In the last decades science has been pervaded by the Popperian paradigm: the acquisition of knowledge proceeds by hypothesis testing. Hypotheses cannot be verified; they can only be falsified. It was, and still is, the same in statistics, much before Popper. You never demonstrate that your hypothesis is true, you always find that the probabilities of its contrary (its falsification) are very low. So you keep your hypothesis until you find it wrong. If your hypothesis is proven wrong, even once, you reject it (since it has been falsified) and try to find a better one. All crows are black! No matter how many black crows you find, if you find a non-black crow you have to reject the hypothesis! What is a better hypothesis on the colour of crows if you find an albino crow? Well…. You can invoke genetics, but then you are begging for ad hoc explanations! You build up another hypothesis to explain the exception so to save the rule, but it is not worth trying, the rule is false! Strange enough, the case of crows is not solved then: what is the law governing the colour of crows if you find an albino crow? It is impossible to find such law; there is no law on the colour of crows. So studying the colour of birds is not science because you cannot find a single statement describing it? Apparently so! But then, why choose this example to explain how to make science?
In another article, the example has been changed, and the law on the colour of crows has been turned into a law on shark distribution: there are no sharks in this bay. No matter how many times you will not find a shark in the bay, you will never be sure that sharks are not there, but if you find a shark the law is falsified and is to be rejected!
An embarrassing thing in Popperian science is that you cannot verify things, you can only fail to prove them false. But then, let’s take the shark example and let’s turn it the other way round: there are sharks in this bay. No matter how many times you will not find a shark in the bay, you will never be sure that this law is false, but if you find a single shark you have VERIFIED your hypothesis! There is hope, we can verify things!
What is the difference between there are no sharks in this bay and there are sharks in this bay? The first is a universal statement, it covers ALL sharks: the shark universe is out of the bay. If one arrives, the universality of the law, exemplified by NO SHARKS, is falsified. The second is an existential statement, it hypothesises the existence of sharks in the bay. The existence of sharks in the bay is verified even by a single finding: there ARE sharks in the bay! Of course universal statements are true laws and are much stronger than existential ones: a science based on universal laws is surely very hard. Physics is very hard, even though it is turning a little softer in recent times!
What about biology? Can we find universal laws in biology? Is the Popperian paradigm applicable to the sciences of life?
Let’s go back for a while and see what happened in the most philosophical branch of biology: evolutionary biology. One supposed law of evolutionary biology was gradualism. Then saltationists came out. This happened in the midst of Popperian enthusiasm. Gradualism and saltationism were not proven true by the observations of cases of either gradual or saltational evolution. If a gradualist demonstrated a case of gradual evolution, he/she claimed to have falsified saltationism, and of course saltationists rejected gradualism with every example of saltational evolution. We have to live with it; there is no universal law of evolution. Evolution can proceed both gradually and by jumps, and maybe there are other patterns that we are not aware of. There IS both saltational and gradual evolution. So what? Is evolutionary biology softer than physics because it has no universal laws? Of course it is softer, if you like this way of rating science, but this does not imply a lower rank in terms of dignity. Biology is much more complicate than physics and the more complicate a science becomes, the less predictive and law-conducive it becomes. This concept is already contained in physics, with the principle of indetermination. Why do biologists have to envy physics? Because of its laws? Physics envy is not justified. Physics has laws because it is simple. It uses an abstruse terminology, with lots of mathematics, but it deals with the simplest level of organisation of matter. Biology deals with the most complex object of the universe: life. Life is baroque. It obeys the laws of physics, even though one might argue that, with mathematics, one plus one is two, whereas in biology one plus one can become three, or even more (as every couple of parents know very well). The laws, however, govern the lower level of life organisation. Both a sheep and a panther respond to the laws of thermodynamics, but having to deal with them makes it clear that the laws of physics are of no help if you want to survive! Especially if you cannot tell a leopard from a sheep.
As Mayr has claimed one million times, biology in general is a historical science. There is strong correlation in events that are very near in time (one generation is almost identical to the preceding one) and one is the direct cause of the other. But if the time frame is widened, the correlation disappears and even though each generation is almost identical to the preceding one, after one million generations we can have enormous differences! In this framework, it is easy to claim that contingency is the main rule governing evolution and, thus, biology. The change can occur either gradually or saltationally, but you cannot predict its outcome anyway. Little events, like the beat of the wings of a butterfly at Bombay, can cause enormous effects, like a thunderstorm in New York, if given enough time to exert their influence, and life sciences are surely the most chaotic of all sciences. But chaos implies attractors, so that the impredictability of contingency is nevertheless limited by constrain. Contingency led to the almost complete disappearance of brachiopods from marine soft bottoms, constraint made so that their place has been taken by another group of bi-valved filter feeders: the mollusc bivalves! So both contingency and constrain laws explain part of the picture, both are right and it does not matter if both are wrong. We have to learn what is the good law case by case.
It is true that there are recurrent events, just as with history (as explained by G.B. Vico): but no historian is so stupid to think that it is possible to find a nice equation that will predict the outcome of history! Horoscopes pretend to do so, and if one would ask predictivity to historians they would become angry at the question, being offended because asked to behave like future tellers. Sure, historians can make weak predictions, inferred by what they saw in the past. The US are like the Romans, Europe is like the Greeks. The Greeks developed culture, the Romans developed technology, the Greeks became rapidly weak, the Romans ruled the western world for centuries, but then they fell. Like the Egyptians, like the Aztecs, like all the great peoples of the past. Why not the US? It is not a matter of what will happen, it is a matter of when and how. We can easily predict that, sooner or later, the US will fall. Maybe they will last many other centuries; maybe they will dissolve just like the USSR, in no time! Nobody knows. There is no magic formula leading to such a prediction. We can only say that it was always like this and that chances are very good that it will happen again.
If history is not predictable and there is no equation able to describe and predict history, then what can we do with biology, a clearly historical discipline? This leads to the question: is mathematics, clearly the language of physics, a proper language for biology? Can we hope to find equations describing the future of biology, so to make predictions? Strange enough, chemistry, biochemistry and cell biology do not use mathematics so much, whereas genetics, population biology and ecology use a lot of mathematics. Evolutionary biology based on genetics uses mathematics, whereas when the base is palaeontology mathematics disappears! Apparently there are fields of biology that are tackled by mathematics and fields that obviously are not. But are these domains really prone to be treated mathematically? Apparently genetics is, but only if gene action is discrete. What about ecology? The dawn of modern ecology is the mathematical treatment of the relationships between fluctuating species. The equations of Lotka and Volterra (a case of parallel discovery just as Wallace and Darwin’s) describe the variation in number of the populations of two interacting species. The illusion, looking at the equations, is that if you measure the quantities of the two species at a given time, then you can predict the numbers of individuals at another time, in the future. Such predictions, essential to the management of populations, are almost invariably wrong! The reason is simple: there are no systems with two interacting species! There are no systems with three interacting species! Ecological systems imply the interaction of hundreds of species and, at a global scale, all species interact at a certain point! Poicaré, much earlier than Lorenz, discovered chaos with the problem of the three bodies. If there are more than two interacting bodies in a given system, then the long-term dynamics of the system becomes mathematically intractable.
Predictability, then, is an illusion if the investigated system is complex. You can try with reductionism, and deal with couples of features, pretending that the rest of the world does not vary. But sooner or later you have to assemble the couples and, as is written in the first chapter of every ecology textbook, ecology is the science of the emerging properties, and the features of the world are not the simple sum of the features of its constituents, just as a masterpiece of literature is not simply the sum of the words that make it up (otherwise we would be satisfied in simply reading the dictionary). In a complex world only weak predictions are possible, those falling within the upper and lower range of the attractor of the system. Of course we can have short-term predictions that are mostly accurate if we want to predict weather, but the chances of having good predictions in systems in which weather’s complexity is coupled with life complexity are simply desperate. The coupling of weather and life is what ecology is mostly about. Ecology cannot be a predictive science, if we try to say something more than just apply thermodynamics to ecosystem functioning. With thermodynamics, biodiversity becomes irrelevant, but we know (even if we do not understand completely its value) that this is simply incorrect! Black boxes have been useful to understand great processes, but now they have to be opened and looked into, if we want to understand more.
The evolution of diversity
If life is monophyletic, in the beginning there was a single species. Probably a chemiosynthetic and decomposing moneran, able to close the circle of life by itself. With a single species, biodiversity was minimal, but life worked anyway. After that first period, one can imagine that a Red Queen game started. The single species became two and both started to compete with each other, one causing the change of the other and viceversa. Then a third arrived, and a fourth, and the system became more and more complicate. With a chain reaction, life itself is the reason of its complexity and diversity. It might work also in a simple way, it did in the past, but it now works in this way. As simple as that. It is true that the function of ecosystems can be reduced to black boxes in which diversity is irrelevant, but diversity is relevant for the way life is NOW. And we are the products of this diversity and complexity. Everything might work anyway even going back to a single-species system, but that species would not be something within the range of diversity that we can actually see with our eyes. Diversity is valuable by itself, and the meaning of our existence resides in the past history of this diversity. We are the product of it. Preserving biodiversity is preserving our history and posing the base for our future. Life is not endangered by our action, the premises of our existence are. Life can withstand terrible extinctions, and we know from the history of life that the outcome of these extinctions was usually very different from what was out there before.
Conserving biodiversity
Conservation ecology is a mixture of genetics, ecology and evolutionary biology and is becoming a most integrated discipline. Can we hope to be able to say: if we do this then we will have this result? Maybe we can have such result in the short term, but conservation is by definition "preserve something in the long term". The aims of conservation ecology want long-term predictions. And a contingency ruled world teaches us that these are impossible. We can only try some weak predictions.
Decision-makers want us to produce strong predictions, if we do not, here are the engineers, the architects, the self-taught ecologists (often coming from the physics world) who come out with their answers. So we play the game, just to survive. If we say that we cannot predict the future, then somebody else will be listened to, so it is better if we pretend to be able to do so. Let’s invent more and more algorhytms, let’s test tem, let’s reject them when they fail and look for some more. Let’s formalise the obvious in mathematical terms, and we’ll have our secret jargon just like those who practice physics! They are so respectable, and nobody dares contradict them, because nobody but them can speak that jargon. Let’s play the game!
All biologists, deep in their heart, feel that all that mathematics does not belong to them. Some like it, and show it off. To say something into mathematical terms gives such an allure of scientificity! It makes biology look like physics! The poor guys who do not understand do not dare saying that the envelope is nice but the box is empty, they do not dare saying that the formula has just enough predictive value as a simple verbal description. Why make things complicate when they can be simple? Read Darwin. His books changed the world, and there is not a single formula.
The outcome of all this arguing is not that hypothesis testing is not good, that we can survive without mathematics and be finally happy. Unfortunately not. We must still build up hypotheses, we must use statistics to test them, and we can even try to perform experiments and not simply observations to test our hypotheses. But we have to realise that this is not what bio-ecology is all about. To perform proper ecology, evolutionary biology, conservation biology, we need more knowledge in the field of natural history. We have to build up complex verbal and graphical models, identifying all the interacting variables. This kind of exercise is unfashionable, also because the variables are species, and there are very few people who recognise species: less and less. The validity of a theoretical model is to be measured by the value of its level of understanding of natural history. We need rules (that will be invariably broken but that will hold anyway, at least statistically) but we need cases, we need generality and we need detail. In this way we will not build up an elegantly simple building, we will build a baroque, Gaudittian temple. Life is monotonously linear at a certain level, just four bases and you have DNA, but suddenly it becomes overwhelmingly complex. The guys who passed to biology from physics did not discover universal laws, and the reason is simple: there are no universal laws in a complex world. And this is what makes biology so fascinating. Some will say that if it is not expressed with formulas, then it is not science. So what?