The Lagoon

by Armand Marie Leroi

  They saw him as a proto-Linnaeus constructing classifications by the Aegean shore. A gifted one, too. Cuvier, characteristically, was fulsome in his praise:

  Aristotle, right from the beginning, also presents a zoological classification that has left very little to do for the centuries after him. His great divisions and sub-divisions of the animal kingdom are astonishingly precise, and have almost all resisted subsequent additions by science.

  This, of course, is mere hyperbole. Cuvier himself constructed a classification of the animals that was vastly superior to Aristotle’s, one in which the Greek’s major divisions were added to, subtracted from or simply abandoned so that hardly one survived intact. Still, hagiography aside, the view that Aristotle’s project was fundamentally a taxonomic one has progressivist appeal. After all, a science can hardly get off the ground unless its objects have first been pinned down and named. As biology needed Linnaeus’ system, so astronomy needed Johan Bayer’s star atlas, crystallography the Abbé Haüy’s geometries, and chemistry Dimitri Mendeleev’s Periodic Table. But why talk only of science? No sooner did He make the animals than He gave them to Adam to name – even God sees things that way.

  Most of Aristotle’s kinds – genē – correspond roughly to our species. Erythrinos, perkē, skorpaina, sparos, kephalos can all be matched to one or a few modern species of fishes. Sometimes, however, his kinds designate our breeds or varieties: ‘there are several kinds of dogs . . .’ – the Laconian and the Molossian hounds. His animal names were not, as far as we can tell, invented technical names of the sort that Linnaeus devised for his species. Rather, they were the vernacular zoology of his day: he got them from the fishermen, hunters and farmers that he talked to. ‘Near Phoinike [Lebanon], there are crabs that are known as hippos – horse – because they run so fast that they’re hard to catch’ – thus the ghost crab whose Latin binomial, Ocypode cursor, means ‘swift-footed runner’. ‘There is a rock-bird called kyanos [“blue”]. It is most common in Skyros and it spends time on rocks. It is smaller in size than a kottyphos [blackbird], a little bigger than a spiza [chaffinch]; it has large feet and climbs on the face of the rocks; is entirely deep-blue; its beak is thin and long, its legs short, and like those of the hippos [woodpecker]’ – probably the rock nuthatch. The fact that, for Aristotle, a hippos can be a crab, a bird and a horse doesn’t make his zoology any easier to read.*

  There is a widespread, and rather romantic, notion that the fishermen and hunters of traditional societies are extraordinarily skilled taxonomists, able to distinguish at a glance species that mere scientists struggle to tell apart. New Guinea Highlanders are said to identify unerringly 136 different kinds of birds. Maybe they can. Modern Greek fishermen, however, appear to be much less gifted when it comes to identifying their fish. There is no reason to suppose they were ever any more so.

  We were at Skamanoudi, a tiny port on the eastern shore, where, we were told, the remains of ancient Pyrrha’s harbour could be seen just off shore when the light is right. But the wind was up, the Cape exposed and whitecaps obscured the view, so we sat down and ordered some ouzo and a plate of salted fish. Someone said how good the papallinas were. David K., the expedition ichthyologist, demurred. You mean sardellas, he said. The sardella is Sardina pilcharus, the papallina is Sprattus sprattus, he continued – and, to prove the point, he produced Ta psara tis Helladas (The Fishes of Greece) of which he is authorially proud and rarely without, so that we could see his beautiful gouaches of two practically identical fishes.

  We asked the proprietor. They’re papallinas, he said. But, we pointed out, the menu says sardellas. Of course – a papallina is a sardella inside the Lagoon and a sardella is a papallina outside, and these ones came from inside which is why they’re so good. An adjacent table of fishermen intervened. The proprietor had not spoken truthfully or at least not comprehensively. The sardella and the papallina were indeed one and the same species, but the essential difference lay not in their geographic origins, but in their age or perhaps simply their food – but whether one or all of these factors was salient, none could agree. Some contrarians took the scientific view. The sardella and papallina were, they said, quite different species, just as the kyrios said: anyone could taste the difference.

  The diversity of their views on the relationship between the two fishes, and which of them we were eating, was puzzling. Kalloni exports thousands of tons of sardellas or papal linas or at least little silvery fish annually; there’s no Greek supermarket in which they cannot be bought, and you’d have thought that the men who fish them daily would have arrived at a taxonomic consensus. They’ve had a very long time in which to work one out. Does Aristotle recognize the inherent ambiguity of vernacular names? Perhaps. His faith in the zoological prowess of fishermen is limited, and he certainly sees that folk-taxonomies do not capture the diversity of life: ‘the other kinds [of karkinoi, crabs] are smaller in size and don’t tend to have special names’. But he never remedies the deficiency.

  Even so, many of Aristotle’s kinds can be convincingly identified with modern species, among them: dogs in toto, horses, two cicadas, four woodpeckers, six sea urchins and humans. Unsurprisingly, he’s very good on cephalopods, naming the polypdōn megiston genos (common octopus), the heledōne/bolitaina/ozolis (musky octopus), the sēpia (cuttlefish), the teuthos (sagittal squid), the teuthis (European squid) and the nautilos polypous (paper nautilus). He also speaks of another shelled cephalopod ‘that lives in its shell like a snail, sometimes protruding its tentacles’. The identity of this creature has been much disputed. It would be a lovely description of that exquisite mollusc, the chambered nautilus, were it not for the fact that Nautilus pompilius lives in the Indo-Pacific west of the Andaman archipelago – very far outside Aristotle’s range. Some scholars have suggested that he took his description from a specimen seen by someone who had accompanied Alexander the Great to India; others that he is referring to the male of a pelagic octopus, Ocythoe tuberculata, that makes its home inside a salp’s test or else to a pelagic snail, Janthina janthina, that doesn’t look like a cephalopod at all. None of these seems particularly plausible and the identity of the ninth cephalopod remains unknown.

  Aristotle also recognizes larger groups that resemble modern higher taxa such as Genera, Families, Orders, Classes and Phyla. He calls them megista genē – ‘greatest kinds’. Some of their names are obviously vernacular too: ornis (bird), ikthys (fish). But others were apparently invented as part of a technical vocabulary. Aristotle saw that folk-taxonomies aren’t much good at classifying animals into larger groups, especially when the animals in question are the sort that most people ignore. The names of his greatest kinds often have a descriptive air: malakostraka (‘soft-shells’ = most crustacea), ostrakoderma* (‘hard-shells’ = most echinoderms + gastropods + bivalves + barnacles + ascidians), entoma (‘divisibles’ = insects + myriapods + chelicerates), malakia (‘soft-bodies’ = cephalopods), kētōdeis (‘monster-like’ = cetaceans), zōotoka tetrapoda (‘live-bearing tetrapods’ = most mammals), ōiotoka tetrapoda (‘egg-laying tetrapods’ = most reptiles + amphibia), anhaima (bloodless animals = invertebrates), enhaima (blooded animals = vertebrates).

  Aristotle seems to have believed that, in a good classification, kinds are subordinate to kinds and that each kind has a unique, defined position relative to all others – in other words, that they should be arranged as a nested hierarchy. ‘The most important kinds of blooded animals are egg-laying tetrapods, live-bearing tetrapods, birds, fish, cetaceans and any that are unnamed because the group does not exist, merely the simple form in each individual case.’ ‘We must now speak of bloodless animals. There are several kinds’ – which he then lists. ‘There are four greatest kinds of soft-shells: they are called astakoi, karaboi, karides and karkinoi’ – which tells us that lobsters, crayfish, prawns and crabs are subordinate greatest kinds within an even greater kind, the soft-shells. But some of his hierarchies are very shallow: man is a blooded animal but otherwise stan
ds alone.

  It is now obvious that animal relationships should be described as a nested hierarchy. It is the only way to describe the topology of a tree graph using words; and a tree graph is the only way to describe descent by modification from a common ancestor using a picture. But if it’s obvious to us, we may wonder why it was so to Aristotle, who, after all, never read the passage in the Origin where Darwin explains it. (‘The affinities of all the beings of the same class have sometimes been represented by a great tree. I believe this simile largely speaks the truth . . .’) There are, after all, logical alternatives. Aristotle could have built a classification from taxa that were quite independent of each other. In his delightfully disingenuous description of the Heavenly Emporium of Benevolent Knowledge, a Chinese encyclopaedia, Borges tells of one in which each taxon is defined by features such as belonging to the emperor, being embalmed, being a mermaid, being a stray dog or resembling, from a distance, a fly. Aristotle could also have built one from purely orthogonal, rather than nested, taxa. In Politics III, 7 he classifies forms of government in this way on the basis of two features, the degree of concentration of power and their quality:

  In the event he did not apply this structure to animals.

  Perhaps it is just obvious to anyone who studies the diversity of life with care that it should be arranged hierarchically. Linnaeus didn’t need Darwin to tell him to put his animals into genera, orders and classes. Aristotle’s term for ‘taxon’, genos, is also inherently hierarchical for it originally meant ‘family’, by which the Greeks meant a patrilineal clan. But it’s also true that nested hierarchies emerge naturally from his classification method.

  Aristotle’s classification of the animals was probably the first.* But classification is very close to definition, and definition was an Academic obsession. Plato thought that to define something was to understand it. His method of definition entailed successive dichotomous division of the thing’s features. Investigating the nature of monarchy in the Statesman, he began with ‘all human knowledge’, which he then divided into successive specialist branches of knowledge until he could show that a king is a kind of herdsman. But what does a king herd? To find out Plato divided the animals successively by their various features and concludes that a king is a herder of the tame, hornless, featherless bipeds more commonly known as humans. Plato acknowledged that any given class of activities, people or animals can be divided in many different ways yielding many possible definitions (he gave eight or so pedigrees for the sophists, most of them aimed at defining them as unsavoury, mercenary corruptors of youth). Even so, he argued that one can ‘pull together the threads’ of the various definitions and so discern the nature of the beast. His later dialogues show signs of definitional mania.

  In the Metaphysics and Posterior Analytics Aristotle tweaks Platonic division a bit; in Historia animalium and The Parts of Animals he transforms it. He widens its object to embrace classification and subjects its method to a withering assault. He gives many reasons why dichotomous division won’t work, but the most telling is the arbitrary nature of the results. Plato divided animals into ‘water-dwellers v. land-dwellers’ and ‘gregarious v. solitary’ and ‘tame v. wild’, which is all very well except that, whichever of those you choose, birds will end up in both sub-groups and that doesn’t seem right. Living things, Aristotle saw, have a deep, natural order that a good classification should reflect; when dividing, he says, ‘one should avoid tearing each kind apart’. Actually, Plato expressed the same thought more stylishly: ‘we shouldn’t cut across the joints like a clumsy butcher’ – a wise precept that he invariably ignored. It is also one that prompts the question: how do we find nature’s joints?

  XXVIII

  THE WHOLE PROBLEM is that they can be hard to see. Aristotle has much to say against Plato’s approach but less about his own. Nevertheless, his practice and various programmatic passages speak of a sophisticated method of division, one that rests upon two important insights.

  The first of these is his recognition that animal features vary at different scales of nature’s hierarchy. The diaphorai – differences – between kinds within a given greatest kind, between, say, a sparrow and a crane, are relatively subtle. They share the same basic body parts, differing only in shape and size. His term for such variation is ‘the more and the less’:

  Differentiation between bird [kinds] involves excess and deficiency in their parts and is a matter of the more and the less. Some are long- or short-legged, have broad or narrow tongues and so on for the other parts.

  Much of his descriptive biology, then, is about how beaks, bladders, bowels and brains vary in size and proportion.

  The differences between the greatest kinds, say, birds and fishes, are much more radical. They lie in the kinds of parts that animals have and their arrangement. They are architectural. Modern zoologists speak of a ‘body plan’ from the German Bauplan; Aristotle doesn’t have an equivalent term, but he uses the concept. The relative position of hard and soft parts and the number of legs are particularly important. Some soft-bodies (cephalopods) have a hard internal structure (a squid’s quill and cuttlefish’s cuttlebone),* but soft-shells (crustaceans) and hard-shells (snails, clams, sea urchins) have an external hard part or, as we would say, an exoskeleton. Fish have no legs, humans and birds have two, tetrapods have four, divisibles and soft-bodies many.

  Greatest kinds also differ in their geometries. For Aristotle an animal has three axes with six poles: above–below, before–behind and left–right.* Above is the pole of an animal that takes in nutrition, below is that which expels it; before is the pole towards which an animal’s sense organs face and the direction in which it moves, behind is its opposite; his right and left are the same as ours. This geometry is based on humans and distinguishes them from tetrapods. In a tetrapod, above (location of the mouth) and before (orientation of the sense organs) are the same pole, and below (location of the anus) and behind (opposite to the sense organs) are too. This is one reason that Aristotle doesn’t classify us with the live-bearing tetrapods (mammals).

  Modern zoologists will find his way of geometrizing bodies rather odd.* But there’s no reason why he should do things as we do. And it does give Aristotle a genuine insight into the weird geometry of cephalopods. Since their feet are arrayed around their mouths and their guts are twisted into a U, Aristotle asserts that the cuttlefish has the geometry of a tetrapod that has been bent double so that its above and below and before and after all meet in the same place.

  That’s rather brilliant.* But his geometry also leads him into some less perspicacious claims. Ignorant of photosynthesis, he imposes an animal model of nutrition on to plants. Plants, he thinks, get their nutrition through their roots which, therefore, are analogues of animal mouths. They must also excrete something at the opposite end – namely, fruit. These analogies lead him to the conclusion that the above end of a plant is buried in the soil, while its below end bends in the breeze.

  But Aristotle doesn’t just delineate his greatest kinds by their body plans. He also asks whether or not they share the same kind of body part. He appropriates an existing term, analogon – analogue. He never defines it but his examples suggest that he means something like ‘a part in one kind of animal that has the same function or position as a part in another animal yet is different in some fundamental way’. The term is mathematical in origin: ‘as A is to Y, so B is to Z’. In his zoology he applies it metaphorically: ‘as feather is to bird, so scale is to fish’. If two creatures have analogous parts then they belong to different greatest kinds. Analogues differ in their fine structure or physical properties. Crabs and snails both have hard external parts, but step on a crab and its carapace is crushed; step on a snail and its shell shatters. Carapaces and shells must, then, differ in some fundamental way* and so, too, must the kinds that possess them.

  Aristotle identifies quite a few analogues. Live-bearing tetrapods, humans and dolphins have skeletons made of bone, but fishes, sharks, cuttl
efish and squid have bone-analogues: ‘fish spine’, cartilage, cuttlebone and quills. All of these structures have the same function: to preserve and support the soft tissue. Bird feathers and fish scales are both obviously coverings. Blooded animals have hearts, but bloodless animals – particularly cephalopods – have something analogous to blood and something analogous to a heart to handle it. Lungs are analogues of gills. Sometimes he seems unsure whether two parts are analogous or really just variants of the same thing. In one passage he says that the cephalopods have only a ‘brain-analogue’; in others that they have ‘a brain’ which seems to imply an identity with the tetrapod brain.* Aristotle did not invent an antonym of analogon for parts that ‘are the same’, but only speaks, with considerable ambiguity, of parts that are the same ‘without qualification’. It was only in 1843 that Richard Owen filled the terminological gap with ‘homologue’. Aristotle probably thinks that most of the internal organs of vertebrates are homologous in this pre-evolutionary sense; at least he speaks of hearts, stomachs, livers, gall bladders, etc. in egg-laying tetrapods, live-bearing tetrapods, birds and fishes without qualification.

  THE GEOMETRY OF LIVING THINGS AFTER THE PROGRESSION OF ANIMALS 4

  Smaller kinds – breeds, species – are, then, distinguished by variety in the size and shape of the same parts; greatest kinds – higher taxa – are distinguished by variety in body plan and analogy of parts. To put it more abstractly, Aristotle adjusts the weight of his features to the scale of his classification. This logic is still the basis of modern systematics. Yet his sense of unity beneath diversity is often acute. He recognizes that his terms ‘analogy’ and ‘the more and the less’ are ambiguous. When, after all, does a quantitative difference become so pronounced that it becomes a qualitative one? Compare the skeleton of, say, a cow with that of a sardine and the distinction between true bone and fish spine seems clear (at least to him): they are analogues. But, as Aristotle notes, while birds and snakes generally have bone, small birds and snakes have something more resembling fish spine. In such animals he observes ‘nature makes a transition by small steps’. He recognizes that the boundaries between his greatest kinds are not sharp, but shade into one another. Speaking of snakes and lizards he says, ‘Serpents as a kind have parts comparable to lizards (if you increase their length and take away their feet),’ and he even calls them syngennis – kin. And seals may live in water yet their flippers are just odd limbs; they are, he says, ‘imperfect’ or ‘crippled’ tetrapods.