually followable back to the primary source on
which the conclusion is based, rather than trying to count the number
(or average number) of steps in the trail. I don't mind following the
trail of documentation in a so-called "tertiary" source through a few
steps if the trail actually leads somewhere. What I don't like are
the cases where the trail of "documentation" ends in an undocumented
secondary source, or in what might be called a "pseudodocumented"
source, i.e., those works which just list a bunch of sources, without
any indication of which source (if any) is used to justify the key
point(s).

Stewart Baldwin

This thread: