Assessing dFSCI (4) Functional Specification

Now for the final part of the definition of dFSCI:


d) Which conveys the information for a well defined function


So, for example, to quote Gpuccio (with some minor editing) the function of haemoglobin is:

Attach an oxygen atom – which is the biochemical function of the protein itself, but adding some better description of the properties necessary for that function to be biochemically useful for the higher level necessities of the system: the oxygen atom must be bound with a reversible link, in different conditions of pH and so on.

The set of proteins that can perform this function then becomes the “target” as discussed in the previous post on complexity.


The functional specification has two contributions to dFSCI.  It is a way of defining a “target” for the probability calculation but some ID commentators regard the very fact that there is a function as a distinctive indication of design.  I will take each in turn.


Clearly for any biological phenomenon it is possible to describe its function in an infinite number of ways.  It is possible to define the function of haemoglobin in great detail as above or in much less detail. For example:


  • Conveys oxygen to the tissues
  • Acts as means of allowing tissues to get access oxygen
  • Contributes to the respiratory system

Right though to something like:

  • Improves fitness of the organism


These are all things that haemoglobin does.  As the description gets less specific the probability of a “random” protein performing that function increases but also gets harder to estimate.  There are far more configurations of proteins that can improve the fitness of an organism than there are configurations that can attach an oxygen atom with all the conditions above.    Which specification should we use? I can make any outcome of any sort as improbable as I like by defining it in sufficient detail.  The probability of a dice landing with six uppermost is about 1/6. The probability of this dice landing on my desk with six uppermost and with the dots aligned North South is much smaller.  There needs to be some justification for using a particular specification. Gpuccio says he always uses the most specific definition he can find because the calculation is more precise.  But this is hardly a justification.  It is just a convenience.  It happens to have the consequence that it makes the probability of a random protein meeting that specification very small.


Remember that evolution (according to the conventional theory) does not have a target.  It happens across useful innovations.  It may that a specific innovation is one in a billion possible useful innovations.  Evolution happens to stumble upon that one when a billion others that might have improved the fitness of the organism in unpredictable ways were neglected.  From the point of view of the organism with that innovation it may seem extraordinary that evolution should have unearthed that particular solution.  From the point of view of the evolutionary biologist – the chances of stumbling across something useful are much, much higher.  Someone has to win the lottery ticket – although the winner may think it is extraordinary that they have won it.


There appears to be no justification for using the most detailed functional specification.  What about the correlation between function and design?  As Gpuccio puts it:

functional specification and meaning are the real marks of design

As I understand it, the argument is that the very fact that parts of living things have functions at all is an indication that they were designed.  The probability calculation is just a way of eliminating cases where outcomes have the appearance of design but actually may be the result of necessity or chance.  Everything other than living things that has a function is designed by humans, so it is reasonable to conclude that biological outcomes with functions are likely to be designed.


To examine this argument it is necessary to examine why we suppose that things have a function.  What is it that makes us suppose that something is functional?  It does not just come down the physical characteristics of the object.  Imagine a complex heap of fallen trees.  When you step on one fallen trunk it causes a surprising and large movement in another very heavy tree trunk some distance away.  Is this a function?  We are unlikely to settle the dispute by further examination of the heap.  What would settle it is to find some purpose for the operation – for example to open up a pathway to a sleeping place. Function is logically tied to purpose.  Characteristics such as complex causal chains may arouse our suspicions but they are not definitive.  A complex causal chain leads to a hurricane – but we do not say that the function of warm air and converging winds is to create a hurricane. The same applies to man made objects.  It is quite common, for example, to come across a feature of a building that has no function – an extra buttress or such like i.e. it has not purpose.  It may not even be there for aesthetic reasons.  It may be an oversight or a by product of the building process.


Given that function is tied to purpose it is only going to be associated with the kinds of things that can have a purpose.  For those that do not believe in deities this means living things.  People and animals, perhaps even plants, can have purposes.  Rocks and the weather cannot.  Those who are religious will ascribe purpose to deities.  So for some people person almost everything has a function to the extent that fulfils their God’s purpose.  Some believe that the function of the universe is to provide an environment to sustain life.  However, we cannot use function as a sign of purpose, because it is only by ascribing purpose that we know we have a function.  For the atheist, the universe, however complex, does not have a function.  It would be like declaring that being a bachelor is evidence for being unmarried.  There are things that are often associated with being a function that could be used as evidence – such as long causal chains – just as there things that are associated with being unmarried such as long nights out and eating alone – but they are not the same as being a function or being unmarried.


So what is happening in the case of biological outcomes?  How come we can ascribe a purpose to haemoglobin?  Whose purpose is this fulfilling?  Note that it is not the purpose of the organism in question.  Most organisms certainly want to breathe and survive but they do not create or control their level of haemoglobin.  The answer is that  purpose in the case of biological outcomes is an analogy.  It is very useful to consider living things as though they had been designed with the purpose of living and producing.  For a long time most people supposed there was a deity with this purpose.  But the whole point of modern biology is that this now considered to be simply a useful analogy.  There is nothing with this purpose and so strictly speaking the parts of an organism do not have functions. 


Biology is the most common case of explaining something as though it had a purpose, but it is not unique.  I have heard a teacher explain the development of the structure of a river delta as though the river water intended to get to the sea.  So it was therefore possible to talk about the function of a particular marsh or stream in conveying water to the stream.  But no one concluded that the river actually had this purpose or that the delta was designed.


So function can be consider a mark of design in the sense that it is a causal chain created with a purpose. This is true by definition.  But in this sense function is not found in biology.  Or you could change the definition of function to be something like “causal chain leading to well defined outcome” in which case it is found in a wide range of phenomena – not just living things and things we now to be designed.


1 Response to “Assessing dFSCI (4) Functional Specification”

  1. 1 Toronto October 7, 2010 at 12:37 pm

    “I can make any outcome of any sort as improbable as I like by defining it in sufficient detail. The probability of a dice landing with six uppermost is about 1/6. The probability of this dice landing on my desk with six uppermost and with the dots aligned North South is much smaller. ”

    I think this statement is a good description of ID’s improbability argument. By adding detail, any outcome can be highly improbable.

    They also go the other direction though, by not including enough detail.

    As an example, the probability of any male marrying one particular female is about 1 in 3 billion considering the Earth’s population at 6 billion. This holds for any particular woman also.

    That means the odds of any particular couple getting together are 1 in 9,000,000,000,000,000,000.

    At this simple level of detail, we reach a result that suggests that people are more likely to win the top prize in a lottery than actually being married to their current spouses.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: