by Charles Godfray
The growing realisation of the threats to the Earth’s biodiversity
has thrown into sharp focus the plight of taxonomy in the cut-throat world
of modern science funding. There seems to be broad agreement of the need
to maintain our stock of taxonomic expertise, and to support our great
museums and herbaria, but these generalised positive statements seldom
translate into new funds for the descriptive taxonomy that underpins our
ability to describe and understand biodiversity. Why is taxonomy finding
it so hard to attract these new funds, and how might it be remedied?
Taxonomy is one the crowning achievements of modern science:
it has provided a logical framework in which the massive diversity of plants
and animals can be classified, and their inter-relationships resolved.
It is customary today to talk of the bioinformatics crisis caused by the
explosion of genomic and other molecular data, but the first bioinformatics
crisis occurred in the late 18th and 19th century when both the magnitude
of the task confronting taxonomists, and the difficulties of multiple workers
applying Linnean nomenclature simultaneously, became apparent. Taxonomy
solved these problems by creating the Zoological and Botanical Codes of
nomenclature; rigid rules that led to a consistent and coherent nomenclature,
universally adopted by all taxonomists.
However, we can be sure that if Linnaeus and his successors
had access to the informatics tools available today—especially the
web and associated technologies—they would have solved the first
informatics crisis in a different way, more along the lines the molecular
community have responded to the growth in their field’s data holdings.
Today there are “one-stop-shops” where one can access vast
amounts of information on, for example, DNA sequences, the corpus of knowledge
on the fruit fly or other model organisms, three-dimensional protein structures
The early taxonomists did not have the technology to create
these unitary resources and instead taxonomy evolved a distributed data
structure. The taxonomy of a group of particular organisms does not reside
in one place but is an ill-defined integral of all the publications on
that group, supported by type specimens in museums. Taxonomists are adept
at working with this system, and indeed part of being an expert in a group
is understanding the distributed structure of a group’s taxonomy.
But it is a barrier to taxonomy being widely used by non-experts.
An alternative approach, evolutionary not revolutionary, is
to move taxonomy from a distributed paper system to a unitary web-based
system. How might this work? A group would be revised and the product mounted
on the web (after refereeing and revision). The revision would include
all that would be in a traditional paper-based treatment, including references
to type specimens, though the space available on the web compared to paper
publications would allow a much greater use of illustrations and even audio
and video where appropriate. The revision would also include a list of
currently unaccepted synonyms etc. that might in future be resurrected.
Critically, future revisions need not consider any names not included in
the first and subsequent web revisions: taxonomy would be liberated from
the nomenclatural archaeology that slows down research in so many areas.
The taxonomy would not remain static but revisions would be posted for
comment and refereeing and periodically a revised current web taxonomy
would be posted, the site providing seamless links between current and
previous versions. The site would be moderated by an international committee,
the equivalent of an editorial board. Access would be unrestricted.
Such an idea has both positive and negative aspects. On the
minus side it would require new resources to host and moderate the unitary
sites. There is also a danger of authoritarianism, only revisions within
the web framework would be valid. But this could be countered by insisting
that alternative hypotheses to the current web revision are also included
on the unitary site. End-users would then be able to work with the current
consensus taxonomies, while alternatives would be available for research.
On the plus side the product of taxonomy would be a modern informatics
tool that would be much more attractive for funders. It would link produces
and end-users of taxonomy and hopefully create a constituency that would
militate for more taxonomy. It would also modernise the field and ensure
that taxonomy could absorb rather than be absorbed by molecular approaches
to the subject. Surely now is the time to experiment with a unitary web
taxonomy of a significant group of plants or animals,
For a longer version of this argument see Nature 417, 17-19.