A Methodological Science

(This is the third part – of four – of the talk ‘The Science Delusion’ and follows on from the introduction to (conventional, i.e. metaphysical) Physicalism in part 2 to return to the idea of science as a method introduced (through the words of Rupert Sheldrake) in part 1.)

 

10. Methodological Physicalism

The problem with both dualist and physicalist approaches, which leads to the conflict between science and religion, is that they are metaphysical positions. They are both making claims about what really exists, over and above the everyday existence of objects.

In contrast,  Sheldrake talked of science as a method of science. Recall:

… But there’s a conflict at the heart of science between science as a method of enquiry, based on reason, evidence, hypothesis and collective investigation, and science as a belief system, or a worldview…

This conflict between methodology and metaphysics is commonly applied to naturalism:

But in this talk I’m using the term ‘physicalism’ rather than ‘naturalism’, as already noted, to emphasize:

  • Its monist (non-dualist) agenda
  • Its reductionist agenda

Hence I use the rarely-used terms ‘methodological physicalism’ (an oxymoron to some) and ‘metaphysical physicalism’ (a bit of a mouthful).

Whereas ‘metaphysical physicalism’ asserts what is, ‘methodological physicalism’ merely takes a stance –  a hypothesis, ‘as if’ what is claimed were true.

Michael Polanyi’s notion of a ‘personal commitment’ is at the strong end of the scale of claims of a methodological assertion. It is an individual effort, required to fully explore possibilities, defending an idea so that it is not prematurely rejected. This puts it at odds with a Popperian falsification. It is akin to an Athenian rather than Spartan approach to child-rearing: to nurture them until they are mature enough to stand on their own two feet rather than exposing them to the full forces of nature from the beginning. But this personal commitment still falls short of saying ‘it really is this’.

One example of a methodological approach as a systematic approach that may be familiar is Descartes’s ‘Methodological scepticism’ , also known as Cartesian Doubt.

 

11. Dennett the Dualist

To labour the point of a methodological position (or ‘stance’), I will provide another example, namely, Daniel Dennett’s ‘intentional stance’. Or rather, I will present my interpretation of his ‘intentional stance’ to try to illustrate the point.

Dennett’s ‘intentional stance’ is a way of considering an object ‘as if’ it has a mind of its own, with its own intention. This is in contrast to the ‘physical stance’ (example: a thermostat includes a strip of steel and a strip of copper) and the ‘design stance’ (example: a thermostat includes a temperature-sensitive device, invented by John Harrison). We learn to interact with objects around us by applying the appropriate stance towards things. For people, we learn to apply the intentional stance but we can take any particular stance towards any object. The issue is merely whether it is useful for us to do so. So, to take an intentional stance with the thermostat might be to say: ‘a thermostat wants to maintain the temperature of a room by turning the heating on when it’s too cold and off when it’s too hot’. With the intentional stance, it is ‘as if’ we have attributed a mind to an object (be it a person, a thermostat or anything else). And when people think of ‘mind’ they typically conceive of it being distinct from matter. It is as if we’re being a dualist. But with any stance, we are not making any commitment to how things ‘really are’.

To frame this into methodological vs metaphysical terms, when Dennett (someone we consider to be antithetical to dualism) takes the intentional stance, he is being a ‘methodological dualist’ whilst at no point being a ‘metaphysical dualist’.

 

12. Dogmas, Pragmas and Morals

Returning to the list of values of science set out by Christopherou or the list of dogmas by Sheldrake (in ‘Dogma and Habit’), we can apply those values in different ways.

The initial position is that we can apply them in a metaphysical way: ‘I believe in the dogmas of science because they are true’.

But an alternative way is to apply them in an epistemological way: ‘I believe in the pragmas of science because they are useful’ (in as far as knowledge is useful). Pragmas are ‘rules of thumb’, heuristics, tools. They are not absolute.

Philosophy can be divided up into some major sub-disciplines:

  • Metaphysics / Ontology
  • Epistemology
  • Ethics
  • Aesthetics
  • Political philosophy

and many other sub-disciplines, often closely related to the sub-disciplines above. For example, the sub-discipline of the philosophy of science is closely associated with epistemology.

Looking at this list of sub-disciples, we can see there are other ways of applying the values of science. So, a third way of applying the values of science is ethically: ‘I believe in the morals of science because they are right’. I’m not going to say any more about this possibility now; it is just worth noting that it is a possibility. (And I suppose there is a fourth way: ‘I believe in the aesthetics of science because it is beautiful’.)

 

13. The Metaphysical and the Methodological

To hopefully illuminate the distinction between the metaphysical approach and the methodological, here is another crude comparison table with some notes…

Metaphysical Methodological
Dogmatic Pragmatic
Positivist Pragmatist
Absolutist Balanced
“is” “as if” (model)
Ontological Epistemological
Noumenal Phenomenal
Realism Anti-Realism
Objective Subjective
Dogmas Pragmas
  1. Dogmatic / Pragmatic: as already noted, the dogmatic implies fixed/unchanging (a fixed mark, around which everything else must be accommodated) whereas the pragmatic is amenable to change.
  2. Absolute / Balanced notion of truth: metaphysics provides absolute truths whereas methodological does not. But this does not mean that it is just relativist. The previous talk introduced a notion of truth as a balance between internal coherence on the one hand and correspondence with the external on the other. Knowledge is personal (person-specific) but constrained. It is not free to believe just anything.
  3. Objective / subjective: ‘about what is’ versus ‘dependent to some extent on the observer’.
  4.  ‘is’/’as if’: ‘what is there’ versus ‘as if it was like this’, where ‘this’ is a local model here of what is there. (I will have more to say about models later on.)
  5. Ontological / epistemological: ‘what is’ versus ‘what is known’.
  6. Noumenal / phenomenal: Kant’s distinction between the unknowable ‘thing-in-itself’ (the noumenal) and the mere appearance of whatever it is that is behind the phenomenal curtain.
  7. Realism / anti-realism: This is worth looking at in more detail…

 

14. Realism and Anti-Realism

There are many doctrines that go under the name of ‘realism’ – depending upon what is claimed as being real, but the most common is the one that claims that material ‘stuff’ really does exist – it is not a figment of our imaginations (it is not ‘idealism’ = idea-ism) and, more specifically, that only ‘stuff’ exists, denying that ‘mind’ (ideas) is separate. Over the past few decades this term has largely been superceded by ‘materialism’ or ‘physicalism’, the latter becoming more popular. As already noted, I prefer the latter term as is does not have everyday connotations that the former suffers from.

Within the Philosophy of Science, there is the rather separate debate of Scientific Realism versus Scientific Anti-realism. The former has the diluted claim that the world studied by science really exists – that scientific theory is ‘at least approximately true’ (whatever that means). And the latter also has a diluted claim in that it generally no longer denies that the everyday (approximately human-sized) objects that obviously seem to exist really exist. The issue is whether concepts postulated by science that are not obvious really exist – for example, sub-atomic particles.

There is the danger that we can then get bogged down in debates about whether the trees over there in the distance really exist because I can only recognize them when I have my glasses on. At this point, I want to avoid these problems and just effectively equate a non-realist position with methodological physicalism and so remain non-commital (sceptical) as to whether even everyday objects ‘really exist’.

There are many factions within the anti-realist / non-realist camp. The most significant are the instrumentalists and the constructivists. Instrumentalists treat science as an ‘instrument’ –  a tool to explain and predict phenomena. Constructivists see science as a community measuring and making models of the natural world. (Note the underlined terms highlighted are recurring themes – the use of models, to predict.) But all anti-realists are rejecting the idea that science is converging on ‘the truth’.

 

15. A Confutation of Convergent Realism

Recall that:

  • Realism is the doctrine that ‘stuff’ really exists.
  • Scientific realism is the doctrine that scientific theories really tell us things about what really exists. It is claimed that the theories are at least approximately true.

‘Convergent realism’ is the doctrine that, as science progresses, it gets closer and closer to ‘the truth’ – the approximations are better and better.

Now, why should an anti-realist think that science is not converging on the truth?

The problem is that there is a long history of ideas being accepted as being ‘really true’ only for them to be subsequently rejected by the scientific community. They are then just seen as being plain wrong – going up a dead end: they weren’t even approximately true! Larry Laudan’s paper ‘A Confutation of Convergent Realism’ (1981) provides a well-known list of such theories:

  • The crystalline spheres of ancient and medieval astronomy
  • The humoral theory of medicine
  • The phlogiston theory of chemistry
  • The effluvia theory of static electricity
  • ‘Catastrophist’ geology, with its commitment to a universal (Noachian) deluge
  • The caloric theory of heat
  • The vibratory theory of heat
  • The vital force theories of physiology
  • The electromagnetic aether
  • The optical aether
  • The theory of circular inertia
  • Theories of spontaneous generation

Most people will be unfamiliar with most of these theories, which are now consigned to the history of science. If some are known, they will almost certainly be among the few near the start of the list.

James Burke’s BBC book and TV series ‘The Day the Universe Changed’ (1985) looked at revolutions of ideas and included some more commonly-known examples of obsolete scientific theories in discussing transitions:

  • From the Ptolemaic (geocentric) system via the Copernican (heliocentric) system to the modern expansionary model due to Georges Lemaître and Edwin Hubble.
  • From the Newtonian ‘billiard ball’ universe to Einsteinian relativity;
  • From (Newton’s) corpuscular theory of light, via wave theory to the wave/particle duality.

Burke’s title is a good one: of course, no one believed that the Universe actually changed the day a new theory was adopted. It was only ‘as if’ it had changed; it looked different because our knowledge of it had changed.

Imagine a realist who said the geocentric system was ‘really true’ until presented with evidence to the contrary and who then adopts the heliocentric system, saying that that theory was ‘really true’. They are then presented with evidence for the expansionary universe. They lose credibility for asserting that their adopted latest theories are ‘really true’.

 

16. Instrumentalism, or Alternatively, Modelism

One of the isms I am subscribing to here is Instrumentalism – the major ism countering scientific realism, which is sometimes referred to as a form of anti-realism and sometimes as non-realism, outside of the realism/anti-realism debate.

(Note that scientific realism is the prevalent position among philosophers – see the PhilPapers survey: 75% compared with 11% for scientific anti-realism. Non-realism was not an explicit option. Those in favour of instrumentalism will be drawn from some of the anti-realism voters and some of the ‘other’ 14%.)

Now, consider a graduated scale with, at one extreme, ‘ideas’ and at the other end, ‘reality’. In the simplest terms, scientific knowledge is a set of ideas that tell us about that reality. Scientific method (such as it exists, c.f. Paul Feyerabend’s ‘Against Method’) is about how we generate those ideas from that reality (and it can be chaotic).

Consider a number of intermediate points between ideas and reality on this scale:

  1. Ideas
  2. Descriptions
  3. Models
  4. Instruments
  5. Phenomena
  6. Reality

We can’t access ‘reality’ i.e. the noumenal world, but only the phenomenal world. And those phenomena may only be observable via instruments e.g. telescopes and microscopes.

At the other end of the scale, we generally access scientific knowledge not from ideas (conscious or otherwise) inside our head but from written descriptions. But normally, descriptions, in ordinary language, are vague and prone to subjective misinterpretation. They are often metaphors (we use ‘is’ rather than the similes ‘like’/’as if’ – a metaphysical/methodological blurring).

In the middle of the scale are models. Traditionally, they were codified in the international language of mathematics. But the model may alternatively be codified in a computer language instead – in order to describe behaviour that is very complex. For an example of a model at the most complex end of the spectrum, a major aspiration of the Human Brain Project is a whole brain simulation.  Models, unlike descriptions, are precise. If I simulate using the same model and the same inputs as you, I will get the same results as you. This allows repeatability, independent verification, and precise communication of ideas.

But having good models is not everything. A good model of the human brain that enables us to run accurate simulations will not provide explanations (descriptions) that will provide us with intuition (ideas, theories) about the model (or its concomitant phenomena/reality).

(A good model of the human brain will however allow us to assert that brain behaviour is entirely physically based and then provide us with a platform to tinker around with in order to abstract to higher levels in order to gain an understanding of the brain, in the way we cannot ethically tinker around with real brains.)

So science is not about just one of these steps – that leads to an eccentric position:, e.g. idealism, realism, phenomenalism, …). Instead it is about all these steps in parallel. Individual scientists may be focussed on a single stage but science requires all stages in appropriate proportions.

However, if we were to latch on to one of these steps to give the whole approach a name, ‘instrumentalism’ might have been adequate 100 years ago but now ‘modelism’ seems more appropriate.

An instrument is a tool, but so is a model.

We are increasingly becoming aware of the importance of models and of models being precise behavioural descriptions far beyond traditional mathematical formulae. It is only in recent years that the phrase ‘model-based’ is being applied to various approaches (methods) in science and engineering. We should place models at the heart of our thinking about science, as it is in the middle of the scale.


‘The Grand Design’ (2010)

Even some self-styled realists subscribe to a model-driven approach. For example, in ‘The Grand Design’ (2010), Stephen Hawking and Leonard Mlodinow talk of ‘model-dependent realism’ and of ‘reality accessible only through models of phenomena’.

To summarize, here is the scale again, with a bit more flesh on it…

  • Ideas: (i) conscious and (ii) tacit
  • Description: in vague language, including metaphorical language
  • Models: precise simulations, predictive. (i) mathematical: formulae in the mathematical language and (ii) computational: complex algorithms in an artificial language
  • Instruments: (i) Measuring: converting from some physical measure to an observable measure and (ii) Observing: converting from an insensible realm to a sensible one.
  • Phenomenon (the phenomenal)
  • ‘Reality’ (the noumenal)

 

17. Instrumentalism and Constructivism

Daniel Dennett’s ‘Tower of Generate and Test’ provides an evolutionary scale of creatures:

  • Darwinian: evolutionary (randomly varying i.e. mutating)
  • Skinnerian: adaptive
  • Popperian: using internal models (hypothesizing)
  • Gregorian: using external tools (physical and linguistic)

In ‘Scientific Creatures’ I confined Gregorian creatures to just physical tools and extended the scale with various stages of scientific creatures, which extend their capabilities by applying the generation, variation and adaptation of models to outside of the bounds of the creature:

  1. Scientific-I: create external models of the environment.
  2. Scientific-II: create objects in the external environment to ‘hypothesize’, i.e. automatically stimulate the environment and detect the response.
  3. Scientific-III: create external adaptive models.
  4. Scientific-IV: create external model generators.

This scale of individual intelligence was then complemented by an orthogonal scale of social intelligence, which was far less detailed but basically extended from individuals through pre-linguistic communicating groups, language and writing to elaborately organized societies of scientists, interacting via peer review.

IntelligenceAndCooperation

Science = Instrumentalism & Constructivism

Now the focus is about science rather than intelligence and so I present the same picture as follows:

  • The vertical axis represents non-realist Instrumentalism, dealing with the individualistic aspects of science.
  • The horizontal axis represents anti-realist Constructivism, dealing with the social aspects of science.

To elaborate a bit further with Constructivism, the horizontal axis is where the scientific community constructs jargon and standards that are ‘just’ social conventions – they could quite easily be different – but are essential for scientific progress. Two examples:

  • The definition of scientific terms such as ‘planet’, which saw Pluto being reclassified it as a dwarf planet in 2006.
  • Our model of the Earth has changed over the many centuries from flat, to a sphere, to an oblate spheroid but for greater accuracy, scientists now use the WGS-84 standard, which provides the coordinate system for GPS. That standard is the culmination of international agreements following pragmatic conventions ultimately dependent on historical contingency – the use of the Greenwich Meridian followed by early radio navigation systems (so that the Greenwich Meridian now lies about 100 metres West of WGS-84’s 0°).

Neither the Instrumental nor the Constructivist axes on their own are adequate. It is only when the two are viewed together that we get a good picture of what science is about – a methodological science.

 

(Talk to be concluded in the next part.)

 

Advertisements
This entry was posted in Uncategorized and tagged , , , , , , , , , , , , , , , , , , . Bookmark the permalink.

Leave a Reply

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

WordPress.com Logo

You are commenting using your WordPress.com 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