[Lexicog] gloss vs definition (was: question about markers)

[Ronald Moe]

Kim Blewitt wrote:

“The MDF formatter does, however, pull in \ge for a dictionary entry if the
record contains no \de, thus saving us from entering single-word definitions
twice for each entry.
Someone please correct me if I'm wrong here...”

 

I find that many new lexicographers (and some experienced ones) are confused
over the use of the definition and gloss fields. I’ll describe here what I
consider to be “correct practice” and the reasons why.

 

The gloss field is designed to contain a gloss (and not a definition). A
gloss is a very short, preferably one word, translation equivalent in a
second language. There is no such thing as a “vernacular gloss”. Such a
thing would be a synonym. If the vernacular has a word meaning ‘older
brother’ and a second word meaning ‘younger brother’, then the glosses would
be:

 

\ge older.brother

 

\ge younger.brother

 

Notice the use of a period to join two (or more) words in the gloss field.
In linguistics it is conventional practice to join two words of a gloss with
a period or underscore. One reason is because in an interlinear display a
space is used to separate vernacular words and align columns. So a space in
a gloss causes confusion. If you also wanted to use ‘older brother’ and
‘younger brother’ as the definitions, you would need a separate field,
because the two would not be identical:

 

\ge older.brother

\de older brother

 

\ge younger.brother

\de younger brother

 

A definition that consists of a single word is almost always (I would say
always) insufficient to describe the meaning of a word. A vernacular
definition in a monolingual dictionary that consists of a single word is
always going to be misleading because there are no exact synonyms:

 

\lx big

\de large

 

The words ‘big’ and ‘large’ in English are not identical in meaning and
usage, although native speakers of English would call them synonyms and
non-linguists would have a difficult time describing the difference. The
problem is even worse when trying to define a vernacular word with a single
gloss from an analysis language. A gloss can be used to give an
approximation of the meaning of the vernacular word, but it is entirely
insufficient to describe the meaning and usage.

 

It is best to fill in the gloss and definition fields for each sense of each
entry. If you do this you can sort on either field. You can also guarantee
that each sense has been defined. It would be rather embarrassing to find
that an entry in your published dictionary lacked both the gloss and
definition fields and therefore had no indication of the meaning. If you
only fill in the gloss and definition fields for some entries, there is no
easy way to guarantee that each entry has one or the other. (It can be done
by simultaneously sorting or filtering on both fields.)

 

The gloss and definition fields are designed for very different purposes. So
it is a mistake to use one of them for the other purpose. A gloss has three
primary purposes:

 

(1) to indicate the meaning in an interlinear display:

 

em-bwa

C1.Sg-dog

n.pfx-n

 

(2) to indicate the meaning in a linguistic paper:

 

“The noun embwa ‘dog’ illustrates this process.”

 

(3) to indicate the meaning in a Browse view in lexicography software when
space is at a premium. Unless you fill in the gloss field for each sense,
you can’t use it for this purpose.

 

In contrast, the definition has only one purpose—to describe the meaning of
a lexeme in a published dictionary. This description should be as complete
as possible. If you are publishing in print, you will need to keep your
definitions relatively short to keep the price down. If you are publishing
electronically, there is no such limit. In either case you should aim to
adequately describe the meaning. So the primary characteristics of glosses
and definitions (gloss-short, definition-complete) are at odds. If you wish
to include translation equivalents in a bilingual dictionary, it is
generally best to place them after the definition:

 

\lx embwa

\ps n

\ge dog

\de A medium sized, domesticated, carnivorous mammal (canis familiaris) kept
primary for the purpose of guarding the home; dog.

 

Notice also that the use of capital letters and punctuation in the
definition also conflicts with their absence in the gloss field.

 

When you print your dictionary, MDF will export the gloss field and use it
as the definition whenever there is no definition field in the sense.
Fieldworks Language Explorer does this too. I consider this to be poor
practice at best and downright dangerous at worst. It would be far better to
copy the gloss to the definition field whenever the definition field is
empty (and then fix up the definition field). This enables you to format the
gloss field one way (for instance joining multiple words with a period) and
the definition field a different way (for instance capitalizing the first
word and ending with a period). Copying the gloss field into the definition
field (or the definition into the gloss) can be done very easily and
quickly. I have a CC table that you can download from the DDP website that
will do this for Toolbox. It can be very easily done in FieldWorks using
Bulk Copy. So there is no reason to use this export feature of MDF or
FieldWorks.

 

By the way, the correct way to capture the data in Dimitris’ original entry:

 

kaboo n in the expression "kaboo and happy" stoned

 

is as follows:

 

\lx kaboo

\ps n

\de A word meaning ‘stoned’, used only in the phrase “kaboo and happy.”

\xv By the end of the party we were all kaboo and happy.

 

I’m assuming here that “kaboo and happy” is not an idiom, since the meaning
is transparent from the meaning of each word. If a word is only used in an
idiom, it would be handled in the following way:

 

\lx fro

\mn to and fro

 

\lx to and fro

\de An idiom used to describe back and forth movement.

\xv The battle raged to and fro between the river and the forest.

 

The word ‘fro’ may have an archaic meaning on its own, but today is
impossible to define on its own.

 

These two entries would be formatted thus (I hope the bold and italics fonts
come through on your screen):

 

fro  see to and fro

 

to and fro  An idiom used to describe back and forth movement. The battle
raged to and fro between the river and the forest.

 

Ron Moe


No virus found in this incoming message.
Checked by AVG.
Version: 7.5.519 / Virus Database: 269.21.7/1333 - Release Date: 3/18/2008
8:10 AM



No virus found in this outgoing message.
Checked by AVG. 
Version: 7.5.519 / Virus Database: 269.21.7/1333 - Release Date: 3/18/2008
8:10 AM
 
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listserv.linguistlist.org/pipermail/lexicography/attachments/20080323/4b0bc937/attachment.htm>