Bairdia bradyi

Bairdia bradyi View in CoL and The Illusory Truth Effect

Cognitive psychologists warn us of the Illusory Truth Effect: The perceived truthfulness of a statement increases as a logarithmic function of the number of repetitions ( Hasher et al. 1977; Brashier et al. 2020; Hassan & Barber 2021). Politicians and writers of advertising slogans profit by exploiting this tendency. But, repetition of an assertion does not make it true. In the case of species identifications, caution is necessary in the compilation of regional species lists. Even if a name was originally applied with reservations or uncertainty, its repetition in regional checklists, secondary compilations, and electronic databases tends to confer incremental authority to that identification.

The supposed occurrence of B. bradyi in the New World derives from the influence of two eminent authorities, G.S. Brady and W.A. van den Bold. For Caribbean researchers of the mid-20 th Century, Brady’s (1880) identification of B. foveolata at Bermuda and at five other locations around the globe was considered credible, in spite of his expressed reservations and the numerous discrepancies of the drawings. W.A. van den Bold and his students reported B. bradyi throughout the Caribbean, although the species to which this name was applied were heterogeneous.

Three generic re-combinations have been proposed for B. bradyi : Holden (1976) identified late Cenozoic fossils of Bairdoppilata sp. aff. B. bradyi (Bold) from drillholes on Midway Island but did not discuss the reclassification. Kontrovitz (1978) reported Paranesidea sp. cf. P. bradyi (Bold) from the Pleistocene of South Florida but did not mention any reason for the generic re-assignment. The brevity of accompanying taxonomic discussion suggests that both Holden and Kontrovitz considered only the material under examination (from Midway and Florida), not the nominotypical population of B. bradyi in New Caledonia. In a regional checklist, Hanai et al. (1980) accepted Brady’s (1880) subsequent misidentifications of B. foveolata from Java and Hong Kong. They re-assigned the species to Neonesidea , without taxonomic discussion, and without citing the original description from the type locality in New Caledonia.

Both Holden and Kontrovitz deliberately recorded uncertainty about the species identification by their use of open nomenclature. “Cf.” suggests that critical information is missing, perhaps by reason of poor preservation, whereas “aff.” leans toward the probability that the material does not truly belong to that species but to a different, unnamed species ( Bengston 1988; Sigovini et al. 2016). Taxonomists understand these conventions and depend on them for recording uncertainty about species identifications. Species synonymies, secondary taxonomic compilations, and regional species lists that are published in books and journals generally repeat the species identification exactly as given by an author, including cf. or aff., thus ensuring that the doubts of the original author remain a matter of record. In the Caribbean region, 5 of the 16 misidentifications of B. bradyi were recorded with uncertainty, using cf. or aff.

The ICZN (1961, 1999) has made no Rule or Recommendation concerning Open Nomenclature. Likewise, it does not require that generic recombinations be justified, or even intentional, because the Rules explicitly refrain from restricting taxonomic decisions. In the case of B. bradyi , two of the three generic recombinations were applied to uncertain identifications in open nomenclature, while the third was based on a probable misidentification. None of them considered the New Caledonian population.

With the emergence of electronic databases and search engines, there is a tendency toward the “rounding up” of such untidy designations as cf. and aff. This tends to distort taxonomic communication, giving the appearance of certainty where originally there was none. With a single click, a mistake (such as misunderstanding B. bradyi to be a sp. nov., rather than a nom. nov. for an existing species) can effectively transfer the application of a name to a different species in a different part of the world. The option to select one generic recombination as “accepted” (WoRMS instructions to editors, https://www.marinespecies.org/aphia.php?p=manual) encourages a taxonomic decision by an editor, perhaps without new or sufficient information. With each click on a database entry, the recorded mis-information takes on more apparent authority. As databases multiply and draw their data from other databases, inconsistencies propagate and become difficult to correct. If databases become the primary resources for future researchers, subtlety will evaporate, allowing AI to extract the desired, though perhaps spurious, consensus.