Turing 12 - thoughts from AISB/IACAP World Congress

Monday and Tuesday this week I spent at the AISB/IACAP World Congress (Turing 2012), at Birmingham University. (The event is continued all week, but I just went for the two days.)  It was very worthwhile - many thanks to the organisers - and I've come away with lots to think about and, in an ideal world, blog about.

Here's just a few notes to flag some of the things I would write about (for a variety of different reasons) if I had time:

- Floridi's 'Level of Abstraction' (LoA). I agree with his emphasis of the utility of the idea, but I call it a map rather than an LoA.
- The requirement for information to be true to count as information (adding that to the General Definition of Information, GDI). Yes but there's more I want to say about that. (James Cussens very clear comments on the Barwise/Carnap paradox)
- The Big Picture about information - paradigm shift? Grand narrative? Philosophia prima?
- Various thoughts about ontology
- Defining intentionality, whether a machine can 'intend' something - see my blog post about intentionality in the genetic code, 'irreducible first person perspective'; synthetic semiotics
- distinction between ontological and epistemological emergentism;
- Kants claim that only the creator of the world can have direct access to it - the designer of a simulation can have access to the noumenal
- Suzanne Embury on "Forget Dimensions, Define your Information Quality Using View Patterns". About a specific task to do with information quality, but surely links to all sorts of ontologies.
- (Floridi) "since 2007, information created has exceeded storage capacity" (so you can't just store everything and worry about what to do with it later)
- Intuition and Ingenuity: Alan Turing Centenary Exhibition

The 'Author meets Critics' session, when five eminent academics gave comment (it could hardly be called 'criticism' in any sense, to be fair!) on Luciano Floridi's excellent new book: "The Philosophy of Information" (from the left: Luciano Floridi; James Cussens; Phyllis Illari; Jean-Gabriel Ganascia; Chryssa Sdrolia and Barry Cooper).

Choice & information

My GP (family doctor) referred me for physiotherapy but the surgery doesn't have a physio of its own, so the receptionist held a card in front of me with a list of about ten and told me to choose which I wanted. This is the zeitgeist: consumer/patient/customer choice.

I wasn't really in a position to make an informed choice, but I pointed: "I'll take that one".

My choice is communicated somewhere - it's not important where - so this can be nicely modelled by the Shannon communication channel. Each time a patient chooses a  physio, we have a message selected from 10 possible messages, and, if each physio is equally popular, the message conveys log2(10) =  3.32 bits.

What I'm thinking, is that choice is needed for information. In 'the difference that makes a difference", implicit in the 'difference' (#1 in the phrase), is a choice of one the alternatives.  It is tied up somehow with intention.

Can a difference without a choice makes a difference?

But in my example I wasn't really making a choice, or rather there was no information in my choice. If we'd gone a layer up, there was no selection. The apparent 3.32 bits of information was spurious.

This needs working through further.

Intention in the genetic code

I commented in the previous post that John Maynard Smith argues[1] that there is intention in the genetic code, and also that the genetic code uses arbitrary symbolism. The argument is that evolution has selected the genetic codes - the specific DNA sequences together with biochemistry to decode the sequences - for specific purposes. Thus, the DNA sequences and the biochemistry that ends up with the growth of my eyes exists in order to produce an eyes - which themselves exists so that I can see.

In other words, this code was selected by evolution so that I can see.

Contrast this with the tree rings. When you cut down a tree you can find out how old the tree is by counting the rings, so the rings contain information about the age of the tree. But there was no intention for the tree rings to contain that information, it is a chance by-product of the way a tree grows.  The  information in the DNA sequences that are the instructions for building an eye are by contrast there with the intention of building an eye. Furthermore, they are arbitrary, in the sense that other instructions to build an eye would do just as well, provided they built an eye.  I'm not a biologist, but I guess that different organism use slightly different biochemistry to grow eyes, and certainly there are different types of eye which all nevertheless still serve the same function of seeing.

Notice, incidentally, that this is in no way suggesting that the genetic mutations that led to the ability to grow eyes were themselves intentional (there's no intelligent design in here).  The intention came from the competitive advantage of having eyes, or, rather, being able to see, which led to the selection of the organisms containing mutations that led to eyes.

In conclusion: tree rings are, in the dichotomy of Floridi, natural [environmental] information; and genetic information is semantic [non-natural] information.

(A peacock's tail is also, by the way, semantic information from this line of argument.)

Ref 1

Smith, J.M.

The Concept of Information in Biology

Davies, P. and Gregersen, N. (ed.)

Information and the Nature of Reality

Cambridge University Press, 2010, pp. 123-145

Abstract & arbitrary symbolism, information and life

Serendipity led me to: Information, Intelligence, and the Origins of Life by Randy Isaac. (Perspectives on Science and Christian Faith, v63 n4 pp219-230)  The paper tackles the theory (put forward by Intelligent Design proponents, especially William Dembski) that consideration of information leads to the need for a creator God. Isaac disagrees with that thesis. (And so do I, for the record!)

Isaac's paper very nicely summarises many key issues in the field, but for this post I want to pick up on one specific issue.  He discusses the distinction between what he calls "physical symbolism" and "abstract symbolism". Abstract symbolism is what I've previously referred to as arbitrary symbolism (well, maybe, see below), and this binary distinction links to the distinction between environmental and semantic information (which I discussed here).

Here's some quotes from Isaac:

In physical symbolism, the symbol has a physical property which serves as the message. For example, a shiny, smooth metal surface can serve as a symbol of high reflectivity. Or a north pole of a magnet serves as a physical symbol of attraction to a south pole magnet. The meaning or significance of a physical symbol is derived from the physical properties of the symbol itself. 

In abstract symbolism, the symbol has a meaning assigned to it which does not necessarily derive from its physical properties.

So far so good. He later says:

Abstract symbolism is a hallmark of intelligence. ... The ability to associate abstract symbolic significance with a distinguishable physical pattern is a key indicator of intelligence, though not the only factor. ... When abstract relationships are a necessary part of information systems, then an intelligent agent must be involved to generate or interpret or design that system.

So, where does that leave the peacock? Is the peacock's tail not an abstract (arbitrary) sign? Or is the peacock intelligent?  Actually, when I wrote that about the peacock I was aware or a possible problem with my claim that "the sign is arbitrary since other birds manage with different signs meaning essentially the same thing": maybe the sign is not arbitrary to the peacock (or peahen). A peacock couldn't one day decide to woo a peahen with, say, a crest, instead of his tail. He wouldn't 'know' to do that and if he did the peahen wouldn't understand.  Still, I think non-human animals can work with abstract symbols, but it needs teasing out a bit more carefully, and I'm still not sure whether that being the case means that animals display intelligence or that arbitrary signs don't need intelligence.

John Maynard Smith in his chapter (The Concept of Information in Biology, page 130) in Davies and Gregersen (Information and the Nature of Reality) says of the genetic code:

The correspondence between a particular triplet and the amino acid it codes for is arbitrary. Although decoding necessarily depends on chemistry, the decoding machinery (tRNAs, assignment enzymes) could be altered so as to alter the assignments. Indeed, mutations occur that are lethal because they alter the assignments. In this sense the code is symbolic.

This is the same sense of arbitrary that I was using about the peacock's tail. (I wonder, maybe there is a difference between Isaac's 'abstract' and Maynard Smith's 'arbitrary'?)

Something to flag for discussion another day. Maynard Smith argues that there is intention in the genetic code. I didn't understand that when I first read his chapter but I think I do now. (His ideas have been digested by my subconscious and were regurgitated when I was reading Isaac's paper!) But that is for a later  post.

Environmental and Semantic Information: Natural and Non Natural Meaning

Luciano Floridi, in his map of information (see, for example, "Information: A very short introduction" Oxford University Press 2010, or Semantic Conceptions of Information on the Stanford Encyclopedia of Philosophy), has the primary distinction between Environmental and Semantic information.

Floridi's informational map, Source: http://plato.stanford.edu/entries/information-semantic/

This is closely associated with Grice's distinction between Natural and Non Natural meaning (H. P. Grice, "Meaning" The Philosophical Review, 1957, Vol. 66(3), pp. 377-388 http://www.jstor.org/stable/2182440).

I talked about information being data plus meaning in a previous post, which is what Floridi refers to as "The General Definition of Information (GDI)", so we see that the two categories of information derive directly from the two categories of meaning.

Floridi points out that his concept of environmental information might not be 'natural'. Here's how he explains it in the Stanford Encyclopedia of Philosophy (op. cit.):

One of the most often cited examples of environmental information is the series of concentric rings visible in the wood of a cut tree trunk, which may be used to estimate its age. Yet “environmental” information does not need to be natural. Going back to our example, when you turned the ignition key, the red light of the low battery indicator flashed. This signal too can be interpreted as an instance of environmental information. 

Environmental information is defined relative to an observer (an information agent), who is supposed to have no direct access to pure data in themselves. It requires two systems a and b to be coupled in such a way that a's being (of type, or in state) F is correlated to b being (of type, or in state) G, thus carrying for the observer the information that b is G [...]:

Environmental information:
Two systems a and b are coupled in such a way that a's being (of type, or in state) F is correlated to b being (of type, or in state) G, thus carrying for the information agent the information that b is G.

However, I don't think this in any significant way makes it different from Grice's 'natural meaning', or not in any way that matters for my concerns at the moment at least.

My interpretation is that semantic information/non-natural meaning requires intention, whereas environmental information/natural meaning 'just happens'.

Switching to the language of semiotics, a point about signs is that the signifier can be arbitrary. I'm wondering if there's a distinction between, say, natural signs and non-natural signs, where the signifier in a natural sign is necessary, but the signifier in a non-natural sign is arbitrary?  We might interpret the rings visible in the wood of a cut tree as the signifier in a sign meaning the age of the tree (or maybe not, I'm not sure about that), but in that case the rings can't be said to be arbitrary: they are a necessary consequence of the way the tree grows. However the word 'tree' as the signifier for a tree is arbitrary, in the sense that we could equally say "arbre" or "der Baum" (http://translate.definitions.net/tree).

One thing I'm still not quite ready to concede, though, is that non-natural/semantic information/meaning/signs necessarily requires people.

Certainly I want to suggestion that animals can be responsible for semantic/non-natural signs. I want to say that when a peacock displays his tail this is a sign with meaning, that there is intention on the part of the peacock, and that the sign is arbitrary (since other birds manage with different signs meaning essentially the same thing).

Source http://commons.wikimedia.org/wiki/File:Peacock.jpg

I'm still trying to get to grips with what we mean by 'meaning', however, and I'm not sure saying there is a distinction between natural and non-natural meaning gets us any closer to what meaning is. I'm still ploughing my way through Ogden and Richards, 'The Meaning of Meaning' (Routledge and Kegan Paul, 1949) but can't say it has been especially enlightening so far, and has been made a little more difficult by their style and the casual 'of-their-time' racism in one the examples they use to illustrate their argument. Wittgenstein's perspective that I mentioned in the previous post sounds rather more promising so perhaps I need to track that down.

Information and votes, electoral reform and 'the electorate'

A couple of random thoughts on the electoral processes following the recent UK election.

1 An election is of course an information/communication process. What is being communicated and between whom is not so straightforward. Well, maybe to whom is OK, since I guess it's the state (perhaps the Queen) - ie who or whatever forms the parliament. From whom is more difficult. A lot of talk in the media has been in terms of 'the electorate' (or 'the British people') 'saying' something (eg, that they wanted a hung parliament). I'm struggling with that. It's not that I have a problem per se with groups having agency. I think I accept that Britain does things, and even when I disagree strongly there is some sense in which Britain is doing that in my name. But that's after the election. There's a sense in which the present government is 'my' government and I am implicated in anything they do: but there is NO sense in which I voted for this government.

Somewhere in there there is something about semantics and intentionality. You can't conclude from looking at the results of the election that there is this thing called 'the electorate' that 'said' something or other.

2 There's a lot of talk about the need for electoral reform and about 'fairness' in the electoral system. I want electoral reform, but I'm not 100% sure what I want. Actually I quite like the idea of the AV system being proposed because I want to keep a constituency MP (though I've never voted for him my - Conservative - MP has listened and responded to me on a number for issues: I like that) but I want to be able to vote for who I really want rather than feeling the need to vote tactically. It seems AV would allow me to do that - though I get the impression I'm about the only person in the country who does want AV. Most people seem either to insist that there should be no change at all or else will settle for nothing short of full proportional representation (like the Electoral Reform Society).

Anyway, that's a bit of a digression. What I wanted to draw attention to was an item on BBC Radio 4's excellent More or Less last week. The short discussion of the maths of voting systems was good, but, in particular, I was interested in Arrow's Impossibility Theorem. What I liked about it was the approach that breaks down the assessment of a 'fair' system into simple statements of things that are required for a system to be considered fair. And the fact that it concludes:

no voting method can satisfy a certain set of desirable criteria, implying no voting method is ideal

(It reminds me of the conclusion of thermodynamics: you can neither win nor break even.)

Meaning in biology

A tweet from Richard Dawkins:

"46% of our DNA was provided in the past by invading viruses, 1.5% from primate & animal ancestors."

led me to this piece* "Friendly viruses aided human evolution" in the Irish times. I find it fascinating the way in which life intermingles, and it is an example of why the 'tree of life' model is a misleading over-simplification.

But the specific thing I started to think about arose from things like the following:

The same is the case with HIV/Aids in humans. “HIV 1 is an acute retrovirus like the viruses in our genome,” says Ryan. “Even at its worst, we know that it is changing and selecting specific genotypes that it feels most comfortable with, while killing off others.

(My emphasis.) You get this anthropomorphism all the time, imbuing meaning and intentionality to things that are 'really' purely mechanistic. (It is in there in the phrase 'selfish gene'.) I'm sure the authors would mostly dismiss it as a figure of speech, a shorthand. But one thing I'm arguing with my layered ideas of information, is that at each layer boundary we see information/data acquire meaning. Thus high and low voltages mean 1s and 0s, a particular pattern of 1s and 0s mean the letter 'a' etc. The thing is, is the meaning I'm talking about here the same as semantic meaning? I think the usual response would be: no, it isn't. But I'm wondering, do the layers go all the way up (and down)? Is there a point where there is a qualitative change to semantics?

I think it relates to the age-old problem of how you get conscious thought through physical mechanisms - mind-matter dualism. I've no answers, of course, as ever, just flagging something of interest.

---

*John Holden reporting on the work of Frank Ryan. "Frank Ryan will be publishing his third paper on viral symbiosis in the Journal of the Royal Society of Medicine later this month".

Intentionality: landscape art, and more on Damien Hirst

I caught Channel 4's programme about landscape art in a series on sculpture last night.

Led me on a train of thought...

Waldemar Januszczak, feeling the stones at Stongehenge, talking about the 'presence' of big sculpture... Angel of the North etc... I know what he means... set me thinking: "is this not about information then?" But he then went to Nancy Holt's Sun Tunnels in the Utah desert. These are aligned with rising setting sun at the summer and winter solstices, and also have holes in them that make the sun project patterns of the stars. These facts - the solstices and the stars - only have meaning because of the information. Or rather the intention of doing this.

... why are the stones at Stonehenge art but not stones left around by glaciers?

It is the intentional information. Same reason that painting of a tree is art but the tree itself is not art.

What about the nest of a Bowerbird? Or the song of Blackbird? (Which I happen to find most beautiful.)

A related question - is the nest of a Blackbird engineering?

And on to Damien Hirst. In the Guardian on Saturday, Robert Hughes explains his objection to Damien Hirst. "One might as well get excited about seeing a halibut on a slab in Harrods food hall" he says. Perhaps it is intentionality again - the halibut in Harrods is not put there with artistic intent. Hirst's sharks in formalin were.