Nigel Collar
BOU Council feels that the continued existence of national groups, each understandably feeling an obligation to retain control of the taxonomy used for their own national lists, is not in the best interests of ornithology and is a barrier preventing the establishment of a unified European taxonomy (https://www.bou.org.uk/british-list/taxonomy/). the scientific community's failure to govern taxonomy threatens the effectiveness of global efforts to halt biodiversity loss, damages the credibility of science and is expensive to society (Garnett & Christidis 2017). It is hard not to sympathize: surely nobody welcomes or benefits from mismatches between taxonomies (taxonomy being the phylogenetic classification of living organisms; mismatches can also occur as a problem of nomenclature – the rules-based system of assigning names in phylogenetic classifications – but this is not the issue under consideration here). Uniformity and agreement – harmony and homogeneity – on regional and indeed global taxonomic scales are sensible, desirable and perhaps even achievable goals, and ostensibly it follows that we should take or support all reasonable measures to reach them. In ornithology, four world lists are now, in Garnett & Christidis's view, ‘competing for legitimacy’: two long-established and constantly evolving works which take their names from their founders, ‘Howard & Moore’ (Dickinson & Remsen 2013, Dickinson & Christidis 2014) and ‘Clements’ (Clements et al. 2017), and two more recent offerings, the online-only International Ornithological Congress (IOC) list (Gill & Donsker 2017) and the HBW-BirdLife illustrated checklist (del Hoyo & Collar 2014, 2016). Each has its legitimacy already. ‘Howard & Moore’ enjoys universal respect for its pioneering attention to matters of nomenclatural priority. ‘Clements’, twinned with eBird, serves an ever-expanding community of birdwatchers. IOC enjoys the cachet of the foremost global ornithological institution. Finally, HBW-BirdLife forms the basis for the IUCN Red List and the ‘HBW Alive’ online resource. Garnett and Christidis illustrate their argument with a graph labelled ‘Mix ‘n’ Match’ which shows the widely discrepant totals of species that these four lists recognize, ‘mostly because they use different definitions for what constitutes a species’. I think this last point is largely mistaken. What unites the lists is far greater than what divides them; and what divides them may perhaps be less to do with fundamental disagreements over species definitions than with working schedules and systems. The HBW-BirdLife list, which in Garnett & Christidis's graph shows a sudden sharp leap in species recognized in 2016 (from under 10 500 to over 11 000, including extinct forms), is merely the product of a proactive application of a set of criteria to the global avifauna in a way which the other lists have not sought to emulate. In due course, however, some at least of the HBW-BirdLife changes may well be adopted by the others, particularly if validated by independent study, as in the cases of (among others) Lesser Crescent-chested Puffbird Malacoptila minor (Ferreira et al. 2017), Lesser Sri Lanka Flameback Dinopium psarodes (Fernando et al. 2016), Turquoise-winged Parrotlet Forpus spengeli (Bocalini & Silveira 2015) and Peruvian Pipit Anthus peruvianus (van Els & Norambuena 2018). But there is something else that divides them, of course: pride, the first and worst of the seven deadly sins (I display it simply in rolling out the list of confirmed splits above). The pride people take in their work is the best of it, but this seemingly irreproachable quality links directly to that irrational loyalty and partisanship we all show to the various institutions – family and faith, country and college, football teams and fashion-houses – in which we human beings have a deep emotional stake. In science this tribal mindset is kept in check by the demand for evidence and proof, the scrutiny of peer-review and the expectation of replicability. Taxonomy, however, is at the looser end of science, where facts need the ample mediation of human judgement, belief and opinion to become meaningful. Consequently, if you question someone's taxonomy, you question their authority, even their identity; and everything suddenly becomes rather tense and personal. Simply therefore for reasons of professional pride – why make all that effort only to let someone else overrule your considered decisions? – authors/editors of the four lists will naturally, inevitably, want theirs to prevail. Even so, one thing that largely unites the four lists (although IOC is unique in remaining silent on the conceptual framework of its species-rank decision-making) is their discomfort with the phylogenetic species concept (PSC), perhaps the greatest difficulty with which concerns the constant potential for subdivision of taxa as their populations are found to exhibit smaller and smaller diagnosable differences (Mace et al. 2003). Recent research has postulated the existence of 1.97 phylogenetic species per biological species of bird, thereby increasing the global total of ‘species’ under that concept to 18 043, with a confidence interval of 15 845–20 470 (Barrowclough et al. 2016). It is perhaps the width of that confidence interval, reflecting so much uncertainty and indicating the need for so many years of work to reduce it, that most entitles the authors/editors of world lists to confine the PSC to the sidelines of avian taxonomy; and for the most part they exclude species defined solely under that concept from consideration. Yet is this not to replace individual tribalist loyalties with a form of collective tyranny? It is hardly as if we want to deny the existence of the kinds of variation that PSC-driven analyses identify, or set them aside as inconveniences. The Barrowclough study is an important reminder of the biological diversity that exists within what we currently label species and it is wholly legitimate to seek to enumerate and recognize it. In fact, one of the most exhilarating and simultaneously depressing aspects of the modern surge in molecular phylogenetic studies is the way new and unexpected characterizations of species keep emerging (exhilarating for obvious reasons, depressing because it so often adds new burdens on the shoulders of conservationists already staggering under the weight of their responsibilities). We are increasingly discovering that avian populations, from boreal waders and oceanic seabirds to passerine migrants and tropical residents, are subtly adapted to local conditions – the Hawaiian Petrel Pterodroma sandwichensis differs morphometrically on each breeding island (Judge et al. 2014), Red-necked Nightjars Caprimulgus ruficollis are larger in one breeding habitat than another (Camacho et al. 2016) – and the ‘implications for conservation’ (a subheading that is becoming standard at the ends of modern taxonomic papers) are, collectively, enormous. In truth, it is not just conservationists who are staggering under the strain of all these new insights: the sheer volume of research into variation within bird species published in the course of the present century has also overwhelmed the authors/editors of the world checklists. The Introduction in del Hoyo and Collar (2016) records that publications on passerine taxonomy alone appeared at the rate of 0.5 per working day for the 6½ years between the start and the end of the HBW-BirdLife publication project. This level of pressure is in my view far too great to permit dependable or stable decisions (which are likely to be made either too quickly or not quickly enough) and consequently no-one and no cause is well served. If there are reasons to lament the passing of the BOU Taxonomic Sub-committee, this is one of the best: to act as a point of reference on the content and import of some of these publications, helping to ensure they are not overlooked and to clarify their worth. Sifting through myriad papers to assess the taxonomically transformative outcomes their authors claim they produce is a long and lonely business. Being able to cross-refer to other reviews, irrespective of their standpoint or conclusions, as a means of discovering and shaping one's own thought and position, was and remains hugely helpful. … vagueness is not compatible with conservation. To protect biodiversity, laws need to compartmentalize it, and those compartments must have legally defensible boundaries. Having myself sought to create such boundaries (Tobias et al. 2010) and ended essays on the same point about law (Collar 1997, 2013), I naturally warm to Garnett & Christidis for their unflinching forthrightness, but I suspect that many members of the avian taxonomic community – see Raposo et al. (2017) for early evidence – will sense the spectre of tyranny (‘restrict the freedom of taxonomic action’!) animating so revolutionary an idea as a single global authority to pass judgement on their work. Moreover, ‘legally defensible boundaries’ for species may take some time to be acknowledged as a necessity, and much longer to be agreed. Beyond the class Aves it is entirely normal for species to be delineated by molecular means only, although not without scepticism (e.g. Monaghan & Sartori 2009). Within ornithology, the BOU Taxonomic Sub-committee laudably sought to make progress in this area 15 years ago, but the resulting ‘guidelines’ (Helbig et al. 2002), involving a set of uninterpretable ambiguities and partial definitions, could not be made to work when applied to uncontentious European taxa (Collar 2013). This failure was in fact the spur to the development of the criteria in Tobias et al. (2010), which were used as the basis for the taxonomic decision-making in del Hoyo and Collar (2014, 2016). It would seem that Garnett & Christidis's requirement almost inevitably predicates very similar criteria, involving multiple phenotypic characters and yielding accountable transparent results. Laws take time to draw up, pass and train people to enforce – this is a crucial circumstance that the zoological community must grasp – so one might expect real resistance to the use of weaker criteria or lower thresholds than those in Tobias et al. if they risk producing outcomes more open to challenge and reversal. However, lower thresholds will still be needed for ‘other taxonomic units’. Infraspecific variation urgently needs a standardized system of classification, precisely in order to capture the biological diversity to which Barrowclough et al. (2016) draw attention; the eBird-Clements, Howard & Moore and HBW-BirdLife lists all make informal gestures in this direction by identifying ‘subspecies groups’, but much more work is required to bring rigour and coherence to the issue (although here again the Tobias criteria could perhaps make a useful contribution). At present, definitions of subspecies are far from fixed; and even molecular scientists cannot always agree over process and evidence (e.g. Zink et al. 2013, 2016, McCormack & Maley 2015, Patten 2015). There is, however, little point in establishing any of these ‘boundaries’ unless they can then be used on the global avifauna with sufficient speed for conservation – which is Garnett and Christidis's (2017) guiding consideration – to benefit. On the other hand, the manpower required to reassess the global avifauna – let alone all the other faunas of the planet – and keep pace with the relentless march of new publications cannot come without a significant cost. At present, the four world checklists fund themselves, and their work is undertaken on budgets that are dwarfed by the investments in molecular research around the world which produce most of the studies the checklists have to deal with. Garnett and Christidis mention that the International Commission on Zoological Nomenclature (ICZN), on which we all depend for final arbitration on the scientific names of animals, is a branch of IUBS, the body to which they propose to transfer all taxonomic authority, but they do not mention that ICZN recently had to be rescued from major financial difficulties by the National University of Singapore (Normile 2013). If it is true that the scientific community's failure to govern taxonomy is ‘expensive to society’, someone needs to muster the evidence to convince ‘society’ to cut its losses in one area by investing wisely in another. The final and most important issue is, of course, how a new commission could be created that will possess the internal resilience – impartiality, scholarship, industry, capacity, finance – to function efficiently as the ultimate taxonomic arbiter of the measurement of biological diversity. Given the almost inconceivable spectrum of that diversity, it is hard not to think that the commission's first action would be to collapse under the weight of its own remit; the spectre of tyranny might yet be banished, but the distinction between inertia and intolerance is not obvious when answers are urgently needed. At any rate, barring some remarkable generosity by a host government or major foundation, it seems wholly improbable that progress will be made any time soon. Still, if Garnett & Christidis's idea has traction, could the birds perhaps lead the way? Should the ornithological community attempt to build a cut-price prototype, whether under the aegis of IUBS or not? The four world checklists are, as noted, buckling under the burden and, if nothing more, well out of sync and half out of pocket. Could it be time for them to disband and pool their pride in a new venture to approach their task afresh and work towards the goal Garnett & Christidis have set? The editors/authors of the four checklists have already, without external prompting, held friendly talks to explore the issue. The most obvious challenges are to agree what might constitute ‘legally defensible boundaries’, to develop a process that is consistent, fair and robust, and to address the fact that none of the persons involved has a mandate from the ornithological community to undertake the work in question. If I can see the benefits of a unified list, I cannot forget my debt to the BOU Taxonomic Sub-committee and other such bodies: it will be crucial to the enterprise that it is understood to be driven by a spirit of goodwill that acknowledges – and even draws strength from – the diversity of taxonomic standpoints, and not by delusions of global hegemony. This needs some thought. Learned ornithological societies might provide neutral ground, independent advice and intellectual input that would help decide the framework, the constitution and the schedule (a 4-year cycle geared to the IOC perhaps) under which such a collaboration could flourish. The financial resources of the four checklists might be combined and, if possible, supplemented. Some or all of the editors/authors might need to recuse themselves or at least seek the approval of authorities representing ‘the best interests of ornithology’; and they, too, would surely need to be supplemented. Above all, to give this initiative to synthesize a single world list of birds the chance of lasting long enough to confer any recognizable benefit to society, the resulting taxonomy will have to be widely and rightly perceived as the offering of a scrupulously observant democracy and not the imposition of a heedless and self-serving tyranny. Stuart Butchart, Lincoln Fishpool, Josep del Hoyo, Guy Kirwan and Robert Prys-Jones very helpfully commented on the first draft of this note.