If, as he argues, life were bound to arise, not as an incalculably improbable accident, but as an expected fulfillment of the natural order, then we truly are at home in the universe.

Note "natural order", not supernatural order, and you purport to be an opponent of naturalism. For Kauffman the natural order is "self-organization, selection, and chance". You need to read the book and take it in if you want to understand his position.

Among the estimated one hundred billion solar systems in the known universe, evolving life is surely abundant. That evolution is a process of "becoming" in each case. Since Newton, we have turned to physics to assess reality. But physics alone cannot tell us where we came from, how we arrived, and why our world has evolved past the point of unicellular organisms to an extremely complex biosphere.


Hence - not physicalist. Science itself has evolved beyond materialism, 1960's academic philosophy is about 100 years behind. — Wayfarer

I haven't read the new book. I don't believe he's changed his position, but I may be mistaken. He says in At Home in the Universe that physics alone cannot describe the processes of phase transition and autocatalytic systems because they are statistical in nature; that is they rely on nothing beyond chance. Hence complexity theory.

To say that something cannot be described or accounted for in terms of physics is not say that there is anything beyond the stochastic behavior of physical systems at work. There does not need to be any "divine" influence, for example, and as I read him in the earlier work Kauffman is certainly not aiming to support the notion that there is any such thing. If you want to argue that he is supporting such ideas in the later work you will need to present citations from the author himself.

An article worth reading: Confirmable and influential metaphysics. — Banno

The paper makes a good point : not all "influential" theories are "confirmable". :smile:

Watkins : "There are two other kinds of existential statement which are unempirical. The first alleges the existence of something un-localised and abstract, e.g. 'There is a law of nature governing these phenomena'. This statement will get verified, or at least strongly confirmed, if a hypothesis happens to be invented from which can be derived numerous precise and successful predictions about the phenomena in question. But it will not be refuted if no one manages to hit upon such a hypothesis".

For example, Quantum Physicists have, to the dismay of Einstein, hypothesized non-local phenomena, that meet the criteria for spooky Metaphysics, but have nevertheless been useful and "influential" in Physics. :nerd:

"Nonlocality suggests that universe is in fact profoundly different from our habitual understanding of it,"
https://www.physicsoftheuniverse.com/topics_quantum_nonlocality.html

So, as the paper suggests, we should not automatically reject all metaphysical theories out-of-hand, just because they are not empirical, but merely inferential. In some cases, they help us make sense of realities that are "outside of human sense perception". Hence, they are "Influential", true or not. :chin:

Metaphysics : referring to an idea, doctrine, or posited reality outside of human sense perception. . . .
. . .Under the skeptical analyses of the philosophical movements known as postmodernism and deconstructionism, all of these facts have resulted in a modern repudiation of both metaphysics and science. Their criticisms are based on the cultural and historical relativity of all knowledge. These two philosophical "schools" deny any existence at all of an objective or universal knowledge. Thus, metaphysical claims stand today between the absolutist claims of science (scientism) and the complete relativism of postmodernism and deconstructionism.
https://www.pbs.org/faithandreason/gengloss/metaph-body.html

Note "natural order", not supernatural order, and you purport to be an opponent of naturalism — Janus

As conceived by ‘Enlightenment’ philosophy, wherein there’s an irreconcilable division between ‘science and religion’, and scientific naturalism is to supersede religion, by ‘explaining’ it, and everything else, in terms of a biological adaptation. And of course you then must conclude that when Kaufman, whom you brought up, says that ‘physics alone can not tell us’, that actually he’s saying that there’s nothing beyond physics.

metaphysical claims stand today between the absolutist claims of science (scientism) and the complete relativism of postmodernism and deconstructionism. — Gnomon

:up:

And of course you then must conclude that when Kaufman, whom you brought up, says that ‘physics alone can not tell us’, that actually he’s saying that there’s nothing beyond physics. — Wayfarer

I remember that he says in the book that he is not proposing anything beyond complex physical systems; systems, however, which cannot be explained just in terms of physics. I think you just don't want to get that. Biology in general is not reducible to physics, even though biology is not anything other than physical. Read the book again, with an open mind, and then get back to me.

Pretty much.

It's important to note the specialist use of "metaphysical" to mean "unfalsifiable"; and "unfalsifiable" as being of specified logical form - those containing uncircumscribed existential statements, and their variations... anything that does not fit modus tollens...

Isn't this dealt with in the article - being the topic of Section VII? Or have I misunderstood you? — Banno

It defers the problem of how you go about "weighing" an all and some statement to the context of the all and some statement:

I do not think that any single criterion, such as conformity with existing science, can be laid down for assessing haunted-universe doctrine. This task is more like assessing the worth of a man's character than the legality of his acts...

But although these doctrines cannot be proved or refuted they can be criticised and weighed

Which seems a cop out to me. The account is happy to quantify over research programs and abstract over their content, it's also happy to talk about the types of relations between components of research programs - relating them is what all and some statements do. But when it comes to a general description of how this regulative role of all and some statements occurs - it defers to context, appealing to a pragmatic inability to abstract in a similar way to what the article's already done.

IE:

But although these doctrines cannot be proved or refuted they can be criticised and weighed

How?

Well, three levels are offered; disguised analysis; Lewis-Carol-like nonsense; and the intermediary of not having any truth value. For my part I don't see that a statement without a truth value has much by way of meaning...

But that's not quite right, either.

This has flared up in physics in respect of string theory and the multiverse; one side is arguing that these theories are not falsifiable in principle, so, not empirical, so, not really science; the other side is accusing those critics of being popperazi. — Wayfarer

I don’t know much about string theory but IF it cannot possibly be falsified, then I agree it is not a scientific theory. The problem with unfalsifiable theories is that they have no predictive value. You cannot use them to predict what will happen next. Therefore they do not add to our control and abilities. All they do is help us make sense (describe) an event or condition after it occurred.

I personally disagree with Popper on psychoanalysis. True that Freud himself kept tinkering his theories all the time and did not base them on much evidence, but his theories are nevertheless inherently falsifiable, and his tinkering of them shows an effort to fit the data better. E.g. he added ‘Thanatos’ (desire of death) to ‘Eros’ (the libido) after WW1, because he could not explain otherwise the extent of hatred, death and destruction seen in this war. So he postulated an instinct of death and destruction, operating side by side with an instinct of love and life. That shows a bona fide effort to explain experience. Other tenets of Freud have been invalidated by experience, such as ‘penis envy’.

Finally, psychoanalysis have led to the development of many related theories and approaches, some of which are practical and effective, e.g. transactional analysis, or Bateson’s analysis of family dynamics and ‘double-binds’.

Falsification fails to demarcate science from non-science both because scientists make use of non-falsifiable theories (as Watkins shows) and because falsification fails to solve the problem of induction. — Banno

The key to understanding the relationship between philosophy (metaphysics) and science (physics) is to realize that philosophy is a science. And the conclusions of one branch of the investigation of reality must not contradict those of another. All knowledge must be integrated

At root, science identifies and integrates sensory evidence (which is the nature of reason). Science is essentially based, not on experiment, but on observation and logic; the act of looking under a rock or into a telescope is a scientific act. So is the act of observing and thinking about your own mental processes (proof of one's conclusions to others comes later, but that is argumentative, not inquisitive.) Science is willing to accept and integrate information from any observational source, without concern about persuading other people.

They are both the same in that they both gather knowledge through observation and then classify this knowledge, and through classification, elaborate general principles or ideas. Science is simply organized knowledge.

Well, three levels are offered; disguised analysis; Lewis-Carol-like nonsense; and the intermediary of not having any truth value. For my part I don't see that a statement without a truth value has much by way of meaning...

But that's not quite right, either. — Banno

The intermediary and the disguised analysis are the interesting cases.

If it's disguised analysis, the all and some statement should somehow engender a falsifiable statement; but then if you treat that engendering as logical entailment the original all and some statement can be modus tollens'd through the falsifiable one. That would make the all and some statement falsifiable, so it can't be that. So what sense of entailment is it in the disguised case?

Referring back to the op, I haven't read the article, I confess. But I think we definitely need to consider these as distinct categories, "confirmable" and "influential". If I look for propositions which are as close to being truly unfalsifiable as possible, I am lead toward mathematical axioms which deal with the concept of infinity, or the proposed infinite. Such axioms are extremely difficult to falsify, being a matter of intuition, definition, or maybe as some would say a priori, so the falsifiability of them is very low, yet they prove to be very influential due to their usefulness. The problem is that the status of influential is provided for by usefulness, and this does not necessarily imply confirmation.

But this implies that no proposition is decidable prior to test. Which further implies that for all propositions for which no appropriate test is available, truth or falsity is indeterminable/undecidable. — tim wood

We actually do make propositions which are supposed to be true by definition, therefore not requiring testing, as is the case in mathematical axioms. But I do not consider these axioms to actually be "true", as you might know from my postings on other threads. Further, these axioms do tend to get tested through application and usefulness. However there seems to be confusion on this forum as to how usefulness relates to true. The usefulness of a mathematical axiom does not verify it as true. So these axioms get into a sort of limbo position where they are unverified, and unfalsified, yet heavily used, therefore very influential.

Therefore, if any propositions ought to be represented as unfalsifiable, and unverifiable, these would be mathematical axioms. Our mode for testing them is usefulness rather than truthfulness. So in science we have a clash of these two distinct forms of judgement. We tend to think that scientific hypotheses are judged according to principles of empirical observation. And that is what is supposed to constitute a judgement of truth. However, we rarely take into account how the mathematical axioms which are judged according to usefulness, rather than empirical truth, influence these empirical judgements.

You're forgetting the other half of the picture. If the proposition is also unverifiable, then why should we believe it is true? — Janus

Unverifiable and unfalsifiable are supposed to be two opposing extremes, a proposition which is claimed to be both ought to be simply be an irrelevant subjective statement, of no import to science. However, we have the status of mathematical axioms being "useful" as explained above, which demonstrates otherwise. So we are tempted to believe that such a proposition (like a mathematical axiom) is true because it is useful and influential.

As the article pointed out there are kinds of propositions which are unverifiable: "all x are Y", but falsifiable, and there are other kinds of propositions which are unfalsifiable: "some x are Y", but verifiable. In the latter case your position would entail that it is necessarily true that some x are Y, but that is nonsense; we could not know that until and unless it had been verified. — Janus

You misunderstand what I said. What I said is that the statement "some X are Y" ought not be classified as unfalsifiable simply on the basis of the apprehended human capacity to judge the truth or falsity of it. That would make "unfalsifiable" a subjective judgement without any objective principles by which we might make that judgement. It does not render "some X are Y" as necessarily true, because only propositions which are truly and objectively "unfalsifiable" would be necessarily true, and this proposition is not.

...all and some statement should somehow engender a falsifiable statement; — fdrake

Circumscribing reduces all-and-some statements to falsifiable form. Conservation laws, for instance, are used in particular cases to predictable and falsifiable ends. Rolling marbles, for example, can be used to display conservation of momentum in a limited case. IF this did not occur, the universal application of the conservation law would not be falsified, but it would have been undermined. So when you rolled a marble int a string of marbles in a track, one marble flies off the other end, and apparently, momentum is conserved. If, instead, the marbles simply stoped dead still, and the experimented mumbled something about the momentum being taken up at some unspecified place in the universe in order to defend the conservation law, one would be less inclined to take the law as true.

There's quite a body of discussion on falsifiability. SOme familiarity with that would be helpful.

Circumscribing reduces all-and-some statements to falsifiable form. — Banno

That seems to be one way of doing it. "Reduces" hides an awful lot of work though.

Rolling marbles, for example, can be used to display conservation of momentum in a limited case. — Banno

Let's go through it as a worked example then.

A really general statement of the conservation of momentum would be "For all systems which have a translation invariant Lagrangian there exists a conserved quantity under that action; linear momentum". That's an all and some statement with a proof; so it's verifiable and falsifiable in some sense of both words. However, by itself it's not a description of a reality; it's not put to that use. If that statement was taken as the hard core, it's an all and some statement of a different type since it can be shown to be true, and it can be shown to not apply (if the Langrangian isn't translation invariant).

However, it is quite suggestive of the claim that "Every physical system whose dynamics aren't sensitive to where in space it takes place will conserve linear momentum". That one has a lot of flexibility to it; points of ambiguity for auxiliary hypotheses at "physical system", "dynamics", the concept of sensitivity, how "conserve" applies when friction's thrown into the system and so on. Seems more like the all and some statement talked about in the paper.

Then, allegedly, someone looks at a series of marbles in a row, plinks one into it to start a series of impacts, and sees the end one moves off at a comparable speed as the one they threw into it. Through some series of conceptual moves, this is linked right back to the maths.

Going from the first stage - the mathematical theorem - to the second isn't just a case of quantifier restriction, it's actually changing the type of entity considered. A similar alchemy of types occurs in the transition from the second to the third; you go from mathematically suggestive generic descriptions of things to considering concrete, manipulable, particulars. The first lot is just about maths, the second one is some weird mix of math and reality, and the last one's about some marbles in a row that shunt about.

If you want to construe that as an operation of circumscription, it doesn't seem to conserve the types of entities considered. It's not, I guess one way of putting it, a metaphysically inert operation; you change what the statement is considered to apply to between stages. Another way of putting it, the "for all" in the first bit refers only to mathematical objects, the "for all" in the second bit seems to refer to some weird halfway stage between concrete particulars and mathematical abstractions, the "applied instance" of the marbles seems just to consider concrete particulars.

Nevertheless, it seems true that the first is suggestive of the second and the second is suggestive of the third. You know your logic, so I'm sure you can see that they're not following by modus ponens. They seem a lot more artful, more similar to transcendental deductions or interpretive links than strict logical entailments.

You know your logic, so I'm sure you can see that they're not following by modus ponens. They seem a lot more artful, more similar to transcendental deductions or interpretive links than strict logical entailments. — fdrake

Oh, yes. Feyerabend would have torn a new arse hole in such an argument. It has the pretence of rationality but on analysis, fails.

That's pretty much the case for any supposed scientific method.

At root, science identifies and integrates sensory evidence (which is the nature of reason). Science is essentially based, not on experiment, but on observation and logic; the act of looking under a rock or into a telescope is a scientific act. So is the act of observing and thinking about your own mental processes (proof of one's conclusions to others comes later, but that is argumentative, not inquisitive.) Science is willing to accept and integrate information from any observational source, without concern about persuading other people. — Harry Hindu

This errs in failing to notice that science is social. One individual making their own observations is not science. A group actively engaging in a conversation aimed at explaining what they see, and willing to adjust their view to account for what others claim, is at least a start.

There's quite a body of discussion on falsifiability. SOme familiarity with that would be helpful. — Banno

You should have noticed, from what I've posted, that I'm not at all interested in the conventional interpretation of "falsifiability". I believe it tends to be way off the mark. So I really don't know why you would make this suggestion to me. If you're content to sink into the quicksand of that interpretation, then so be it.

No worries. If you are not sufficiently interested in the article to read it, let alone look at the background material then I'll go back to ignoring your posts.

So, are mathematical axioms concerning infinity not level 4 statements? Are they verifiable or falsifiable?

This errs in failing to notice that science is social. One individual making their own observations is not science. A group actively engaging in a conversation aimed at explaining what they see, and willing to adjust their view to account for what others claim, is at least a start. — Banno

So Galileo wasn't doing science when he devised the modern scientific method and performed his experiments in private, away from the watchful eyes of the theocracy?

If philosophy is social, then we have both made the same point - that philosophy IS a science.

Besides, other peoples claims are are not evidence. That is where you err. You still have to personally verify their claims. Other peoples' claims is just another personal observation anyway.

As I explained, these two are contradictory. Unfalsifiable means impossible to falsify, which implies necessarily true, therefore proven. — Metaphysician Undercover

I don't think it implies necessary truth. For example, the claim that there is some particular configuration of stars and planets beyond the edge of the observable universe. That's unfalsifiable, because we can never check it out, no matter how close to the speed of light we accelerate a probe. But it's certainly not necessarily true.

Watkins seems aware of the problem of spelling out this sense of conceptual suggestiveness he's leaning on.

In "Between Analytic and Synthetic" (available for free on jstor with a public account), he sets out a hierarchy of concepts; at the top are things like haunted universe doctrines, at the bottom are observation statements. An observation statement is a report that something has been found to be the case by observation. Like "This substance dissolved in my sample of sulphuric acid". Their distinguishing property is that they are verifiable and falsifiable by observation alone.

At the next level of the hierarchy, he places instantial hypotheses. Instantial hypotheses predicate what they concern using only observable predicates. So "Every metal would dissolve in my sample of sulphuric acid" is instantial because you can check if a particular metal would dissolve in my sample. He calls an instantial hypothesis empirical if the conjunction of an instantial hypothesis and an observation statement entails another observation statement. So "Every metal dissolves in my sample of sulphuric acid" and "This is metal" gives you through modus ponens "This is would dissolve in my sample of sulphuric acid". Presumably "instantial" derives from taking an instance of the universal quantifier at the beginning of the statement, "all" goes to "this".

At the top level of the hierarchy, he places non-instantial (systems of) hypotheses. Non-instantial hypotheses are characterised by the property that if you take a conjunction of them with a collection of instantial hypotheses, they give rise to (his words!) observation statements, while themselves being neither falsifiable nor verifiable. A statement like "Every metal dissolves in some type of acid" is non-instantial - for any given metal it can't be checked without trying all the acids.

Observation statements are verifiable and falsifiable. Instantial hypotheses are falsifiable but not verifiable. Non-instantial hypotheses are neither falsifiable nor verifiable.

Notice however that "Every metal dissolved in some type of acid" actually contradicts the falsifiable statement "Gold is insoluable". Falsifiable because if you found a solvent for gold (eg aqua regia), it could be shown to be false by contradicting an observation (statement).

The paper tries to flesh out the regulatory role in a predictably falsificationalist way; haunted universe doctrines when taken in conjunction with observation statements stand in contradiction to some other observation statements. They are thus empirically meaningful.

However, it's got exactly the same hole as the other paper, and he knows it. The paper concludes:

Finally, since influential haunted universe doctrines are neither demonstrable nor testable, it becomes urgent (my italics) to investigate the ways in which they may be rationally supported or criticised

.

The extent to which "rationally criticising and supporting" haunted universe doctrines saturates our investigations will be the extent to which the active "gives rise to" Watkins uses in characterising the regulatory role of haunted universe doctrines can be fleshed out by the passive constraints haunted universe doctrines place on the logical space of scientific hypotheses. Inspiring the investigation of a hypothesis is much different from constraining what hypotheses can be investigated.

So Galileo wasn't doing science when he devised the modern scientific method and performed his experiments in private, away from the watchful eyes of the theocracy? — Harry Hindu

...as if he never published.

Another article I read long ago. I gather you are enjoying this?

Between Analytic and Synthetic. The title is curious. Quine's two dogmas was then a recent publication. It was also an attack on logical positivism. Watkins wants to highlight propositions that are neither synthetic nor analytic - hence the title; Quine wants to reject the very distinction between analytic and synthetic. So we have Watkins arguing for a spectrum of analyticity, while Quine argues that analyticity depends on a circular definition.

"Every metal would dissolve in my sample of Sulphuric acid" - but here is a metal that does not dissolve! "Ah, that looks like metal, but it isn't. It can't be because every real metal will dissolve in my acid..." Watkins takes little heed of the meanings of the words used; for Quine, the meaning is central, and yet open to such ambiguity. Feyerabend sees science as much more dependent on the whim of the scientist; the ad hoc hypothesis as a part of the scientific method.

So, are mathematical axioms concerning infinity not level 4 statements? — Metaphysician Undercover

Give an example.

Good paper, thanks. Definitely Popperian. So there is a metaphysical core (or several) to any scientific theory.

I don't think it implies necessary truth. For example, the claim that there is some particular configuration of stars and planets beyond the edge of the observable universe. That's unfalsifiable, because we can never check it out, no matter how close to the speed of light we accelerate a probe. But it's certainly not necessarily true. — bert1

I don't agree that such a claim is unfalsifiable. Just because we do not have the means to falsify it right now does not mean that we will not develop the means.

Give an example. — Banno

How about the obvious one then, the axiom of infinity? How would it be empirically verified? How could it be falsified?

I don't agree that such a claim is unfalsifiable. — Metaphysician Undercover

All that means is that you are misusing the term "unfalsifiable".

...the axiom of infinity? How would it be empirically verified? How could it be falsified? — Metaphysician Undercover

It can't be, nor has anyone claimed that it can be.

All that means is that you are misusing the term "unfalsifiable". — Banno

Uh huh, so you say. But you're not interested in any interpretation which is not consistent with yours, so it really doesn't say very much.

It can't be, nor has anyone claimed that it can be. — Banno

Do you agree then, that these sorts of mathematical axioms, which are fundamental to the mathematics which is commonly used in scientific endeavours, are what the op refers to as Level 4 statements?

No.