Darwin’s Dangerous Idea: Chapter 2

In chapter 2, Dennett addresses what he sees as the fundamental "dangerous idea" that Darwin had: speciation through natural selection. He has a very useful section at the beginning on the difference between accidental and essential properties and between real essences and nominal essences. Pre-Darwin, the world was "carved up at the joints" and biological entities were divided up based on what was seen as real essences: essential properties that defined a thing such that absent those properties, the thing ceased to be what it is. Darwin overthrew this type of essentialism. Terms like "natural kinds"-- where this term described silos of living things which are not interrelated with each other--no longer described the biological world.

Dennett describes natural selection as an algorithmic process by which change occurs in living organisms by a regular, repeated, chance-driven process that always works. Natural selection is essentially a set of algorithms that sort, winnow, and build things (52) but which have no goal or purpose behind them. Dennett makes some effort to present the idea that what seems like complex results can be achieved through very simple steps using algorithms. Natural selection, then, provided Darwin with his mechanism for the diversity which his research produced.

Dennett also continues to promote the idea that Darwin's idea is beyond dispute.

"But the fact of speciation itself is incontestable, as Darwin showed, building an irresistible case out of literally hundreds of carefully studied and closely argued instances." (44)

"To put it bluntly but fairly, anyone today who doubts that the variety of life on this planet was produced by a process of evolution is simply ignorant--inexcusably ignorant, in a world where three out of four people have learned to read and write." (46)

"It is reasonable to believe that an idea that was ultimately false would surely have succumbed by now to an unremitting campaign of attacks." (47)

Dennett appears to be convinced that Darwin made the case for speciation through evolution and for natural selection. The single book The Origin of Species seems to have been sufficient to establish the fact of evolution by natural selection even if it was greatly insufficient to establish the mechanism by which this natural selection occurs.

Interestingly, Dennett says that Darwin didn't really explain the origin of species. In fact, Dennett seems to say, no one has yet explained it. "Controversy about the mechanisms and principles of speciation still persists, so in one sense neither Darwin nor any subsequent Darwinian has explained the origin of species." (44).

Algorithms and Natural Selection

This section is important as a foundation for natural selection and Dennett does his readers a great service by laying it out so plainly. This section doesn't contain much if any scientific evidence for natural selection as it lays out a philosophical foundation for how natural selection might work. The model of algorithms that Dennett proposes has three main features.

1. Substrate neutrality. The algorithm is logical in nature, not tied to any particular material per se.

2. Underlying mindlessness. Dennett is unclear here. The feature is described as "mindlessness" but the description is about the process of an algorithm using small, incremental, and fundamentally simple steps to produce very complex results. Of course, this says nothing about mindlessness. That something is simple doesn't say anything about it not requiring specific and perhaps irreducible features of mind namely intention, goal-directed behavior, persistence, and the like. But I think Dennett's point is that an algorithm can build on itself to produce complex results from very modest and simple processes. Perhaps a better title for this feature would be "underlying simplicity."

3. Guaranteed results. An algorithm always produces the same result every time it is executed.

Dennett then formulates Darwin's fundamental idea: life on earth has been generated over billions of years in a single branching tree--the Tree of Life--by one algorithmic process or another.

Dennett's discussion of how algorithms might work in natural selection is important in that he's trying to present a foundation for how the process might have worked. Notably, he doesn't bring in any physical evidence and argues that algorithmic processes are demanded by the physical evidence. He summarizes what he believes Darwin proved and concludes that algorithms are the best underlying explanatory model for what Darwin discovered.

Dennett is very clear that the algorithms of natural selection are mindless, goalless, chance processes. This, of course, is important given a naturalist presupposition in metaphysics. Dennett seems to hold that naturalism and Darwinian evolution are mutually dependent.

"Darwin was offering a skeptical world what we might call a get-rich-slow scheme, a scheme for creating Design out of Chaos without the aid of Mind." (50)

"But a procedure doesn't fail to be an algorithm just because it is of no conceivable use or value to anyone….Algorithms don't have to have points or purposes." (56)

"We can now expose perhaps the most common misunderstanding of Darwinism: the idea that Darwin showed that evolution by natural selection is a procedure for producing Us." (56, emphasis and capitalization in original)

"Can it [the fancy biosphere] really be the outcome of nothing but a cascade of algorithmic processes feeding on chance? And if so, who designed that cascade? Nobody. It is itself the product of blind, algorithmic process." (59)

For Dennett, Darwinian natural selection not only describes an algorithmic process for how evolution takes place but also can tell us that this process is not guided by intelligence of any sort. However, Dennett does little to demonstrate this. In fact, his description of the algorithms of natural selection and how they might work in evolution are not based in physical evidence neither do they seem to have even prima facie entailments for what is "behind" them. Yet, Dennett seems to think that an algorithmic description of natural selection entails naturalism and, in fact, this was a key driver of Darwin's idea that makes it dangerous. If Dennett isn't implying anything as strong as entailment, he certainly sees a strong relation between Darwin's idea and purposeless processes. I'm not quite sure why. Certainly he hasn't yet provided any evidence to think that this is the case--in this chapter it's been mere assertion. In fact, based on what Dennett says in this chapter, algorithms make little sense without Mind.

Natural Selection, Axiology, and Teleology

In his discussion of algorithms, Dennett uses language, metaphors, and examples to illustrate the idea. These are laden with teleology. He's already established that Darwin's dangerous idea implies that purposiveless, mindless, chance forces operating in an algorithmic way developed in small, incremental steps the diverse forms of life we see on earth today. To discuss algorithms consistent with this view then demands language that avoids concepts involving goals, purpose, and value. While many Darwinians use such language as “semantic devices” designed to communicate clearly, a realist description of natural selection would avoid such metaphor and this chapter does not provide such a description.

"But the algorithms that will concern us have nothing particular to do with the number system or other mathematical objects; they are algorithms for sorting, winnowing, and building things." (52)

"But what, exactly, does this algorithm do? It takes as input a set of competitors and guarantees to terminate by identifying a single winner. But what is a winner? It all depends on the competition." (53).

"The power of an algorithm to yield something of interest or value is not at all limited to what the algorithm can be mathematically proven to yield in a foolproof way, and this is especially true of evolutionary algorithms." (57)

"It can be proven mathematically that these rearrangements will tend to get better and better, approaching the optimum or strongest total structure…" (Dennett here is using the example of metal annealing to demonstrate the use of an algorithm. 57,58).

"Here, then, is Darwin's dangerous idea: the algorithmic level is the level that best accounts for the speed of the antelope, the wing of the eagle, the shape of the orchid, the diversity of species, and the other occasions for wonder in the world of nature." (59)

Of course, if naturalism is true, then there are no winners or losers in evolutionary process, there are no changes that are more valuable than others, there are no algorithms that are for any purpose, there's no sense to be given to notions of "optimal" or "strongest" or any such thing. I suspect that Dennett employs such language (as Dawkins does) simply as a rhetorical device to get some complex ideas across. However, if he doesn't move away from such language, I think he can easily be accused of being inconsistent with his naturalism.

In this chapter, Dennett sets up a very useful foundation for the mechanism of natural selection. He does not provide much by way of evidence that his idea is true and no evidence that the diversity of life was brought about in the algorithmic way. But his is a philosophical discourse, not a scientific one.