Darwin

Tim Lewens

• Publisher: Routledge
• Year: 2006
• ISBN:
  9781134280100
• Edition: 1
• Topic: Philosophy
• Subtopic: Philosophy History & Theory

In this invaluable book, Tim Lewens shows in a clear and accessible manner how important Darwin is for philosophy and how his work has shaped and challenged the very nature of the subject.

Beginning with an overview of Darwin's life and work, the subsequent chapters discuss the full range of fundamental philosophical topics from a Darwinian perspective. These include natural selection; the origin and nature of species; the role of evidence in scientific enquiry; the theory of Intelligent Design; evolutionary approaches to the human mind; the implications of Darwin's work for ethics and epistemology; and the question of how social and political thought needs to be updated in the light of a Darwinian understanding of human nature. A concluding chapter assesses the philosophical legacy of Darwin's thought.

Darwin is essential reading for anyone in the humanities, social sciences and sciences seeking a philosophical introduction to Darwin, or anyone simply seeking a philosophical companion to Darwin's own writings.

Four
Evidence

1. SCIENCE AND GOD

Darwin framed the first edition of the Origin with two epigraphs, both from people who are probably best known now for their contributions to the study of scientific method. One is from Francis Bacon (1561–1626), a man famous for his insistence on scrupulous and exhaustive observation if scientific knowledge is to be acquired, and who, it is said, died from a cold following an experiment in which he stuffed a chicken with snow:

To conclude, therefore, let no man out of a weak conceit of sobriety, or an ill-applied moderation, think or maintain, that a man can search too far or be too well studied in the book of God’s word, or in the book of God’s works; divinity or philosophy; but rather let men endeavour an endless progress or proficience in both.

(Bacon: The Advancement of Learning)

Bacon was a highly regarded figure among many of Darwin’s most influential contemporaries. They thought of him as the father of the ‘inductive method’ in science; that is, a method that places great store on the meticulous gathering of numerous experimental results before confidence is placed in any hypothesis. More specifically, this method warns against leaping to general theoretical positions simply because they are consonant with some small number of observations.

The ‘Baconian’ method is sometimes caricatured as one that exhorts the scientist to begin their work by accumulating diverse facts, thereby allowing observations to speak for themselves without the distorting bias that theoretical presuppositions might bring. Darwin claimed, rather unconvincingly, to have fashioned his theory in this way: ‘I worked on true Baconian principles, and without any theory collected facts on a wholesale scale’ (Autobiography: 72). But thoroughly Baconian science of this kind is almost impossible to conceive, and hardly seems commendable. The philosopher of science Karl Popper is particularly well-known for deriding it. He was fond of going into classrooms full of students and asking them to ‘observe’. Unsurprisingly, the students were confused – how could they observe unless they had some idea of what they were supposed to be observing? Popper used this baffling request to make the point that naive observation, unclouded by any theoretical assumptions, is impossible, for we need some kind of theory to tell us where to look, and how to interpret what we see. Darwin’s notebooks make it clear that he was no naive Baconian of this sort: he designed experiments in order to test the presuppositions of his transmutationist views. But it was important for him to be seen to credit Bacon, in order to signal to Bacon’s Victorian admirers that Darwin’s theory was built on a firm empirical foundation, and that unlike Lamarck and Chambers, he should not be viewed as one prone to wild flights of fancy.

The Origin’s other epigraph comes from one such influential Victorian Baconian – William Whewell:

But with regard to the material world, we can at least go so far as this—we can perceive that events are brought about not by insulated interpositions of Divine power, exerted in each particular case, but by the establishment of general laws.

(Whewell: Bridgewater Treatise)

As we saw in chapter one, Darwin knew Whewell from his days in Cambridge. Philosophers and historians invariably describe Whewell as a polymath: his work on the philosophy of science will be of particular concern to us here, but he also wrote on an array of subjects from astronomy to law, from architecture to mineralogy. The quotation comes from Whewell’s Bridgewater Treatise. The Bridgewater Treatises were a series of works commissioned from different authors by the Earl of Bridgewater to demonstrate the existence of God, as evidenced by the natural world. Whewell’s comments show that, at least as far as the physical sciences went, he did not believe that God intervenes directly to influence individual events; rather, he thought that God had set up natural laws (such as Newton’s laws of motion), and those laws dictated the patterns of the Universe. As we saw in chapter one, Darwin’s general view of the natural world at the time he wrote the Origin was of the same type. He did not believe that an intelligent God was directly responsible for creating individual species, nor for fitting species to their environments. Darwin thought that natural laws alone (primarily the law of natural selection) were responsible for these phenomena. But Darwin did not intend natural selection to rule out a God ultimately responsible for natural laws themselves.

Darwin’s epigraphs raise two interrelated themes, which we will address in sequence in this chapter. First, we will look at what Darwin understood by scientific method. Specifically, we will look at what Darwin and some of his contemporaries thought it took for a theory to be well supported by evidence. We will then look at today’s debate between the theory of intelligent design and the theory of evolution by natural selection. Armed with a good understanding of what it takes for a theory to have a good base of evidence, what should we make of the claim that both natural selection and intelligent design have claims to be taught alongside each other in biology classrooms? Is there really good evidence in favour of the intelligent design hypothesis?

2. INFERENCE TO THE BEST EXPLANATION

Darwin gives a strong hint in the Origin of his view about what makes a theory a good one. In the final chapter he summarises the varied facts his theory is able to explain – facts about anatomy, embryology, the distribution of species around the globe, even about the characteristic arguments had by natural historians – and he notes how ill-equipped are rival theories for explaining the same facts. In the Origin’s sixth edition he adds that:

It can hardly be supposed that a false theory would explain, in so satisfactory a manner as does the theory of natural selection, the several large classes of facts above specified. It has recently been objected that this is an unsafe method of arguing; but it is a method used in judging of the common events of life, and has often been used by the greatest natural philosophers.

(Darwin 1959: 748)

In other words, the fact that a theory is able to successfully explain diverse phenomena is, Darwin thinks, strongly indicative of the theory’s truth. This mode of reasoning has become known today as Inference to the Best Explanation, often abbreviated to IBE (Lipton 2004). Darwin is right that this is a highly intuitive conception of how theories of all kinds are supported by data. Why do we think it is likely that the butler killed the Earl of Wensleydale? Because if the butler did kill him, then this would best explain our data – the Earl’s blood on the butler’s jacket, the Earl’s blood on the knife found under the butler’s bed, the sighting of the butler running away from the crime scene just after the Earl died. When a hypothesis explains the data better than its rivals, we often think the hypothesis is true.

The Origin is full of arguments which seek to show how much better our understanding is of diverse phenomena if we assume common ancestry rather than special creation. Consider this example, where Darwin explains the different distributions of species in the Galapagos Islands (in the Pacific, nearest continent South America) and the Cape de Verde Archipelago (in the Atlantic, nearest continent Africa):

The naturalist, looking at the inhabitants of these volcanic islands in the Pacific, distant several hundred miles from the continent, yet feels that he is standing on American land. Why should this be so? why should the species that are supposed to have been created in the Galapagos Archipelago, and nowhere else, bear so plain a stamp of affinity to those created in America? There is nothing in the conditions of life, in the geological nature of the islands, in their height or climate, or in the proportions in which the several classes are associated together, which resembles closely the conditions of the South American coast: in fact there is considerable dissimilarity in all these respects. On the other hand, there is a considerable degree of resemblance in the volcanic nature of the soil, in climate, height, and size of the islands, between the Galapagos and Cape de Verde Archipelagos: but what an entire and absolute difference in their inhabitants! The inhabitants of the Cape de Verde Islands are related to those of Africa, like those of the Galapagos to America. I believe this grand fact can receive no sort of explanation on the ordinary view of independent creation; whereas on the view here maintained, it is obvious that the Galapagos Islands would be likely to receive colonists, whether by occasional means of transport or by formerly continuous land, from America; and the Cape de Verde Islands from Africa; and that such colonists would be liable to modifications;—the principle of inheritance still betraying their original birth place.

(Origin: 385–86)

The hypothesis that an intelligent God produced each species individually to suit its surroundings leads us to expect that similar habitats will contain similar species, and different habitats will contain different species. This is not what we see. Compared with Darwin’s hypothesis of common ancestry, special creation is a poor explanation of our data, and using IBE it is far less likely to be true.

We are undoubtedly prone to think theories true when they have explanatory power. But some explanations have lots of attractive attributes – they tie disparate phenomena together, they show us that the events we have witnessed are ones we should have anticipated – yet they are, in spite of all this, false. Conspiracy theories are often like this. So although ‘Inference to the Best Explanation’ is an attractive slogan, ideally one would want to flesh it out with an account of what makes an explanation good, and an account of why, when a hypothesis ties up some body of facts in a neat, explanatory way, it is therefore likely to be true. One cannot solve the first problem by insisting that only true explanations are good ones. IBE’s defender needs to pinpoint characteristics of good explanations, and then show that explanations with these characteristics are more likely to be true than bad ones. Requiring that an explanation does not count as good unless it is true misses the point of IBE.

As a first pass, we might understand a good explanation to be one that cites factors which would raise the probability of the event we are trying to explain. The more the probability is raised, the more satisfying the explanation (Mellor 1976). The most satisfying explanations are the ones that show that what happened had to happen, with 100% probability. On this view, we do not define a good explanation as one that is true, we define it as one that, were it true, would make the facts we seek to understand probable. We can explain a fire by reference to a short-circuit because in the circumstances (a ready supply of oxygen, a warehouse without sprinklers), a short circuit would make a fire highly probable.

When a hypothesis makes some set of observations very probable, philosophers and statisticians say that the hypothesis has a high likelihood. I will use italics throughout this chapter to refer to this technical notion of likelihood. It is important to remember that the likelihood of a hypothesis is a function of how probable the hypothesis makes some set of observations, not of how probable the observations make the hypothesis. So, for example, given the observation that there is clear liquid dripping down my kitchen walls, the hypothesis that there is a flooded bath upstairs has high likelihood. This hypothesis makes my observation likely. The proposal we have been considering is that good explanations of observations are hypotheses with high likelihoods in the light of those observations. Are explanations that are good in this sense also generally true? They are not, as the philosopher Elliott Sober often reminds us (e.g. Sober 1993). For any given set of data, there are typically numerous alternative hypotheses that make those data probable. They all have high likelihoods, but we must choose which is true. The drips on my walls are made likely by the flooding bath, but could also have been made by spilled vodka, errant guttering, a burst pipe and so forth. Some hypotheses with high likelihoods are patently absurd. If I hear a drumming noise coming from my ceiling, then I can explain this either by hypothesising that it is raining, or by hypothesising that tiny fairies are having a disco on the roof. Both hypotheses make my observations probable – if there were a fairy disco on the roof, then I would hear a drumming noise – but I do not regard the fairy hypothesis as probably true in virtue of this, even though that hypothesis has a high likelihood, and is, by the criterion we are currently discussing, a good explanation.

This teaches us an important lesson. We should not accept hypotheses merely on the grounds that they make some body of data highly probable. This does not mean that the likelihood of a hypothesis is irrelevant to its truth, but it does show that likelihood is not sufficient to justify belief in a hypothesis.

3. HERSCHEL AND WHEWELL

The methodologists of science who influenced Darwin most heavily – especially John Herschel – insisted that scientific theories should appeal only to what they called verae causae, or ‘true causes’. Darwin read Herschel’s major methodological work – A Preliminary Discourse on the Study of Natural Philosophy – during his student days:

During my last year at Cambridge I read with care and profound interest Humboldt’s Personal Narrative. This work and Sir J. Herschel’s Introduction to the Study of Natural Philosophy stirred up in me a burning zeal to add even the most humble contribution to the noble structure of natural science. No one or a dozen other books influenced me nearly so much as these two.

(Autobiography: 36)

We can understand adherence to the vera causa ideal as a recognition of the point we made in the last section. High likelihood does not by itself constitute serious evidence in favour of a theory. A theory must do more than explain some narrow class of phenomena if it is to command our confidence. There is no evidence in favour of the fairy hypothesis beyond the fact that it fits with our observation of noise from the roof. Rain, on the other hand, is something we have plenty of additional evidence for. ‘True causes’ are causes like the rain, rather than dancing fairies.

What, precisely, does it take for us to have ‘additional evidence’, of a kind that might begin to warrant our believing an explanatory theory? Equivalently, we can ask what it takes for a theory to meet the vera causa ideal (for details see Ruse 1975; Hodge 1977). Herschel gives various different answers to this question in his Preliminary Discourse on the Study of Natural Philosophy. At some points he adopts a proposal consonant with a more plausible version of IBE: verae causae are those which, in addition to explaining the phenomena that initially lead to their invocation, are also able to explain many other phenomena. They are causes ‘competent, under different modifications, to the production of a great multitude of effects, besides those which originally led to a knowledge of them’ (Herschel 1996: 144). Maybe we should believe in fairy discos if such discos also explained phenomena beyond the drumming noise on the roof (if, for example, they explained the appearance of tiny ethereal beer cans, left discarded on the mornings after drumming is heard). At this point, Herschel says, the scientist can be fairly certain that he has found ‘causes recognized as having a real existence in nature, and not being mere hypotheses or figments of the mind’ (ibid.). Best of all is when, once we have posited some cause, we are then able to form further explanations we had not expected, as well as explanations of phenomena that, at first glance, seem opposed to our theory:

The surest and best characteristic of a well-founded and extensive induction, however, is when verifications of it spring up, as it were, spontaneously, into notice, from quarters where they might be least expected, or even among instances of that very kind which were at first considered hostile to them. Evidence of this kind is irresistible, and compels assent with a weight which scarcely any other possesses.

(Ibid.: 170)

In these respects, Herschel is close in his methodological stance to Whewell, who claims in his Philosophy of the Inductive Sciences that:

. . . the evidence in favour of our induction is of a much higher and more forcible character when it enables us to explain and determine cases of a kind different from those which were contemplated in the formation of our hypothesis. The instances in which this has occurred, indeed, impress us with a conviction that the truth of our hypothesis is certain.

(Whewell 1996: 230, emphasis in original)

Both men recommend confidence in hypotheses that explain many diverse phenomena. When a hypothesis does this we have what Whewell calls the Consilience of Inductions. One difference between the two men, and it is a mild one, is that while Herschel puts the stress on explanations of phenomena in domains initially thought hostile to the theory, Whewell stresses explanations of kinds of phenomena which the theory was not designed to accommodate (Laudan 1981).

At times Herschel makes an even stronger demand on a theory, namely that we should be able to directly perceive eith...

Table of contents

• COVER PAGE
• TITLE PAGE
• COPYRIGHT PAGE
• ACKNOWLEDGEMENTS
• CHRONOLOGY
• A NOTE ON TEXTS
• INTRODUCTION ‘A PHILOSOPHICAL NATURALIST’
• ONE: LIFE
• TWO: SELECTION
• THREE: SPECIES
• FOUR: EVIDENCE
• FIVE: MIND
• SIX: ETHICS
• SEVEN: KNOWLEDGE
• EIGHT: POLITICS
• NINE: PHILOSOPHY
• GLOSSARY
• REFERENCES
• RELATED TITLES FROM ROUTLEDGE HUMAN NATURE AFTER DARWIN
• RELATED TITLES FROM ROUTLEDGE