Feyerabend 8

8

‘Initial difficulties caused by the change are defused by ad hoc hypotheses which thus turn out occasionally to have a positive function; they give new theories a breathing space, and they indicate the direction of future research.’

‘This is the place to mention certain ideas developed by Lakatos which throw light on the problem of the growth of knowledge, and which to some extent, undermine his own quest for law and order in science.

It is customary to assume that good scientists refuse to employ ad hoc hypotheses and are right to do so. New ideas, so it is thought, go far beyond the available evidence and must go beyond it in order to be of value. Ad Hoc hypotheses are bound to creep in eventually, but they should be resisted and kept at bay. This is the customary attitude as it is expressed, for example, in the writings of K. R. Popper.

As opposed to this, Lakatos has pointed out that ‘adhocness’ is neither despicable, nor absent from the body of science. New ideas he emphasizes, are essentially entirely ad hoc, they cannot be otherwise. And they are reformed only in a piecemeal fashion, by gradually stretching them, so that they apply to situations lying beyond their starting point. Schematically:

Popper: New theories have, and must have, excess content which is, but should not be, gradually infected by ad hoc adaptations.

Lakatos: New theories are, and cannot be anything but, ad hoc. Excess content is and should be created in a piecemeal fashion, by gradually extending them to new facts and domains.’

Popper’s notion of excess content –

a new theory is to have ‘excess content’ – relative to the older theory – if it is to be preferable to the older theory

if the idea is that  a new theory is only preferable to an older theory – if it adds ‘excess content’ to that older theory –

then all you have is ad hoc additions to the original theory

on this view there is no place for a genuinely different theory – with different content

science as an investigation of different theories comes to a stop

and as for Popper’s prohibition on ad hoc theories –

his notion of excess content depends on ad hoc theories

Popper’s view of the growth of knowledge – of scientific knowledge – is rendered  incoherent because of his view of ad hoc theories

the deeper problem is that Popper fails to see that the issue of scientific method – is open to question – open to doubt – and is uncertain

we need to be open to the many and varied descriptions of how science works –

and we need to look for new ways of describing science

it is not a matter of simply insisting that there is only one way for scientists to proceed –

to take such a view is to adopt and authoritarian attitude – which in the end comes down to nothing but pretence and rhetoric

for Lakatos – all new content – excess content – is ad hoc

now the question here is – are you dealing with new and different theories – or not?

if you are just talking about additions to the current theory – then you can say that such are ad hoc

but in that case – you have no new theory – just an older theory added to

if a genuinely new theory is proposed – what will mark it out – is that it is different

it will present a different picture of reality – it will come from a different perspective – its content will be different

Lakatos I think is a philosophical bureaucrat – he has no notion of genuine theoretical difference –

anything that comes into the in-tray gets absorbed into the balance sheet –

there is nothing that can’t be fitted into the expanding ledger

hard to say – but this just may be a good picture of a great deal of scientific practise

but it doesn’t account for genuinely new theories – genuinely different content

with their focus on ad hocness both Popper and Lakatos are in the same boat – they cannot account for new theories

if a new theory is not taken up as a prospect – for whatever reason – it stays on the outer

if a new theory is brought in from the cold – it plays a role –

just what that role is – will be determined by those involved

it may be regarded as adding to the current debate – or it may be seen as turning things in a new direction

if the later – is that  ad hoc – or a real change in propositional perspective?

however it is regarded – however it is described – the propositional action is open to question – open to doubt – and is as a matter of fact – uncertain

we can get too hung up on this idea of the ad hoc –

I’m inclined to think it is a non-issue

the basic problem with Lakatos’ view is that it renders the notion of ad hoc meaningless

ad hoc to what – if every hypothesis is ad hoc?

the best you can say of Lakatos’ contribution here is that he proposes that scientific theories or hypotheses can be or are related to each other

‘being related to each other’ – is not a strong enough criterion to distinguish anything from anything

the result is that Lakatos’ view is of no consequence for science

Feyerabend says he sides with Lakatos over Popper and wants to show that the early history of Galileo’s mechanics ‘tells exactly the same story’

he looks at some statements on the nature of motion from Galileo’s early works Du Motu and Dialogue on Motion and concludes that Galileo is in agreement with Aristotle’s general theory of motion – where a mover is postulated for every motion

‘Galileo seems to accept this part of the theory, both when letting rotating spheres slow down and when accepting the ‘force of the intelligences’. He also accepts the impetus theory which attributes any motion to an internal moving force similar to the force of sound that resides in a bell long after it has been struck, and is supposed to ‘gradually diminish.

Looking at these few examples, we see that Galileo ascribes a special position to motions which are neither natural or forced. Such motions may last for a considerable time, even though they are not supported by the surrounding medium. But they do not last forever,

and they need an internal driving force in order to persist even for a finite time.

Now if one wants to overcome the dynamical arguments against the motion of the earth (and we are here thinking about its rotation rather than its motion around the sun) then the two underlined principles must both be revised. It must be assumed that the ‘neutral

motions’ which Galileo discusses in his early writings, may last forever, or at least for periods comparable to the age of historical records. And they must be regarded as ‘natural’ in the entirely new and  revolutionary sense that neither an outer nor an inner motor is needed to keep them going. The first assumption is necessary to account for the daily rising and setting of stars. The second assumption is necessary if we want to regard

motion as a relative phenomenon, depending on the choice of a suitable co-ordinate system. Copernicus, in his brief remarks on the problem, makes the first assumption, and

perhaps the second. Galileo takes a long time arriving at a comparable theory. He formulates permanence along a horizontal line as a hypothesis in his Discorsi, and he seems to make both assumptions in the Dialogue.’

Feyerabend says –

‘My guess is that a clear idea of permanent motion with(out) impetus developed in Galileo only together with his gradual acceptance of the Copernican view. Galileo changed his view about the ‘neutral’ motions – he made them permanent and ‘natural’ – in order to make them compatible with the rotation of the earth and in order to evade the difficulties of the tower argument. His new ideas concerning motions are, therefore, at

least partly ad hoc. Impetus in the old sense disappeared partly for methodological

reasons (interest in the how, not in the why – this development itself deserves careful study), partly because of the vaguely perceived inconsistency with the idea of the relativity of all motion. The wish to save Copernicus plays a role in either case.’

Feyerabend says that Galileo in dropping the Aristotelian ideas and replacing them with the idea of permanent motion without impetus – was adopting an ad hoc strategy – a strategy designed to support the Copernican view

dropping the Aristotelian ideas and replacing them with the idea of permanent motion –

can be seen as an addition to the Copernican view –

this is a fair enough interpretation –

however this issue is really just and only a matter of perspective

you could also argue that the key theory that Galileo was concerned to advance was the theory of permanent motion

and that the Copernican view – was an addition to this theory – an addition designed to give the permanent motion theory substance and credibility

my overall point here is that we can look at the development and function of theories – and hypotheses – from different perspectives –

there is no one way of viewing the matter – just as there is no one way of proceeding in science

                                                                                                                                      

Feyerabend proceeds –

‘Now if we are right in assuming that Galileo framed an ad hoc hypothesis at this point, then we can also praise him for his methodological acumen. It is obvious that the moving earth demands a new dynamic. One test of the old dynamics consists in the attempt to establish the motion of the earth. Trying to establish the motion of the earth is the same as trying to find a refuting instance for the old dynamics. The motion of the earth, however, is inconsistent with the tower experiment if this experiment is interpreted in accordance

with the old dynamics. Interpreting the tower argument in accordance with the old

dynamics, therefore, means trying to save the old dynamics in an ad hoc fashion. If one does not want to do this one must find a different interpretation for the phenomena of free fall. What interpretation should be chosen? One wants an interpretation that turns the motion of the earth into a refuting instance of the old dynamics, without lending ad hoc support to the motion of the earth itself. The first step towards such an interpretation is to establish contact, however vague, with the ‘phenomena’ i.e. with the falling stone, and to establish it in such a manner that the motion of the earth is not obviously contradicted. The most primitive element of this step is to frame an ad hoc hypothesis with respect to the rotation of the earth. The next step would be to elaborate the hypothesis, so that

additional predictions become possible. Copernicus and Galileo take the first and primitive step. Their procedure looks contemptible only if one forgets that the aim is to test older views rather than to prove new ones, and if one also forgets that developing a good theory is a complex process that has to start modestly and that it takes time.’

‘The most primitive element of this step is to frame an ad hoc hypothesis with respect to the rotation of the earth’

yes – in terms of the Galilean / Copernican problematic –

relative to the theory of the motion of the earth – an hypothesis with respect to the rotation of the earth – is an addition to that theory –

in that the idea is that it reconciles free fall – with the theory of the motion of the earth and challenges the tower argument of the old dynamic

you would have to say this additional hypothesis strengthens the theory of the motion of the earth – as an alternative to the old dynamic

however the ‘addition’ – is only an addition – if view historically –

in terms of the overall theory – it could well be seen as integral

there is nothing against an historical view of theory development – however it is only one view – and a plodding one at that –

‘piecemeal’ is Lakatos’ term here

‘Their procedure looks contemptible only if one forgets that the aim is to test older views rather than to prove new ones, and if one also forgets that developing a good theory is a complex process that has to start modestly and that it takes time.’

science just is this argument – an argument of different perspectives – different theories –

and in terms of the logic of any scientific argument – it is irrelevant which is the old view and which is the new –

that is to say the history of any scientific argument – is from a logical point of view – irrelevant –

it is history – here – that is ad hoc –

and of no value to the problematic per se

now with regard to how Feyerabend uses the notion ad hoc –

Lakatos’ view is that every theory and every addition or subtraction to or from a theory is to be regarded as ad hoc

Feyerabend also runs with this version of ad hoc when he says –

‘The historical material I have been discussing ….lends unambiguous support to the position of Lakatos. The early history of Galileo’s mechanics tells exactly the same story.’

no description – no historical theory – if it is held rationally – is without question – is without doubt – is certain

the matter is not ‘unambiguous’ –

I would have thought that just this would be Feyerabend’s point –

after all – he represents himself as the ‘anything goes’ man

and if  ‘anything goes’ – then one particular view – is not the end of the story –

you would expect at the very least the presentation of a range of different views

this argument over ad hoc hypotheses really goes back to Popper – who make a big deal of it – when in fact it is a non-issue

what’s in and what’s out of a theory – what strengthens and what weakens it – or what is irrelevant to it –

these are matters that are never finally settled –

nevertheless in practise – decisions get made –

and any decision here – is open to question – open to doubt – and is uncertain

in conclusion I find Feyerabend’s support of Lakatos’ view quite puzzling

which brings me to the claim that Lakatos is a fellow anarchist –

that is a stretch – that is very hard to explain

and by the way fellow anarchist to who?