Dear Donald,
Thanks for your e-mail. I agree with Richards reply, the used
definition of taxon names and taxon concepts are a bit confusing and
not directly matching the zoological point of view.
For a zoologist a taxon name is the naked name plus the
authorship (author and year). At the species level this includes the
epithet and authorship and often a link to the original genus. In the
TCS I thought a naked taxon name is called a name-element. For
botanists this part of nomenclature is rather implicit and the
zoologist's (pre-)occupation on these details quite obscure.
When a species-group name (so the epithet/author/year) is linked
to a genus-group name we obtain a scientific species name (so the
actual binominal), often also called the species concept, but in
general we understand as species concept the species name plus the
associated subjective synonyms (although the semantics on this point
are a bit fuzzy). I think your 'taxon name' is equivalent with the
scientific species name.
The taxon concept is considered the use of a species concept in
literature, so that equals your definition, however, the issue of
taxon concepts is very much a botanic subject. This is caused by the
fact that plants are sessile and have to adapt to local circumstances
(animals run away from unfavorable environments). Due to local
adaptation (phenotypic plasticity) each plant specimen can be
considered as a crucial observation to verify a species concept.
Another reason is that in contrast to plants asexuality is an
exception in most non-sessile animal groups which provides a more
randomly cross-breeding population and more morphological homogeneity.
For zoologist the issue of taxon concept is more implicit and often
merged to the issue of species concept, but of course this much
depends on the taxonomic group.
The reason why zoologists are more focused on epithets compared
to botanist is twofold:
(1) Zoologists (particularly entomologists) are not bookkeepers
of nomenclatural history in a way botanist are. Especially objective
synonymy is (i) not fully catalogued and (ii) solved at the generic
level (as synonyms of genus names) if relevant. The historical reason
to this practise is highly pragmatic: The -compared to botany- few
entomologists have to deal with so many species names, so many
parallel descriptive efforts in the past, and commonly incomplete
taxonomic treatments that a nomenclatural bookkeeping sensu botany is
out of question. Within Fauna Europaea we experienced that many
entomologists even don't feel the slightest need to (as they call it)
"maintain the trash of history".
(2) In zoology we have less homonomy compared to botany,
especially when the authorship is taken into account. This could be
partly due to the compared to botany extended authorship (also
including year) and partly to other constraints (e.g. the fact that
the zoological society is more divided and therefore more original in
its naming). As a consequence zoologists are less drowning in a sea of
homonyms and therefore there is less need to refer to identification
anchors like for instance original publications. Thus I think taxon
names in zoology are far more suitable for GUIDs compared to
botany.
------------------------------
How is Fauna Europaea organised?
- Basically we have a list of 'naked' taxon names (from the
infraspecific level to Kingdom) including full authorship for at least
the generic and (infra)specific levels. All those taxon names (read:
name elements) have unique id's.
- All epithets are linked to their original genus, so original
combinations can be reconstructed. This is nearly, but not completely
similar to basionyms in botany because gender unequivalencies are not
necessarily corrected.
- Species names are created by recursively linking species-group
names to genus-group names. Genus names are parents of specific
epithets, and specific epithets are parents of infraspecific
epithets.
- Subjective synonyms are specific epithets recursively linked as
childs to 'accepted' specific epithets. For objective synonymy genus
names synonyms are recursively linked as childs to 'accepted' genus
names.
- Conclusion: Names (=naked name element/author/year plus
original genus for epithets) are for eternity. Species names (so your
taxon names) are temporary concepts (if they are not basionyms) which
don't have identifiers and which are not kept in the database after
taxonomic changes else then by version control.
- However, an unique species name id can be easily artificially
created by merging both generic and epithet id's (a trick we also use
for Species2000).
- Taxon concepts are not explicitly covered.
------------------------------
About TCS, GUIDs and nomenclators.
During the TDWG meeting at Christchurch I was happy the TCS did
allow the flexibility of building up species and taxon concepts from
scratch (so from the name elements). In St Petersburg I discovered
that the adapted TCS had been botanized and the starting point moved
to the binominal. Due to that zoologists evidently lose some
nomenclatural playground and a possibility of 1:1 matching of their
names to the schema. However, that's not a major problem because for
most mappings zoological databases need to set up views anyway. The
root for taxonomic modelling now is the original combination (=
basionym) which is persistent, hence OK. My only remaining concern is
the implementation of objective synonymy which is fundamentally
different in entomology.
For practical reasons I think the starting point for assigning
GUIDs should be basically nomenclatural. This mean that we should
establish nomenclators and leave taxonomy out for the moment. With
such nomenclators (plus GUIDs) we have the right tool to backlog
scientific names in any database for unambiguous cross-linking.
Taxonomy can be kept by the databases themselves and initiatives like
the Catalogue of Life. Aside we should distinguish basionyms for
unequivocal linking to type specimens and genetic resources.
Kind regards,
Yde de Jong
Zoological Museum Amsterdam
[ Another topic for comments. Please keep
the Topic number in responses.
]
Topic 3: GUIDs for Taxon Names and Taxon
Concepts
Another key area in which TDWG has recognised the need
for globally unique identifiers is in connection with taxon names and
the various concepts associated with them. This issue actually
also intersects with that of identifiers for taxonomic
publications.
Definitions
In the following discussion, a "taxon name" is a
scientific name string which simply identifies a name assigned in the
taxonomic literature. In many cases such a name may have been
applied in different ways by the original author and subsequent
taxonomists. Each such application of a taxon name by a
taxonomist to a set of organisms is here referred to as a "taxon
concept". An understanding of the taxon concept adopted by a
researcher is frequently essential if data are to be interpreted
correctly. In its most basic form a "taxon concept" can be
considered to be the use of a given "taxon name" in a given
"taxonomic publication", in other words something that could be
represented as, "Agenus aspecies Author1 Year1 sec. Author2
Year2". One possible approach to assigning
identifiers to taxon concepts would therefore be to assign identifiers
to taxon names and to taxonomic publications and to use a combination
these identifiers to identify each taxon concept.
Note that a taxon concept may be defined at least in
part by a set of assertions about the relationship between the present
concept and the concepts adopted by earlier taxonomists. In
addition it is possible for other researchers to make their own
assertions about the relationships between the concepts published by
different taxonomists. Much of the interest and value to be
gained from modeling taxonomy relates to the interpretation of these
asserted relationships.
Although the distinction between taxon names and taxon
concepts may seem (over-)subtle, it is important that we should know
whether we are referring simply to a nomenclaturally valid name, quite
independently of any set of organisms to which it may be applied, or
to a taxon concept which somehow applies such a name to such a set of
organisms. Without this distinction, we will be restricted in
our ability to develop biodiversity informatics, although of course
there will be many cases in which all we can say is that a data set
refers to some unspecified taxon concept associated with a given taxon
name.
Identifiers
Clearly there are many situations in which a taxon
name can itself be treated as a unique identifier without any apparent
ambiguity about which name is being referenced (e.g. Turdus
merula; Poa annua), but the existence of homonyms prevents
this from being generally true. Even when taxon names include
citations of the original publications (e.g. Turdus merula
Linnaeus, 1758; Poa annua L.), they can be very difficult to
compare since the form of the citations may vary greatly. Even
where there is no ambiguity about which name is being referenced, such
a name does not by itself serve to identify which associated concept
is being referenced.
There are many different systems in place for
associating other identifiers with either taxon names or taxon
concepts. ITIS (http://www.itis.usda.gov/, http://www.cbif.gc.ca/pls/itisca, http://siit.conabio.gob.mx/) assigns Taxonomic Serial
Numbers (TSNs) to each name in its system. Other species
databases have their own identifiers for taxon concepts.
Recording schemes often have their own identifiers for taxa (e.g.
Bradley and Fletcher numbers for Lepidoptera in the UK, various
systems of four-letter codes for North American bird species).
These are often used to provide some stability and clarity in the
taxonomy used by a given project.
Questions
I would like therefore to ask the following questions
of any of you who use scientific names in your databases (either
taxonomic databases recording a list of taxa, or databases recording
information about taxa, specimens, observations,
etc.):
1. Is your data organised using
taxon names or to taxon concepts?
2. Do you assign any reusable
identifiers to taxon names or concepts (i.e. identifiers used in more
than one database)?
3. If so, what is the process in
assigning new identifiers for additional taxa and for accommodating
taxonomic change?
4. Where are these identifiers
used (other organizations, databases, data exchange, recording forms,
etc.)?
5. Do you use identifiers from any
external classification within your database?
6. Would there be any social or
technical roadblocks to replacing these identifiers with a single
identifier that was guaranteed to be unique?
As before I am looking for information on existing
practices and any requirements that would need to be accommodated
within any general system of identifiers.
Thanks,
Donald
---------------------------------------------------------------
Donald Hobern (dhobern@gbif.org)
Programme Officer for Data Access and Database Interoperability
Global Biodiversity Information Facility Secretariat
Universitetsparken 15, DK-2100 Copenhagen, Denmark
Tel: +45-35321483 Mobile: +45-28751483 Fax:
+45-35321480
---------------------------------------------------------------