William Paley’s modern counterpart is Michael Behe. Behe has been a prominent exponent of ID and is credited with coining the phrase ‘irreducibly complex’ in his book, Darwin’s Black Box(1996):

‘By irreducibly complex I mean a single system composed of several well-matched, interacting parts that contribute to the basic function, wherein the removal of any one of the parts causes the system to effectively cease functioning.’

Behe argues that some biological systems are too complex to have been formed by increments over evolutionary time. In other words, if you take away one part, the rest falls apart completely, like a house of cards. Such complexity could only have come into existence by the hand of a designer, and an intelligent one at that; there’s no room for Darwin here… or is there?

The bacterial flagellum is an example of a supposedly irreducibly complex trait. So what is a flagellum and what does it do? Bacteria need both to find nutrients and to avoid noxious substances and in order to do this they need to be motile. One way for bacteria to move around is effectively to stick a propeller to one end of themselves. This propeller is made of one or more long filaments which are shaped a bit like corkscrews. This then turns and propels the bacterium forward or, by reversing direction, allows the cell to turn around.

This ingenious apparatus is the bacterial flagellum and the modern argument for the presence of a designer often hinges on the assertion that this tiny molecular motor is irreducibly complex. That is to say that if even one component was absent, it would no longer function and so could not have developed incrementally over millions of years. This would provide a strong argument against natural selection if it wasn’t flawed in several ways.

Firstly some proteins in bacterial flagella have been shown to be helpful but non-essential to the flagella’s function. Several of these structures have been removed through genetic manipulation and the cells, although they may not be quite as happy, can still swim showing, that the flagellum is functional. In addition, different bacteria have different flagella which require different and unique proteins. As such this irreducibly complex structure shows a large amount of variability and requires a range of different proteins in different organisms arguing against a single irreducibly complex structure.

Secondly the proteins making up the flagellum are very often similar to each other and to proteins used for other functions within the cell. This allows a model based on gene duplication to be built. In this model a section of DNA is duplicated resulting in a second copy of a gene within an organism, and this second copy is not maintained by selection since the first copy is still fulfilling the original role. Subsequently, therefore, this gene may diverge and can be co-opted into the flagellum. The existence of these similar proteins provides an indication that the flagellum did arise from existing structures rather than being produced in fully functional form by an intelligent external creator.

The gradual formation by recruitment of proteins previously used for other structures might still seem incredible. Several groups of scientists have now, however, developed models of flagella evolution relying on the formation of intermediate structures which had other functions. These often start with a passive pore structure allowing transport of molecules across the bacterial membrane. Then this pore developed into a form of secretory apparatus, similar to those still observed in some bacteria. Finally this ‘proto-filament’ could be coupled to ion pumps in the membrane allowing rotation; this would be a fairly crude flagellum but would be expected to provide some selective advantage by the small degree of movement it would allow. This small amount of movement would, for example, allow a bacterium to move from areas where it has excreted waste products into areas where nutrients haven’t been depleted. This simple system could then be elaborated producing the flagellum seen today.

Finally scaffolding is another possible explanation for how a flagellum could appear to be irreducibly complex. What does that mean? Well there may have been a more unwieldy and inefficient flagellum which developed over time bit by bit. This in itself was not irreducibly complex and it could easily be seen how it built up. This could then be stripped down over time, as proteins became adapted to be more efficient whilst others were lost, making a sleeker and more efficient motor. This new motor may not look like anything could be removed from it if it is to remain a working motor, but then who knows what was there holding it together before?

Before the flagellum became the hobby horse of ID, many anti-evolutionists turned to the eye. Our eyes and those of other animals are incredibly complex. But we now have a good understanding of how these structures evolved.

We are bathed in a sea of light constantly, and any knowledge that animals might get from perceiving this light will give it advantages. Light can be used to regulate body clocks, to detect predators, and to search for food and mates. Many simple organisms have light sensitive pigments (photoreceptors). These are very simple, consisting just of proteins that respond to light and kick off a series of signals which help the organism make use of this information. From this, the next step in eye evolution is a simple one, to back the light sensitive cells with an opaque layer. Since light can now reach the photoreceptor from one direction only the animal can infer what direction the light is coming from, very useful for orientation, especially in water (where eyes first evolved) since it allows organisms to move up and down the water column depending on food availability and predation risks.

One way to make inferences of direction more reliable is if the light sensitive area is formed into a basic cup shape. In this situation light directly ahead stimulates all of the photosensitive cells whilst light off to one side stimulates only some of the cells, and from this it is easy to tell the direction of the light. Does this remind you of anything? The retina at the back of our eyes is just such a cup shape.

We can see that creating a primitive eye capable of detecting both direction and intensity of light is relatively simple. Have we hit an impasse? Whilst we have the ability to detect direction we're missing any ability to form an image. The solution again is simple, and as required by evolution can come in a series of gradual improvements. Now all we need to happen is for the opaque layer that backs the photocells to extend to form, in effect, a pinhole camera. This works because if you have a tiny hole you can create a sharper image of the scene, and the smaller the hole the sharper the image. The short steps are clear. Evolution keeps making the hole smaller until it reaches an optimum. As before there are animals that use just this type of eye evident in nature. The Nautilus (kind of like a snail of the sea) in fact has a remarkable eye that works on just this pinhole principle.

There is however a problem. There are physical limits to the effectiveness of a pinhole. The hole can become so small that the effect of light behaving as a wave is to blur the image and make it very dim. The size of the hole therefore must strike a balance between a clear image and a bright one (at least down to a minimum absolute size). Thankfully for us, nature has found a way around this: the lens. As you may recall from early school science classes light travelling from one transparent substance to the next can change speed and therefore bend - it is this effect that makes your straw look bent in your drink or makes white light split to form a rainbow. This property also permits a lens to focus light onto a single point by bending all the incoming rays through suitable angles. This means you can have a large opening but still have a sharp image.

How then did this evolve? Recall that the primitive eye had an opaque backing. What was not made explicit however was that to protect the photocells there was usually also a transparent front layer (for protection). As the cup shape evolved it was often filled with this transparent layer (as our eyes are filled with a transparent gel). Now what had to happen was for part of this layer (the lens) to change so that light slowed differently. This is not a difficult property to manipulate as all it requires is a change of something called the refractive index of the material (a measure of the speed of light in the substance compared to the speed of light in space). Again, we have an easy climb up the evolutionary ladder as each rung need simply work as a slightly better lens - even a small improvement would provide a selection advantage

We have now reached a very fair approximation of the human eye. Interestingly, the final advantage of our eye is that we can change the exact shape of the lens so as to focus light coming either from distant objects or close objects (some animals can only focus light from the middle distance as something of a compromise). We managed to evolve this trick because muscle groups near our eyes affect the focus. Over time these muscles were adapted into the eye's system and became part of how we see. Other animals (such as snakes and fish) do not in fact focus light this way, but instead move the lens backwards and forward in their eye. In this case it seems the muscles that were commandeered by their respective eye systems were different and that those differences have left one set more readily able to adjust the shape of the lens, the other more readily able to move it.

A team of Swedish scientists have demonstrated the ease with which natural selection can produce eyes. They started with a computer model of an 'eye' that was simply a straight strip of light sensitive cells, backed with an opaque screen and protected by a transparent layer (a very primitive system of the sort that featured right at the start of our journey to follow our eye to its modern form). They then programmed the computer to randomly make very slight adjustments to various properties of the eye – for example a 1% change to parts of the transparent layer’s refractive index, or a tiny change in the shape. Their computer then picked any 'eye' that was better at being an eye (using the kind of criteria we discussed such as light intensity at the photocells and how well focused the image produced was). The ones that were better were kept and allowed to randomly change again. Over thousands of generations the strip changed into something very much recognisable as a human-like eye - lens, retina and all. Thousands of generations may seem like a long time but actually, even using a very pessimistic estimation of the rate of change, this was a remarkably swift evolution.

All of the above leaves us where we are today, with a beautifully intricate (indeed irreducibly complex) system, one that would not function without all of its 'parts', yet also one that can be seen to have evolved. The evidence charting this path is also very good, with examples of all the different stages leading up to us still evident in nature today.

We’ve discussed two examples often put forward by proponents of ID as evidence of irreducible complexity, showing how we can provide evidence that these structures evolved by the gradual accumulation of small changes just as Darwin supposed.

Darwin never claimed to have all the answers to spell out the evolution of every trait. He showed us the powerful mechanism by which evolution works, natural selection, but he left it to the generations of scientists following in his footsteps to provide conclusive evidence. Scientists have spent 150 years doing so, not by bias, but because they tested Darwin’s ideas and found that they are right. We still however do not know the evolutionary history of every part of every living thing. There are gaps in our knowledge that scientists work to fill and the future generations of scientist will continue to work on.

The problem with ID is that it suggests that we should just give up when we find a problem that seems tricky to solve! If that was how science worked we wouldn’t be where we are today. We probably wouldn’t have even invented the wheel, or learnt to control fire!

The most fun part of science is finding difficult problems and working hard to solve them; gathering facts and observations, developing hypotheses and testing them – that is what science is about. It’s fun and it works! Why give up on it now?!

After 150 years of research we are starting to appreciate the intricacies of evolution and the power of Darwin’s theories to explain even the most complex traits. When we can’t yet do so, eventually we will, but it’s worth remembering Orgel’s second rule: ‘Evolution is cleverer than you are’.

Written by Ed Roberts & Andy Maddox

Darwin's Black Box
by Michael Behe, The Free Press: 1996

An Intelligent Person's Guide to Genetics
by Adrian Woolfson, Duckworth Overlook: 2004

The Blind Watchmaker
by Richard Dawkins, Penguin: 1986

Climbing Mount Improbable
by Richard Dawkins, Penguin: 1996

Unweaving the Rainbow
by Richard Dawkins, Penguin: 1998