by Stephen Moss
The US team’s suggested changes to classification were contentious, to say the least. They proposed wholesale changes in the family tree of the world’s birds, involving some dramatic – and to some, frankly bizarre – new relationships. For example, they put those masters of the air, the albatrosses, into the same ‘super-family’ as the flightless penguins, and yoked together the apparently dissimilar families of grebes, shearwaters, waders, falcons and flamingos into another massive ‘super-family’ (Ciconiiformes).
Less controversially (at least at the time), they suggested that the bare faces and broad wings of storks and the New World vultures were not, as had previously been thought, the chance result of convergent evolution brought about by their scavenging habits, but had arisen because they were genetically close cousins.
Almost immediately, the backlash began. There was a widespread feeling amongst the wider ornithological community that although the overall aims might be laudable, the pioneering scientific methods being used were simply not accurate enough to support the conclusions. There was also a suggestion that Sibley was manipulating some of his experimental results to fit prior assumptions – either through carelessness or to get the answers he wanted – making it hard to know whether any particular set of results was valid or not.
During the quarter of a century since this bombshell first hit the world of ornithology, some of these radical conclusions have indeed been discredited.14 These include, ironically, the one proposal with which most people had agreed – the placing of New World vultures in the same group as storks. This was later retracted as having arisen from erroneous data, and most authorities now place the vultures back with the Old World hawks and eagles in their original order, Accipitriformes.
Right up to his death in 1998 Charles Sibley passionately continued to defend the new approach, pointing to the fact that many of their findings had been accepted by the wider scientific community.15 These included the once radical idea that Australasia’s songbirds had – like that continent’s marsupial mammals – evolved separately from those found elsewhere in the world, and so despite their superficial differences were in fact more closely related to one another than they were to their Old World counterparts (see Chapter 4). That one turned out to be spot-on.
Today, thanks to the pioneering work by the US evolutionary ornithologist Professor Richard O. Prum of Yale University, the big picture of avian phylogeny – at least at the level of orders and families – is becoming clearer, though there are still many gaps in our knowledge. In a 2015 paper in the journal Nature,16 Prum and his colleagues produced a series of what one of their peers called ‘robust conclusions, representing a decisive technical advance on Sibley and Ahlquist’.17
To the ordinary birder, their findings are actually quite comforting, as they appear to reflect the way we see and understand relationships between birds in the field. For example, one grouping consists of those aerial acrobats the nightjars, swifts and hummingbirds, while two huge groupings are broadly separated into waterbirds (including all diving, wading and shorebirds) and landbirds.
The authors believe that these splits occurred in the wake of the last mass extinction, the Cretaceous-Palaeogene, some 66 million years ago. This was almost certainly the result of a massive meteorite hitting the planet, creating catastrophic changes to the world’s climate and atmosphere, and killing off three-quarters of the plant and animal species on Earth – including the non-avian dinosaurs. This left the playing field clear for new lineages to evolve, including modern birds (or as we should perhaps now call them, avian dinosaurs).18
*
These days, however, the argument about the potential relatedness at the higher level of orders and families has faded into the background. A second revolution has been taking place in the world of ornithology, putting a potentially far more iconoclastic issue under the spotlight: the whole way we decide what defines a species. It has had a profound effect on the names we give to birds, as the result has been a huge increase in the number of different kinds of bird found around the globe.
When I was growing up in the 1960s and 1970s, the number of different species of bird in the world was thought to be roughly 8,600. This figure had been arrived at many years earlier, in 1951, by the German/American evolutionary biologist Ernst Mayr and his colleague, the US ornithologist Dean Amadon. At the time it was generally assumed that this total would increase only very slowly, as a handful of new species (or re-discovered species hitherto thought to be extinct) might still be found in the remote rainforests of South America, Africa or south-east Asia.
For several decades, that’s exactly what happened. Then things began to change, with the number of species suddenly rising dramatically, to more than 9,700.19 As Charles Sibley later noted, only fifty or sixty of the thousand or so new species were previously unknown to science; the rest – around 95% – were the result of a new and radical way of classifying birds.
Before this time, the most widely accepted way of defining a species was Mayr’s Biological Species Concept (BSC). This assumed, as we have seen earlier in the debate around the identification of the willow tit as a British breeding species, that if two different populations of birds overlap in range without interbreeding, they must be two separate species. But a problem arises when there are two populations that look or sound slightly different, but whose ranges do not overlap, which means that the hypothesis cannot be tested in the field.
Under the strict rules of the BSC, it had previously been assumed that these were not separate species, but merely races or subspecies. To get round this problem, Sibley and Monroe made one crucial change. They promoted the ‘allospecies concept’, which gives birds that look or sound significantly different from their relatives the benefit of the doubt, and regards them as separate species – whether or not this can be confirmed in the field.
This may all sound like an arcane and meaningless distinction: the academic equivalent of medieval scholars debating how many angels can fit on the head of a pin. But in fact it was a critical paradigm shift: not just scientifically, but from a practical viewpoint too. For birders all over the world, it has made a huge difference, as it revealed the hidden presence of an extraordinary number of new species – perhaps as many as 1,100.
At this stage it is worth reminding ourselves of the old joke, ‘a species is whatever a competent taxonomist says is a species’.xx Given the ambiguity, even the greatest experts in their field have been unable to agree on a working definition of when two discrete populations of birds can be defined as separate species, rather than just distinctive subspecies.
But whatever the arguments, the general move towards a looser definition of what makes a species has led to an epidemic of what ornithologists call ‘splitting’. This is the process by which what were once considered distinctive races of a single species are ‘promoted’, thus creating two new species where only one had existed before.
This process has always been with us: when I started birding, back in the 1960s, pink-footed and bean geese were still considered conspecific – i.e. two different races of the same species. Soon afterwards, they were each elevated to specific status. Later on, the ‘bean goose’ was in turn split into two full species, known as tundra bean and taiga bean geese. Likewise rock and water pipits were lumped together as a single species until as recently as the late 1980s, despite their very obvious differences in appearance and ecology.
Today more and more species are being split: according to some authorities the familiar brent goose may comprise three different species: pale-bellied, dark-bellied and the North American version known as ‘black brant’.
Further afield, there are now three different versions of the elegant raptor known as black-winged or black-shouldered kite (one in southern Europe and Africa, one in Australia and one, known as white-tailed kite, in the Americas); there are four different darters or ‘snakebirds’, where there were once two; and no fewer than six different versions of the giant cousin of our moorhe
n that used to be called ‘purple swamphen’, found from Spain and Portugal, through Africa and Asia, to New Zealand.
For keen ‘world listers’, desperate to see every species of bird on the planet, this unexpected development has been a mixed blessing. The good news is that there are suddenly all sorts of new birds to go out and see – or if they are lucky, to ‘tick off’ their list while sitting at home, as they have already seen them without hitherto realising that they were actually separate species. But for some, the new approach has proved deeply frustrating, because unless they have kept very detailed notes about the birds on their various travels, and exactly where they came across them, they have found it impossible to work out which of two or more new species they have actually seen.
The upshot of all this is that the current generally accepted total of different species in the world has now reached a figure of between 10,500 and 10,700, an increase of almost a quarter on Mayr and Amadon’s original estimate. But is that the limit, apart from a handful of new discoveries to be made over the next few decades? Or might it be just the beginning of an explosion in the number of different species of bird in the world? I gained a fascinating insight into this issue in 2004, on a birding trip to Morocco.
*
The Dutch birder and ornithologist Arnoud van den Berg knows more about the birds of the Western Palearctic than virtually anyone. So when I accompanied him on a visit to Morocco I expected – and duly received – the benefit of his vast wealth of knowledge about that country’s birds.
What I didn’t expect was that, while we were watching one of the rarest birds in the world, a flock of northern bald ibises, Arnoud would casually point out a small, cormorant-like bird perching on the rocks below. I was even more surprised when he explained that this bird, the North African subspecies of the European shag, was potentially the most endangered bird we would see on the trip, with a global population of just a hundred pairs.
g Another member of our group tactlessly pointed out that it was ‘not a real species’, which prompted a characteristically thoughtful response from Arnoud. He patiently explained that, given the difficulty in trying to define what is a species and what is not, we should perhaps take a different approach. In his view, we should simply focus on each distinctively different kind of bird – whether we consider it a subspecies or a full species – and do our best to conserve it. That way we would ensure that we retain avian biodiversity, and can leave future generations to argue whether or not a particular bird is indeed a different species from its cousin.xxi
Arnoud van den Berg is not the first naturalist to take this approach to classification and taxonomy. Back in the late nineteenth century, the Victorian ornithologist Richard Bowdler Sharpe, curator of the bird collection at the British Museum of Natural History, took a similarly pragmatic view. Indeed, he went even further: by ignoring the distinction between species and subspecies, he concluded that there were almost 19,000 species of bird in the world.xxii
Astonishingly, some ornithologists now believe that Bowdler Sharpe may have been right all along. In December 2016, a new study led by the highly respected American Museum of Natural History (AMNH) suggested that there could be as many as 18,000 different species of bird on the planet – close to Bowdler Sharpe’s estimate, and not far off twice the current accepted total.20 The authors based their conclusions on the concept of ‘hidden avian diversity’; the idea that there are many birds out there that look so similar to one another that they have previously been either ignored, or thought to be subspecies.
Considering that birds are probably the most studied group of wild creatures in the world, this seems hard to believe. But perhaps we shouldn’t be so surprised: after all, as recently as 1999 scientists discovered that the small bat known as the pipistrelle – one of Britain’s most widespread mammals – was in fact two totally separate species, each echolocating using different sound frequencies.
In some ways this is also the logical consequence of the new way of looking at what makes a species. As Charles Sibley remarked, ‘“Splitting” and “lumping” will continue as we try to make nature fit our concepts. Of course, we like to think that we are making our concepts fit Nature.’
Fundamentally, this comes down to the mismatch between the way we try to classify the natural world and the way it actually arose – something that, of course, we can never truly know or understand.
*
So where does this leave English bird names? Well, for a start, it would suggest that if there really are almost twice as many species out there, then we are going to need an awful lot more names. Perhaps fortunately, this is not quite true: in many cases the various subspecies already have a perfectly serviceable name, which can be adopted when the species is split.
Take the case of one of North America’s most familiar and colourful birds, the Baltimore oriole. The Baltimore Orioles baseball team, based in Baltimore, Maryland, was named in the 1950s after the national bird of that state – a striking, orange-and-black songbird. But in 1973, to the horror of the team’s many fans, the AOU decided to lump this species with its western counterpart, Bullock’s oriole,xxiii into a single species, which was given the deeply unimaginative name ‘northern oriole’. The reason given for the change was that, in the zone where both forms encountered one another, ornithologists discovered that they were interbreeding and perhaps producing fertile offspring – a classic sign, at least according to the more traditional BSC, that they must be the same species.
Fortunately for devoted followers of the Baltimore Orioles, the story has a happy ending. Later research by a team of Canadian ornithologists showed that while there was some interbreeding between the two forms, it was nowhere near as frequent as had previously been thought, so in 1995 they were separated once again, and given back their old (and to my mind far more interesting) names.xxiv
In other cases, when a species new to science is discovered, it is often named after the locality where it lives – especially when it has a very limited territory, as many of them do. Leafing through the final ‘Special Volume’ in Handbook of the Birds of the World,21 published as a supplement to the series in 2013, we find that of the eighty-four newly discovered species featured, the vast majority are named after the location where the bird was found.
These areas range from the geographically large, as with the New Zealand storm-petrel, to the tiny, such as the Rubeho forest-partridge, known only from a single, densely forested mountainside in northern Tanzania. Other species named after very localised place-names include the Delta Amacuro spinetail, from the Orinoco Delta in Venezuela, and the Acre antshrike, found only along a short section of Brazil’s border with Peru.xxv Another recently bestowed bird name, Jocotoco antpitta, sounds as though it comes from a location, but ‘Jocotoco’ is actually the name given by the local indigenous people to the bird – a useful reminder that although we often consider a species to be ‘newly discovered’, it may in fact have been known to its human neighbours all along.
This new way of looking at the relationships between birds does have other, more serious implications for the names we give to birds. On a broader scale, falcons have recently been found to be far more closely related to parrots (and adjacent in the classification system to songbirds) than they are to other diurnal birds of prey such as hawks, buzzards and eagles.
So can we still refer to falcons as ‘raptors’, or indeed ‘birds of prey’? And if we do continue to do so, why do we not include owls under the same term?xxvi After all, they are as closely – or distantly – related to the ‘true’ raptors as falcons are, and early ornithologists used to classify all three groups under the wonderfully evocative term ‘rapacious birds’.
At a more detailed level, the latest DNA studies have revealed some surprising relationships – or perhaps that should be ‘lack of relationships’ – within one of our best-known bird families, the warblers.
We all know – or at least think we know – what a warbler looks like: in Britain at
least, they are (mainly) migratory, insectivorous songbirds, often rather drab in plumage and hard to see, so that we tend to tell them apart by their songs. But as I look through the latest version of the BOUs’ British List (updated December 2016), I see to my surprise that Cetti’s warbler is now separated from the rest of its tribe by the long-tailed tit.22 So does that mean that Cetti’s warbler is not a warbler, or that the long-tailed tit is, or that neither species belongs to the warbler family?
Some authorities go much further, splitting Sylviidae, which used to include all the European warblers, into five separate families. These are the ‘leaf warblers’ (willow, chiffchaff etc.), ‘bush warblers’ (an African and Asian group, of which Cetti’s is the only European member), ‘grass warblers’ (such as grasshopper and Savi’s), ‘marsh and tree warblers’ (from the genera Acrocephalus and Hippolais, including reed, sedge and icterine) and finally the ‘true warblers’, of the genus Sylvia, whose members include our familiar whitethroat and blackcap. As if that wasn’t confusing enough, this last group may prove to not actually be warblers at all, but members of the babbler family.
If you are feeling a little baffled, then you had better get used to it, for this is the future of ornithology. In the meantime, we can either spend time arguing over whether ‘warblers’ actually exist, or take a more relaxed view, and continue to lump all these birds under that convenient (though perhaps scientifically inaccurate) linguistic term.
One result of all this change and upheaval, however, has been a very sensible proposal to standardise the sequence in field guides. When I was growing up the ‘Wetmore Order’ prevailed (established in 1930 by the US ornithologist Alexander Wetmore, and revised again in 1951 and 1960), with divers and grebes at the start and buntings and sparrows at the end. This had in turn displaced the sequence used in The Handbook of British Birds, published at the start of the Second World War, which was more or less in the opposite order, placing crows at the start and gamebirds at the end.