Evolutionary epistemology (0) – an introduction

by Neil Rickert

In a previous post I suggested that traditional epistemology, as studied within philosophy, has developed in a framework based on a creationist or intelligent design way of thinking.  Today, I want to look at what might be a way to look at knowledge from an evolutionary perspective.

It’s important to be clear, here, that I am not concerned with the question of whether knowledge evolves.  Everyone agrees that there has been a growth in knowledge, and that some old ideas (geocentrism, for example) have been discarded.  So there isn’t any important disagreement over whether our knowledge evolves.

The difference being discussed is between the ideas about knowledge that come from a creationist way of looking at things, and those that come from an evolutionist way of looking at things.

The word knowledge itself is one of those confusing words where we are not quite sure what it means.  But we need not allow that to be too great a concern.  In practice, epistemology is concerned with the question of being able to make true statements about the world.  That is, epistemology has to do with having facts, and with what is involved in having facts.

According to the creationist perspective, God created the world based on his own principles.  And those principles are the basis for having facts.  So the creationist perspective is that there are lots of facts just laying around and waiting to be picked up.  The cognitive agent picks up these facts, and then tries to induce the creators principles, based on patterns that can be discerned from within the facts that are picked up.

From an evolutionist perspective, there is no reason that there should be any facts at all.  The world exists in its current form, due to the way that elementary parts of matter aggregated, and due to the more stable aggregations being persistent.  Thus there are no design principles, there are no laws of nature, and there are no facts that are just laying around.  Under the evolutionist’s perspective, if a cognitive agent wants to have facts about the world, then he must first invent ways of having facts (of being able to represent the world).

To illustrate this, consider a statement which we might think of as if it were a putative fact:

At 1 pm, next Wednesday, I shall take a bus trip from Illinois to Wisconsin.

We are able to express this putative fact, only on account of human conventions.  Firstly, there are the naming conventions.  We know that the words used are conventional, because other people might use French or German or Chinese words instead of the English words used here.  But the naming conventions are only part of the story.  Both Illinois and Wisconsin represent parts of the north American continent that result from political conventions in how to divide that continent into individual states.  The use of Wednesday is an implicit reference to our conventional grouping of the days into weeks, where a week is a group of seven consecutive days.  At this time of the year, 1 pm refers to a time expressed in daylight savings time, and that itself is a conventional standard for time.  Even the word bus references our conventional partitioning of kinds of transportation devices into cars, vans, truck, buses, etc.

From a creationist perspective, the problem of knowledge is one of picking up facts that just happen to be laying around, and then searching for patterns within those facts.  From an evolutionist perspective, the problem of knowledge is one of establishing useful conventions such as would make it possible to have facts in the first place.  These are two very different ways of looking at the knowledge problem.


7 Responses to “Evolutionary epistemology (0) – an introduction”

  1. I agree with this to the extent that I think that religious thinking has influenced both philosophy and science, particularly from two perspectives.

    One has been the extent to which both philosophy and religion has come to see thought as a means of getting access to reality. The problem with this is it leaves the imagination to run free and unchecked by empirical experience, so that fairies, ghosts, gods can be invented and believed to exist, and for statements about them believed to be true, without any requirement to verify or challenge those ideas empirically.

    The other has been the imposition of authority. This mode of thought, that something can have a sense of absolute authority leaked into science right up until the 20th century, through out which it both peaked and began to be challenged. While good scientists may have dispelled the myth of authority to a large extent, science was still presented as absolutely authoritative, particularly through marketing, and from government when it suited.

    I don’t have a problem with induction. I can see that philosophers do, in that they tend to want logical proofs, and get a little tetchy when scientists propose that we don’t always need proofs, that a collection of experiences that start to form a pattern is often good enough. It turns out that this is all science ever does for us.

    Laws of nature are entirely constructs in human heads. They just happen to fit the messy nature we observe pretty well. That’s good enough. Until it isn’t. Then we have to look for a better law.

    Another view of laws of nature is as models. All our laws, and all our concepts, are just models. For that matter everything is pretty much a model of something else. So, when DNA is produced, it is modelled on RNA. An infant is a model of its parents, grandparents, and through DNA it’s distant ancestors; but can also be a model of its environment in the womb. Children going through an education systems become models of that system – for example, children in Creationist schools may model their ideas on Creationism rather than evolution.

    I agree that the philosophy of epistemology is flawed. I think that because again philosophers have a tendency to want to pin things down tighter than they can be. I suppose you could class it as being caused by ID thinking, but I feel it’s more to do with the ‘primacy of thought’ problem, which has been influenced by religion; or maybe it influenced religion. Identifying the source of the problem for epistemology is difficult because philosophy and religion evolved together over time. And, particularly in pre-scientific times, they were busy looking for pure and certain truths, which in the various contexts of the times became an inherent part of the theology that refined those ideas into monotheism and a desire to find proof of God’s existence.


  2. “Everyone agrees that there has been a growth in knowledge…”

    This is a point where I’d want to try to get to the bottom of our differences. I do see knowledge as a this ‘something’ that human society has collected and that it has grown, and some of it we have collectively discarded (though not necessarily completely – e.g. astrology).

    I don’t think any one human could hold all of what we might call human knowledge in his head – even if he had time to acquire it. In this context the word ‘knowledge’ has a meaning that should really be replaced by ‘data’. All collections of what we call knowledge outside our heads is really just data. A book by Shakespeare might mean nothing at all to an alien. It will mean very little to a Chinese person who knows no English, though he would no doubt recognise the object as a book. Pretty much as Chinese characters mean nothing to me.

    So it seems that the best use of the word ‘knowledge’ then is what is contained in a human head. Or at least we should distinguish these uses of ‘knowledge’.

    I agree that epistemological definitions of knowledge are flawed.

    One point I would make is that the requirement that knowledge be ‘true’ (leaving its definition to one side) seems unnecessary. It seems easier to say that ‘knowledge’ that humans have in their heads is just information, data. Whether it corresponds to something else, something about a physical object, or something about a language proposition about a physical object, is sufficiently covered by measuring the degree of correspondence in some way. We can say then that this data, this information, this knowledge, is true or false, which we sometimes do, or that it is partly true where it’s a complex piece of data, information, knowledge. The insistence of linking of knowledge inherently to truth seems to cause problems. For example, “I know X is true”; “You are wrong, X is not true, so you don’t know it” seems unnecessarily redundant; rather “You are wrong, X is not true”, which seems sufficient. This seems to match better the feeling that someone does know something, but they are wrong about it.

    The other problem word is ‘belief’, which only adds to the confusion because of its association with religious belief. It’s no good philosophers appropriating words with a common meaning and then expecting non-philosophers to keep up with that. Have you ever had a discussion with a religious believer with little philosophy and tried to explain JTB? It’s hopeless.

    “From an evolutionist perspective, the problem of knowledge is one of establishing useful conventions such as would make it possible to have facts in the first place.”

    I agree with this completely. But I’d link it to the idea of data again.

    There are patterns of matter and energy in the universe, sometimes called ‘fractures in the continuum’, or ‘lack of conformity’. In informational terms there are distinctions – distinct data patterns. These are synonymous to all intents and purposes, though some philosophers object to this – but then I think if they object to this they’ve got bigger problems with solipsism anyway. Certainly from an inductive point of view this acknowledgement of the correspondence between reality and the patterns or distinctions in it is sufficient.

    On this basis, everything is essentially data – including human brains. The change in human brains that occurs when thoughts flit through them or when the remember something is merely brain matter changing state, changing pattern. Conversely, everything is also material – including data, by virtue of the fact that it consists of the organisation of matter, whether that’s configuration of electrons in the capacitive element of a logic transistor or the configuration of synapses in a human brain.

    Even when we think in our minds of abstract data existing in some Platonic plain, that very idea itself has an existence in the formation of matter in the brain. The odd thing to grasp with this is that we have this abstract notion that there is nothing abstract, it’s all real, except the abstraction itself, which isn’t.

    The term (maybe from Luciano Floridi?) is ‘representation = physical implementation’ (no teleology implied by the use of ‘implement’).

    I think it important to note that all ideas, such as ‘idea’, ‘concept’, ‘abstract’, along with religious ideas like ‘soul’, ‘God’, are all inventions of the human mind – as is ‘mind’ or course, so I should really say, of the human brain. No science has ever discovered the existence of a material object, or any trace of energy, that is a ‘soul’, or an ‘idea’, or a ‘concept’.

    So that when philosophers talk about these as if they have some existence it’s pure invention with no verification through evidence.

    What we do find are patterns in matter which are used to represent these, which then invokes something in the brain. So, the word ‘concept’ itself invokes the concept of ‘concept’ in my brain as I read it. But given that this is happening in a material brain then there is little more to expect other than the word on the screen has triggered a recognition of a corresponding pattern in the brain.

    This is why I think that even when talking about human ‘knowledge’ in the brain we are better sticking to terms like data, or information. This view also unifies the idea of knowledge as data within human brains, and outside them, on paper, in books and databases, and even unifies the idea with the material world.

    I accept that as a matter of convenience we will want to differentiate between the places where this data/matter resides. So, on some occasions we’ll talk about ‘the body of human knowledge’ when we mean the accumulation of all of what has at sometimes been in some human brains and has been translated into common media, such as books. On other occasions we’ll talk of how a person ‘knows some proposition to be true’, when we are talking about their commitment to the correspondence of the proposition to some relating thing or event in the world outside the human head. But when looking at this in the whole, and at the same time looking for how all this ‘knowledge’ exists in some detailed but unified way, it’s easier to talk about data, information, matter.


    • This is a point where I’d want to try to get to the bottom of our differences.

      When people say “X has a lot of knowledge”, I don’t think they are referring to the statements that X makes. I think they are referring to his abilities, from which those statements arose. So I see knowledge as mainly abilities.

      For example, you say (on your blog) that you develop software. The programs that you emit are not your knowledge. Rather, your knowledge is an underlying set of abilities that enables you to create these programs. The quality and suitability of your programs for the tasks they address is a better measure of your knowledge than is the quantity of programs that you emit.

      That was, more or less, my starting point. I have since further refined that, but the details are not important at this point.


  3. Yes, I agree with that. So, my question from there is, how is that knowledge actually instantiated in the brain.

    Taking into account principles of conservation and the laws of thermodynamics, and while still waiting for someone like Chalmers to expand or provide evidence for some additional fundamental something-or-other, information, data, needs a physical representation. Knowledge is something we have within us, that I agree is not (or need not be) a one-to-one correspondence with our statements or expressions of what we know. But, it still needs some substrate upon which it is written. There is no evidence of any other substrate than brain material.

    Even if my knowledge is, in your terms, an underlying set of abilities, those abilities reside somewhere in some format, encoded somehow, otherwise they wouldn’t be my abilities to have. When taxi drivers learn ‘The Knowledge’, their brains are seen to physically change to the extent that it is detectable.


    • So, my question from there is, how is that knowledge actually instantiated in the brain.

      I am planning some posts on that. As a brief summary, I see knowledge primarily as the ability to get information from the world and to represent that information. Since there’s measurement involved, it seems that I’ll have to start with a post on measurement.

      It turns out, with some admittedly sophisticated mathematics, that if you know enough about how to get and represent information, then you also implicitly know a lot about the world. The brain doesn’t have to know any of that mathematics. It can just work on getting and representing information. Then, in some sense, we experience the information.

      Looked at this way, knowledge (or what I consider knowledge), and information (roughly the justified true beliefs of philosophers) are complementary to one another. There’s a mathematical duality between the two, and peoples intuition about dualism is probably related to that mathematical duality. Note that the mathematical duality itself is not problematic and does not imply spiritual souls or anything of the kind (except as metaphor).

      I hope to have some posts on this. But it is hard to present in a way that does not confuse people, so I’ll need to do some thinking about presentation.

      Even if my knowledge is, in your terms, an underlying set of abilities, those abilities reside somewhere in some format, encoded somehow, otherwise they wouldn’t be my abilities to have.

      I would prefer to say “structured” rather than “encoded”. The term “encoded” seems to imply an encoder.

      If the neural system just goes about its business, developing ways to measure the immediate environment, then it will structure itself the “right” way. Exactly how to structure itself is vastly underdetermined by the situation. But that underdetermination doesn’t much matter. No matter which possible way it is done, the same world will emerge from information collected by that structured neural system.



%d bloggers like this: