Structural realism and the nature of structure

Imprimirby Scientia Salon

Our new pick for a “notable” paper is Jonas R. Becker Arenhart and Otavio Bueno’s “Structural realism and the nature of structure,” published in the European Journal for Philosophy of Science, 5:111-139, 2015. (Full text here, free.)

Here is the abstract:

Ontic Structural Realism is a version of realism about science according to which by positing the existence of structures, understood as basic components of reality, one can resolve central difficulties faced by standard versions of scien- tific realism. Structures are invoked to respond to two important challenges: one posed by the pessimist meta-induction and the other by the underdetermination of metaphysics by physics, which arises in non-relativistic quantum mechanics. We argue that difficulties in the proper understanding of what a structure is undermines the realist component of the view. Given the difficulties, either realism should be dropped or additional metaphysical components not fully endorsed by science should be incorporated.

The paper is a very thorough, well argued critique of the now very fashionable “ontic structural realism,” as proposed, for instance, by James Ladyman and Don Ross in their Every Thing Must Go. The context for this discussion is the long standing debate in philosophy of science between realists and anti-realists, about both scientific theories and the entities (e.g., electrons) that they posit.

Here are some choice bits from the paper:

The ontic structural realist advances a metaphysical thesis to the effect that structures and relations are the fundamental components of the world; objects are secondary— they should either be eliminated or at best re-conceptualized in structural terms.

Our aim in this paper is to show that it is unclear that a proper characterization of structure suitable for ontic structural realism can be offered. We argue that there are far too many distinct ways of characterizing structure and relations, and as a result, the combination of realism and a metaphysics of structures becomes, at best, prob- lematic and, at worst, incoherent.

On the one hand, ontic structural realists argue that theories are better characterized in accordance with the semantic approach, rather than in terms of the syntactic view of theories and related approaches to structure based on Ramsey sentences.

On the other hand, the semantic approach is typically formulated in terms of set-theoretic structures. But this commitment to set theory, we argue, introduces objects as key components in the characterization of structures, and is responsible for the tension.

This strategy is called the “Poincare` manoeuvre” by Steven French (2012, p. 23). According to it, objects are used merely as heuristic devices or stepping stones to obtain the structure. After the structure is characterized, the objects are left behind: either they are taken as metaphysically irrelevant entities or are only conceived as being derived from the relations, depending on the kind of OSR that is assumed.

This maneuver, however, faces significant difficulties. First, in set theory, struc- tures are obtained as elements of the set-theoretic hierarchy. As noted, on the set-theoretic account of structure, objects are used to construct relations and struc- tures, not vice versa.

The indispensability argument aims to establish commitment to objects that are indispensable to our best theories of the world (for discussion and references, see Colyvan 2001). It was originally designed by W. V. Quine (see, e.g., 1960) to force those who are realist about scientific theories to become realist about the mathematics that is indispensably used in such theories.

We understand the “indispensability thesis” as the claim that scientific theories cannot be formulated without reference to mathematical objects, relations and functions. We understand the “inseparability thesis” as the claim that it is not possible to separate the nominalistic content and the mathematical content of a scientific theory. The indispensability thesis may entail the inseparabil- ity thesis, but not the other way around.

The deflationary nominalist grants that mathematics is indeed indispensable to science, but resists the conclusion that this provides any reason to be committed to the existence of mathematical objects and structures. This is achieved by distinguishing quantifier commitment (the mere quantification over the objects of a given domain, independently of their existence) and ontological commitment (the quantification that commits one ontologically to the existence of something).

The problem with the introduction of ontologically neutral quantifiers in the con- text of structural realism is that, given these quantifiers, it is unclear how structural realists will manage to specify what their realism amounts to. Unless they provide an independent mechanism of access to, and specification of, the structures they are realist about, the use of ontologically neutral quantifiers will ultimately remove all ontological content from structural realism.

Mathematical structures only represent the nominalistic (physical) content, which is the content structural realists are ultimately committed to; they need not be committed to the mathematical content. In other words, the set theory that structural realists invoke only play a representational role; it does not provide any guide to the commitments structural realists have.

In response to the point that OSR does not entail mathematical platonism, the sit- uation is more complex than it may initially appear. On the surface, it may seem that the two views are independent from one another. After all, OSR is a form of realism about the (fundamental) structure of reality. As such, it seems to make no claim about the existence of mathematical structures—which is the scope of a structuralist version of platonism (that is, a form of realism about mathematical structures). But, in fact, if the mathematical content of a theory cannot be separated from its physical (nominalistic) content (Azzouni 2011), it is unclear how the struc- tural realist can restrict ontological commitment only to the physical content without having first already nominalized mathematics.

A structure is characterized (in a loose sense) by both objects and relations, but for the structural realist only relations are primary ontologically. This is a good indication that relations are the fundamental components of the world, and indeed ontic structural realists emphasize this point (see, in particular, French 2010). But this means that in order to understand the nature of structures, we need to understand the nature of relations and of the connections they bear to objects. Metaphysically speaking, relations are far from being uncontroversial. They are at least as controversial as properties. To speak of relations as primary components of reality, one cannot speak of them as being somehow abstracted from objects— since, in this case, they would be ontologically dependent on objects. Rather, in order to have ontological primacy, relations need to constitute such objects.

Given that structure gives rise to objects (which are read off from the relations), how can one make sense of the disparate objects that emerge in distinct theories that share part of an underlying structure? Since some part of the structure is the same in the old and in the new theories, at least one of two options should obtain: (i) some features of the resulting objects should be the same in distinct theories, that is, there is also a form of objectual continuity through theory change, or (ii) since some structural preservation should be maintained throughout, this induces some continuity at the level of objects too, since these objects are characterized in terms of the relevant structures. However, both options entail a form of objectual continuity through theory change, something the structural realist has banned, given the pessimist meta-induction.

By positing some essential structure that gets accumulated, structural realists end up admitting that in the long run (even if it is supposed to be a very long run), as scientific theories get closer to the truth, the objects will get progressively closer to being fixed by the accumulated relations, and so realism about objects will be justified too.

If the structural realist does not allow for some fixed, essential structure to be preserved through scientific revolutions—allowing for modifications even in the parts considered essential—then there is no reason to suppose that in the long run, after many instances of theory change, any structure will be ultimately preserved.

Given the considerations above, ontic structural realists are unable to specify the nature of the structure they are supposed to be realist about. There is underdetermination both at the mathematical and the metaphysical levels.

If OSR is the best combination of realism and structuralism in philosophy of science that is also able to make sense of quantum physics, perhaps the realist component needs to be dropped. The very idea that there is a true, fundamental, underlying structure of the world—in whose existence we must believe—is difficult to make sense of, as the above arguments have indicated.

One needs to acknowledge that the truth or plausibility of the proposed metaphysics will not be settled on purely scientific grounds. By giving up on a strict naturalistic methodology in the metaphysics of science, one can introduce discussions about theoretical virtues in metaphysics, and then invoke those virtues to claim that OSR fares better than the alternatives, at least on prag- matic grounds. However, if a naturalistic metaphysics must go, then we must abandon the idea that OSR is a metaphysics tailored to fit our physics, and without this most cherished motivation, OSR is leveled with other metaphysical packages, disputing priority on a priori grounds.

60 thoughts on “Structural realism and the nature of structure

  1. I doubt anyone else on this site is going to attempt to defend OSR, so I will.

    I count myself an OSRist, but not a typical one since I view OSR as a weaker form of the Mathematical Universe Hypothesis (MUH) which I also subscribe to. In the MUH, the universe is a mathematical object and is no different from any other mathematical object: all mathematical objects are equally real. In the OSR, the universe is a structure and there are no commitments about whether other structures may exist. I take the “structure” referred to by the OSR to be a mathematical object and so the two hypotheses are compatible.

    Firstly, I must confess that I find the technical language of the extract to be quite difficult so I may have misunderstood a few points.

    > On the other hand, the semantic approach is typically formulated in terms of set-theoretic structures

    This is one of the points on which I am lost. By the “semantic approach” as opposed to “syntactic approach” I would think that OSRists want to emphasise that what is important about structures is isomorphisms rather than representations. That is: if two structures, however represented, can be shown to map exactly onto each other by some sort of translation, then those structures are the same. I don’t know enough about this to comment on whether this requires a commitment to set-theoretic structures as the article alleges.

    > As noted, on the set-theoretic account of structure, objects are used to construct relations and struc- tures, not vice versa.

    The OSR view is that those objects have no properties whatsoever apart from their involvement in relations. The objects are therefore meaningless unless viewed in the context of their relations.

    Take for example a structure made by connecting matchsticks end to end. One could make a cube, or a triangle, or a tetrahedron, etc. In this analogy, the matchsticks are relations and the objects are vertices where the matchsticks meet. One needs to talk about the vertices in order to define how the matchsticks connect, but take away the matchsticks and there are no vertices to speak of. The vertices (objects) can be said to exist but only in the context of a structure built of matchsticks (relations). Of course, unlike matchsticks, relations can also only be said to exist in the context of a relational structure.

    So I think OSRists say structure is made of relations only to emphasise that objects have no properties but for their relations, that these objects are nothing but the nodes where relations intersect. One still needs both objects and relations to define structures, but it is the relational interconnections and these alone that form our reality and this is the OSRist point. The objects/nodes form a crucial part of the structural web just as the vertices do in a matchstick-structure. As such, perhaps it is wrong to think about either objects or relations as primary, one needs both at the same time or can be no structure.

    Like

  2. > it is unclear how structural realists will manage to specify what their realism amounts to.

    First, I am actually a Platonist, so I personally am committed to the existence of mathematical objects. OSRists in general need not be Platonist because they are only committed to the existence of the underlying mathematical structure of our universe (which is yet to be discovered). They are not committed to the realism of other mathematical objects, including the models our physics has come up with to explain reality, models which are only approximate and not final.

    Since I am a Platonist, perhaps there is little point in me defending OSRists against accusations of Platonism. Perhaps the paper has a point, although I confess I have trouble following the argument. It doesn’t seem to me to be a problem for OSRists to say they are realists about the relational structure of physical reality while regarding mathematics as but a tool we can use to describe it.

    > Since some part of the structure is the same in the old and in the new theories…

    The article makes the point that since there is some structural similarity between theories, there must be some continuity of objects. But since the structure has changed, and only bears a limited resemblance to the original structure, and since objects are defined only by their position in structures, it is not strictly true to say that there is object continuity. Rather each object in the old theory may have an analogue in the new, with quite similar properties (as defined by their relations).

    Let’s return to the matchstick analogy. Say my original model of reality is a tetrahedron, and my new model of reality is a square pyramid. All the vertices in the original model have an analogue in the new, but I’ve removed an old matchstick, inserted a new vertex and added three new matchsticks. You can call this object (vertex) continuity if you wish, but that’s more a matter of taste than of fact. I think the OSRist preference for denying object continuity is clearer.

    > and so realism about objects will be justified too.

    Fine, but in this case the objects you are realist about are nothing more than pretty abstract nodes in a relational structure — realism about vertices in a matchstick structure. The conception of an object as a physical thing with some kind of meaningful existence independent of its relations is what OSRists dispute.

    > there is no reason to suppose that in the long run, after many instances of theory change, any structure will be ultimately preserved.

    The OSRist is committed only to there being a fundamental theory of physical reality, a Theory of Everything or a Grand Unified Theory. He/she is not committed to any current scientific model of reality being accurate in any respect. Therefore there is no reason to suppose that the OSRist is committed to the preservation of anything as scientific theories change.

    Like

  3. Like DM, I also found this text a bit hard to follow, but I guess this is probably because it is a collage of various passages cut out from the full article. There are various statements in the text which could draw some criticisms regarding various meaning of existence, relationship between physics and metaphysics, mathematics and physics, set theory and mathematics, syntax and semantics, etc. The authors seem to be unaware of category theory (the math discipline), which was invented precisely for the purpose of studying objects and relations, in the most abstract way. And so on.

    However, despite all that, the conclusions at the end are much more transparent:

    Given the considerations above, ontic structural realists are unable to specify the nature of the structure they are supposed to be realist about. There is underdetermination both at the mathematical and the metaphysical levels.

    If OSR is the best combination of realism and structuralism in philosophy of science that is also able to make sense of quantum physics, perhaps the realist component needs to be dropped. The very idea that there is a true, fundamental, underlying structure of the world—in whose existence we must believe—is difficult to make sense of

    As a physicist, I fully agree with this. Metaphysics is (and will always remain) underdetermined by science. But I also don’t think that this represents any breaking news to anyone.

    Second, when speaking what to drop in the OSR, the authors suggest “R”, which I find somewhat confusing. In my opinion, it is much more sensible to drop “O”, and talk about structural realism as an epistemological point of view, not ontological.

    On a general note, I must admit it is puzzling to me why people (philosophers?) put so much emphasis on ontology, at least as metaphysics is concerned. Why so much ado about what “is really out there”? The history of science, IMO, clearly points out that as our knowledge about the world improves, various ontologies swing back and forth like paper in the wind. I can only guess that this behaviour will continue into the future, making any potential metaphysical stance a moving target, and any discussion of ontology a red herring. Having a certain metaphysical point of view is a double-edged sword — it can be used as a good inspirational tool which points to new science, but it can also be detrimental if taken with dogmatic acceptance.

    The best strategy is to be open-minded about changing ones ontology inside-out several times over, if this helps the epistemological advancement, i.e. if it helps doing good science. By rigid subscribing to any metaphysical point of view, one can only get so far…

    DM,

    I doubt anyone else on this site is going to attempt to defend OSR, so I will.

    Good luck… 🙂

    Like

  4. I fear that we are trying to say things that we can never quite say.

    There is always this map/territory ambiguity. On the one hand everybody accepts and says that the model is not the thing that it is modelling and that we should not confuse the two.

    But on the other hand it is not easy to refer to anything in physics without talking as though the model was the reality.

    For example Sean Carroll says of GR : “Spacetime is a curved pseudo-Riemannian
    manifold with a metric of signature (−+++).”, but I doubt that he really thinks that reality is a set of generalisations of numbers with a rule to define how each relates to others in the set.

    Physicists talk of the lowest level of reality in terms of fields which are also groups of generalisations of numbers. Again physicists speak as though the model was the reality, not because they believe it is the case, just that it is almost impossible not to without becoming cumbersome and prolix.

    What I am saying, basically is that the limits of our understanding and the limits of our language are getting in the way of our attempt to describe reality.

    The model may not be reality, but the best model available is still the most accurate and precise way we can talk about the lowest level of physical reality. Once we try to talk beyond the model we are not saying more about reality, we are just speaking less precisely and less accurately about reality.

    I think that ontology and any kind of metaphysics of reality will ultimately fail and the “shut up and do the maths” approach will be as good as any. I think the Ladyman and Ross approach will fail, not because of any particular weakness of their approach, but due to the futility of the whole project of ontology.

    But that does not mean to abandon realism – that is just to accept that we will be limited in the ways we can refer to it.

    So I would suggest that rejecting any or all ontologies will not necessarily involve rejecting realism.

    Like

  5. OSR is, in my view, mistaken. Here is one of my examples: Suppose you have the source code for a program (structure) for a worm. Call the program Worm. Now suppose Worm can be compiled to (at least) two output types: (S) a simulation consisting of code running in a standard computer; (A) an assembly of materials like what is found naturally in worms (i.e., the output of what is called a biomolecular compiler).

    Now all Worm→(S) can do is basically light up pixels and show a cartoon of a worm, whereas Worm→(A) can go out and live in the world interacting with chemicals, just like a natural worm.

    The OSR mistake is that a simulation is not an assembly, and that substrate (the output medium of the compiler) matters.

    Like

  6. To speak of relations as primary components of reality, one cannot speak of them as being somehow abstracted from objects— since, in this case, they would be ontologically dependent on objects. Rather, in order to have ontological primacy, relations need to constitute such objects.

    Hm, I thought that the point was that the relations do “constitute” such objects. I like “process” a lot better than “structure” as a metaphysical primitive because structure has the cognitive baggage of stasis, but the idea is basically the same. If you’re translating (simplistically) between object/property/relation talk and process talk, you’d say that a property of an object is really just a way that the object transforms something else — and that if there is no transformation then there is no property. Having disposed of properties in this way, you can also dispose of objects proper, because the transformations exhaust what the “object” does in relation to other “objects”, and the word “object” itself becomes just a syntactic placeholder that allows you to express a transformation in a system that is not uniform. “Objects”, in other words, arise out of the non-uniformity of transformations. That’s why we can think of a human body as an object and a liver within that body as an object too, with respect to whatever process we’re talking about.

    All that is conceptually difficult, no doubt. But metaphysics is going to be like that because we are going to need to think about things in a way that our brains, built to conceptualize medium-sized things that knock into each other, are not cognitively suited to.

    The semantic v. syntactic thing is really interesting. I’ll have to dive into the paper to see what that’s about.

    I suspect that the “If you’re Ladyman, you must also be Tegmark” idea might be another result of cognitive limitations. Processes only have content by virtue of regularity, and mathematics is the language we use to describe regularity. Maybe Tegmark would have been better off going that route. An awful lot of people seem to be confused by the equation of “describable regularity” and “real mathematic structure”.

    Like

  7. One needs to acknowledge that the truth or plausibility of the proposed metaphysics will not be settled on purely scientific grounds.

    This is a really good point. I think pragmatic grounds are all we’re ever going to get with metaphysical theories. And I think that’s a feature rather than a bug.

    Liked by 1 person

  8. My comment has only to do with the web page design, not the content.

    When the page is converted to pdf, the color of both the abstract and the article text is too light. Contrast is insufficient, making the thing very hard to read.

    Please consider using a darker color. Thanks.

    Like

  9. Hi Marko,

    > Metaphysics is … underdetermined by science.

    I agree too. I don’t like the attempts by some such as Tegmark to call this kind of stuff science. But I still think it’s worthwhile.

    > In my opinion, it is much more sensible to drop “O”

    Fair enough. I actually think that ontology is all in how you look at things. There is no fact of the matter about what does and does not exist, up to and including the universe itself.

    Hi Robin,

    > On the one hand everybody accepts and says that the model is not the thing that it is modelling

    This is *usually* the case but I think it is wrong to say that it is *always* the case.

    Say you have a black box function. You can see the input and the output but you don’t know exactly what the function is doing. You may by experiment extrapolate models of what it is doing and get better and better at predicting what output the function will give for any input.

    In the special case where the model you have produced is precisely the same as what the function is doing, then the map really is the territory, because two mathematical functions that do the same thing are the same function.

    If the OSRist is right that the world actually is a mathematical structure, then the map (the physics in our physics textbooks) is the territory only in the special case that we have got it exactly right. To assert that the map cannot be the model is just to reject that it makes any sense to call the universe a mathematical structure. That may be your intention, but don’t make the mistake of thinking that you have an argument to back it up in the form of the familiar “the map is not the territory” refrain.

    > I think that ontology … will ultimately fail … due to the futility of the whole project of ontology.

    That’s fair enough.

    Hi Philip,

    > The OSR mistake is that a simulation is not an assembly

    Your example does not address OSR, because the two worms are different structures. One worm is made of biological matter, which is one kind of structure, while the other is made of silicon, which is another kind of structure.

    But anyway, even against simulation, this whole argument, akin to Searle’s virtual fire not being hot and Massimo’s virtual photosynthesis not producing sugar, doesn’t really work. For everything a real worm can do, a virtual worm can do an analog. The physical differences we perceive can be explained by our perspective being situated in the same physical context as one worm but not the other. From a virtual perspective within the simulation, the virtual worm appears more substantial and the physical worm appears abstract.

    Hi Asher,

    Nice to see you commenting again!

    You have Ladyman-Tegmark backwards :). If you are a Tegmark you must also be a Ladyman. MUH is making bolder claims than OSR, and the MUH entails OSR.

    Liked by 1 person

  10. Hmm, is this essay really directed at Mainstreet?
    I am going to try to make sense of the subject so for now my only contribution is to make the full paper available:
    http://bit.ly/1QqC8Pr
    And here is another paper that tries to explain Ontic Structure Realism:
    What is ontic structure realism?http://bit.ly/1dQfstX

    To answer my own question, I suspect that the problem is that my end of main street lies in the seedy docklands. The ivory tower is not only invisible to my neighbourhood, they think it has something to do with illegal trafficking in elephant tusks. If they did by chance catch sight of the ivory tower they would probably use it for target practice.

    Liked by 3 people

  11. Disagreeable Me,

    Even if my Worm program was compiled to memristors (the latest “soft hardware” technology), I don’t think that it would be “wormy” enough compared to its being compiled to biomolecular assemblies.

    Maybe I’m too down-to-earth in my position. 🙂

    Like

  12. I wish I could read the entire thing, as I’m a huge fan of structural realism in general (as the SEP notes: “Structural realism is considered by many realists and antirealists alike as the most defensible form of scientific realism”).

    I agree with several other points some of the other commenters have made. Marko mentions category theory, which could definitely be of interest to someone trying to formulate a structural approach in metaphysics. Edward Zalta (one of my favorite contemporary philosophers), also discusses a structural approach in his recent article “Foundations for Mathematical Structuralism,” with an abstract that reads:

    “We investigate the form of mathematical structuralism that acknowledges the existence of structures and their distinctive structural elements. This form of structuralism has been subject to criticisms recently, and our view is that the problems raised are resolved by proper, mathematics-free theoretical foundations. Starting with an axiomatic theory of abstract objects, we identify a mathematical structure as an abstract object encoding the truths of a mathematical theory. From such foundations, we derive consequences that address the main questions and issues that have arisen. Namely, elements of different structures are different. A structure and its elements ontologically depend on each other. There are no haecceities and each element of a structure must be discernible within the theory. These consequences are not developed piecemeal but rather follow from our definitions of basic structuralist concepts.”

    Obviously things get pretty heavy early on, so this stuff is not for the faint of heart, but you can find the PDF’s for Zalta’s writings (including the above article) on his personal homepage at Stanford.

    Asher Kay brings up another great point when he discusses how pragmatic grounds might need to enter into any discussion of metaphysics (reminds me of my final post on the Skepticism article from last week). We may very well not be able to utilize physics completely when it comes to probing the actual nature of reality, with the biggest factor preventing that being the absurdly huge energies we would need to muster to probe the deepest domain of reality (think solar system size accelerators, energies equivalent to the mass-energy of Jupiter, etc.). With that in mind, looking at where a structural take on reality has brought us up to this point is one pragmatic consideration we may need to employ when it comes to figuring out the true nature of existence.

    I recognize that there’s plenty of work to be done on the OSR front, but I strongly believe that its a very promising avenue philosophers (and everyone else who’s interested) can take. Hopefully we continue to see progress in the arena of structural realism, OSR, and similar ilk.

    Like

  13. labnut,

    “Hmm, is this essay really directed at Mainstreet?”

    Clearly not, since this is a paper from the primary literature. The idea of our “notables” feature is to occasionally present the abstracts and some bits from technical paper in science or philosophy, which we think are of more general interest. I always alert the authors, who hopefully will chime in during the discussion. I wish I had the time to actually write a layperson summary of these papers, but I don’t, and obviously not all authors I contact are willing to do that with their papers. But I still think it’s good for more people to have at least an idea of what goes on in academic journals. Besides, I keep being pleasantly surprised at the level of discussion following these posts, evidently people are learning something and enjoying the experience. And thanks for the link to the full paper, I have now updated it in the main post as well!

    DM,

    “I take the “structure” referred to by the OSR to be a mathematical object and so the two hypotheses are compatible”

    While I do think OSR and MUH are compatible, it isn’t event clear whether the structures referred to in the first one are mathematical, or what that means, or how that would then result in such structures somehow underpinning the existence of physical objects — which of course is the reason for this paper to begin with.

    “I must confess that I find the technical language of the extract to be quite difficult so I may have misunderstood a few points”

    As I said, sorry, see my comment above to Labnut.

    “I don’t know enough about this to comment on whether this requires a commitment to set-theoretic structures as the article alleges.”

    Require is too strong a word, though it seems like OSR’s themselves prefer such interpretation. In bits that I did not cite, however, the authors go on to raise substantive issues even in the case of non-standard set theory.

    “The objects are therefore meaningless unless viewed in the context of their relations.”

    I think it’s more than that: it’s that the objects have no ontological status except as embedded in relations.

    “In this analogy, the matchsticks are relations and the objects are vertices where the matchsticks meet”

    I like it, except of course for the obvious disanalogy: you have used physical structures to represent the relations, and non-physical objects to represent the physical ones…

    “One still needs both objects and relations to define structures, but it is the relational interconnections and these alone that form our reality”

    I sense a contradiction there: how can we need objects and yet do away with them as the ontological basis for reality?

    “perhaps it is wrong to think about either objects or relations as primary, one needs both at the same time or can be no structure”

    But that’s clearly not what OSR says, as indicated in Ladyman & Ross’ title: every *thing* must go, there are only relations, without relata.

    “OSRists in general need not be Platonist because they are only committed to the existence of the underlying mathematical structure of our universe”

    True, but as Ladyman & Ross admit, the step to Platonism is really, really short.

    “Perhaps the paper has a point, although I confess I have trouble following the argument. It doesn’t seem to me to be a problem for OSRists to say they are realists about the relational structure of physical reality”

    Well, what the authors are getting at, I think, is that there doesn’t seem to be a sensible way of defending that sort of realism (about structures) without falling into full fledged Platonism, which they reject.

    “The article makes the point that since there is some structural similarity between theories, there must be some continuity of objects”

    Not exactly. In that portion the authors are arguing that OSR is subject to the objection that even structures, not just objects, are sometimes not continuous between theories. Which would mean that being realist about those structures is a bad bet. But if that’s abandoned, there is no more structural *realism*.

    “The OSRist is committed only to there being a fundamental theory of physical reality”

    No, she is committed to say that certain structures described by current scientific theories are real. Otherwise it isn’t realism as it is commonly understood in philosophy of science.

    Marko,

    “Like DM, I also found this text a bit hard to follow, but I guess this is probably because it is a collage of various passages cut out from the full article”

    Indeed, see above. Basically, these “notables” entries are just abstracts plus selected passages that I highlighted when I was reading the paper (and within reasonable copyrights limits!).

    “The authors seem to be unaware of category theory”

    No, they do mention it and discuss it (e.g., pp. 124-127 of the paper), but I didn’t present those bits because they are even more difficult to follow than the rest of the paper.

    “Metaphysics is (and will always remain) underdetermined by science. But I also don’t think that this represents any breaking news to anyone”

    You’d be surprised. Both Lawrence Krauss and Neil DeGrasse Tyson would disagree. Stephen Hawking too…

    “the authors suggest “R”, which I find somewhat confusing. In my opinion, it is much more sensible to drop “O”, and talk about structural realism as an epistemological point of view, not ontological.”

    Both are viable options, from the point of view of philosophy of science, though I can see why the second one is much more palatable to a physicist.

    “Why so much ado about what “is really out there”? The history of science, IMO, clearly points out that as our knowledge about the world improves, various ontologies swing back and forth like paper in the wind”

    Not quite. They change, but I don’t get the impression that they are that variable. Besides, don’t we want to know “what’s really out there”?

    “I can only guess that this behaviour will continue into the future, making any potential metaphysical stance a moving target”

    But by the same token, why espouse any current scientific theory, since it will also change and likely be abandoned?

    “The best strategy is to be open-minded about changing ones ontology inside-out several times over, if this helps the epistemological advancement”

    Agreed.

    Robin,

    “For example Sean Carroll says of GR : “Spacetime is a curved pseudo-Riemannian
    manifold with a metric of signature (−+++).”, but I doubt that he really thinks that reality is a set of generalisations of numbers with a rule to define how each relates to others in the set.”

    Indeed. I find that to be often the case with the way Sean writes.

    “What I am saying, basically is that the limits of our understanding and the limits of our language are getting in the way of our attempt to describe reality.”

    How Wittgensteinian of you!

    “I think that ontology and any kind of metaphysics of reality will ultimately fail and the “shut up and do the maths” approach will be as good as any.”

    There I’m going to part ways with you: I think we want to have the best understanding, not just pragmatic use, that we can derive from scientific theorizing (and metaphysical speculating).

    Philip,

    “The OSR mistake is that a simulation is not an assembly, and that substrate (the output medium of the compiler) matters.”

    As you know, I wholeheartedly agree…

    Asher,

    “I thought that the point was that the relations do “constitute” such objects.”

    Right, but one can’t just say that, one needs an account of what that means. And that’s what the authors think OSR is not providing.

    “Having disposed of properties in this way, you can also dispose of objects proper, because the transformations exhaust what the “object” does in relation to other “objects”, and the word “object” itself becomes just a syntactic placeholder”

    Tell that to my desk (an object), on which I comfortably set my laptop (another object).

    “I suspect that the “If you’re Ladyman, you must also be Tegmark” idea might be another result of cognitive limitations”

    I also think it’s not true, since I’m pretty sure Ladyman rejects Tegmark’s ideas.

    Liked by 1 person

  14. Massimo,
    thank you for the nice summary of the comments so far. I would like to add some further comments:

    “In the MUH, the universe is a mathematical object and is no different from any other mathematical object: all mathematical objects are equally real. In the OSR, the universe is a structure and there are no commitments about whether other structures may exist. I take the “structure” referred to by the OSR to be a mathematical object and so the two hypotheses are compatible.”

    First, if one is going to be realist about mathematical objects, one must specify which mathematics one is talking about and why stick to that one. There are many options: classical mathematics, intuitionistic math and other forms of constructive math, mathematics developed in distinct extensions of set theory, those developed in non-Cantorian set theory, and so on. There are many challenges for Platonism in math that we may suppose friends of OSR are not willing to face. As commented, there is a big issue on whether OSRists will be able to distinguish mathematical and physical structures. Given their inclination to look for structures in current empirical science, they are expected to do so. Ladyman once proposed that a structure is physical in case one can somehow distinguish the objects it talks about, while mathematical structures allow also for completely indiscernible objects. I don’t know how far he pursued this suggestion though… other options are also available to make the distinction, mainly involving the modal nature of physical structures, but this is still a controversial issue.

    “The objects are therefore meaningless unless viewed in the context of their relations.”

    As Massimo said, objects only exist inside a structure, and that is why OSR is mostly viewed as an eliminative position: objects are eliminated from the basic furniture of the world and have, let us say, no ontological dignity. However, the comment applies very well to another version of ontic structural realism, called “moderate”. According to this version, objects exist independently of structures, but all of their features depend of the structure they are in. Objects are somehow bare objects, completely dependent on structures (the naure of this dependence must, of course, be spelled in detail). The paper is mostly directed at the eliminativist version of OSR, although some arguments apply to both versions.

    “Perhaps the paper has a point, although I confess I have trouble following the argument. It doesn’t seem to me to be a problem for OSRists to say they are realists about the relational structure of physical reality”

    The paper tries to put some pression on an important issue for the realist. Given that one is claiming to be realist about structures, then one should be able to specify which structures one is being realist about and what exactly they are. Some of the arguments are directed to show that it isn’t clear that OSR has the resources to specify the kind of commitment required from a realist. So, the situation we have is that one is realist about structures but is not able to specify which structures one is realist about; there are just too many candidates and no way to choose that could be acceptable for a structural realist.

    “Metaphysics is (and will always remain) underdetermined by science. But I also don’t think that this represents any breaking news to anyone”

    Besides those already quoted by Massimo, some OSRists believe that OSR is the best answer answer to metaphysical underdetermination. In fact, they seem to believe that OSR entitles them to be commited only to the common structure of the many distinct metaphysical interpretations of the theory (roughly speaking), so that one is entitled to be realist about those aspects of reality of which we can have some scientific knowledge about. Notice that metaphysical underdetermination appears when one is trying to “read off” metaphysics from science, a reading which may be performed in many distinct ways. As the paper tries to establish, OSR is also prey to metaphysical underdetermination, far from solving it.

    “the authors suggest “R”, which I find somewhat confusing. In my opinion, it is much more sensible to drop “O”, and talk about structural realism as an epistemological point of view, not ontological.”

    In fact, both options are left more or less open in the end of the paper. What is stressed is that it is really difficult to conciliate O with R. There are many difficulties to keep both together. Epistemological structural realism is really an option, but it faces its own difficulties.

    “Why so much ado about what “is really out there”? The history of science, IMO, clearly points out that as our knowledge about the world improves, various ontologies swing back and forth like paper in the wind”

    Well, if one is being realist about science one must estipulate what is this realism about. OSR tries to make it clear that “as our knowledge about the world improves”, we have some cumulative knowledge of reality. That is, there is structural retention as theories change (otherwise it would be hard to claim that our knowledge improves, to begin with). So, it is their task to make it clear what structures are! There is really an issue in discovering what is really out there and how it looks like.

    Liked by 1 person

  15. Massimo,

    “The authors seem to be unaware of category theory”

    No, they do mention it and discuss it (e.g., pp. 124-127 of the paper), but I didn’t present those bits because they are even more difficult to follow than the rest of the paper.

    Oh, I see. After reading it, I have a couple of follow-up comments and questions, but I’ll raise those only if authors decide to join the discussion. 🙂

    “Metaphysics is (and will always remain) underdetermined by science. But I also don’t think that this represents any breaking news to anyone”

    You’d be surprised. Both Lawrence Krauss and Neil DeGrasse Tyson would disagree. Stephen Hawking too…

    Well, then I guess I am living in my own little personal Ivory Tower… 🙂 On a more serious note, IMO, this speaks (volumes) about Krauss, DeGrasse Tyson and Hawking, more than on the “breaking news” themselves. Remember 🙂 — us physicists are notorious for not swaying to authority figures, Hawking or otherwise.

    But I have to ask you — did you quote those three names because they are physicists, or because they are popularizers of physics? As popularizers, they seem to do a fairly equally good/bad job. As physicists, they are worlds apart from each other. So I’m a bit confused — how did you actually come up with them as examples?

    Besides, don’t we want to know “what’s really out there”?

    Well, to be honest — no. I thought that such a question was long gone, defeated by things like the Matrix, brains-in-vats, solipsism, Boltzmann brains, vivid dreams of a butterfly, etc… I’d say that these “alternatives” render any attempt at a conclusive answer completely hopeless. Instead, we should focus on the epistemological question — how is the “out-there” best described by math (or some other suitable language)? That is something one can actually hope to answer (up to Goedel issues etc.). And fundamental physics is all about making models, precisely with that latter question in mind.

    “I can only guess that this behaviour will continue into the future, making any potential metaphysical stance a moving target”

    But by the same token, why espouse any current scientific theory, since it will also change and likely be abandoned?

    Only students are bedazzled by current scientific theories (I know I was…). But an experienced scientist will instead be acutely aware of the shortcomings of those theories, and will never dare to seriously espouse any of them.

    The difference between a physical model and a metaphysical position is that the former, despite being incorrect, can still be useful in a limited, approximate sense. For example, Newtonian mechanics is grossly wrong on a number of accounts, but for the purpose of making simple everyday machines it is useful nevertheless. But metaphysics is qualitatively different — it does not have this practical approximate usefulness. It is always about being conceptually correct or incorrect. I am not aware of any cases where a known-to-be-wrong metaphysics can be considered suitable for any purpose whatsoever (other than as an example of being wrong). Then again, I could be misinformed…

    Liked by 1 person

  16. Massimo wrote:

    she is committed to say that certain structures described by current scientific theories are real. Otherwise it isn’t realism as it is commonly understood in philosophy of science.

    ———————————————————

    Massimo, how would you describe the difference between scientific and metaphysical realism? I have some thoughts forming on this article, but they depend, in part, on how this distinction is cashed out.

    Like

  17. Marko,

    ” I’ll raise those only if authors decide to join the discussion”
    We are waiting for your comments then! 🙂

    ” I thought that such a question was long gone, defeated by things like the Matrix, brains-in-vats, solipsism, Boltzmann brains, vivid dreams of a butterfly, etc”

    That is a very curious thing. Some naturalist philosophers in general dismiss this kind of philosophical dispute (brains in a vat and sceptical scenarios in general) as being too much first-philosophy. They shift to a science-based (or something like that) philosophy and in general adopt either a realist or an anti-realist position. In any case, by sticking closer to science they end up more concerned with “what is really out there” (that is what the debate is about). That is, your denial to give importance to this issue of reality is based on a philosophy of a kind many naturalists believe hopeless, so that science is seen as an alternative (and that is the curious point).

    “But metaphysics is qualitatively different — it does not have this practical approximate usefulness. It is always about being conceptually correct or incorrect. I am not aware of any cases where a known-to-be-wrong metaphysics can be considered suitable for any purpose whatsoever.”

    Well, perhaps practical utility is not something to look for when judging metaphysical theories!

    Liked by 1 person

  18. Labnut said:

    “The ivory tower is not only invisible to my neighbourhood, they think it has something to do with illegal trafficking in elephant tusks. If they did by chance catch sight of the ivory tower they would probably use it for target practice.”

    Hilarious! And a nice break in this serious subject.

    Coel said:

    “The difference between a physical model and a metaphysical position is that the former, despite being incorrect, can still be useful in a limited, approximate sense. For example, Newtonian mechanics is grossly wrong on a number of accounts, but for the purpose of making simple everyday machines it is useful nevertheless. But metaphysics is qualitatively different — it does not have this practical approximate usefulness. It is always about being conceptually correct or incorrect. I am not aware of any cases where a known-to-be-wrong metaphysics can be considered suitable for any purpose whatsoever (other than as an example of being wrong). Then again, I could be misinformed…”

    How about the original metaphysics called religion? Clearly not all religions can be true (from the practioners point of view) but are certainly considered suitable for many purposes including understanding the nature of the universe. Some of the greatest scientists like Newton (alchemy, biblical prophecy) and Kepler (astrology) and even philosophers like Spinoza (pantheism) considered their metaphysics to be useful – at least giving psychological support which then fedback into their science. All of these were “grossly wrong on a number of accounts” to many of us. I can imagine Sir Isaac sitting alone at night by candlelight and deciding what he would work on next: calculus or the book of Revelation.

    I would suggest that metaphysics, while masquerading as a system that “is always about being conceptually correct or incorrect”, is in fact as much about the psychological as the philosophical, as much about feeling as thinking. Descartes’ cogito could then be, “amo, ergo sum”, I love, therefore I am. Feeling goes as deep as thinking. With respect to metaphysics, when I make a judgement and say “it seems to me”, isn’t my subjective preference at least partly how it feels to me?

    Liked by 2 people

  19. Jonas,

    Welcome, and thanks for joining the discussion! 🙂

    However, the comment applies very well to another version of ontic structural realism, called “moderate”. According to this version, objects exist independently of structures, but all of their features depend of the structure they are in. Objects are somehow bare objects, completely dependent on structures

    This is actually connected to what I wanted to ask about category theory. In CT, objects of course cannot be fully eliminated (as you argued in the paper), but they become precisely those “bare” objects, devoid of any nature except the one provided by the structure of the category. In this sense, your discussion of set theory versus category theory strikes me as incomplete — in contrast to ST, which requires a very explicit specification of objects, in CT the purpose of objects is reduced to mere placeholders for “anything that fits” the structure. Objects cannot be completely eliminated from CT, but there is a clear improvement over ST, in my opinion. Was this point just amiss, or deliberately omitted as off-topic for the paper? Or did I miss to read the relevant paragraph? 🙂

    Epistemological structural realism is really an option, but it faces its own difficulties.

    Do you have in mind any particular difficulties?

    Well, if one is being realist about science one must estipulate what is this realism about. OSR tries to make it clear that “as our knowledge about the world improves”, we have some cumulative knowledge of reality. That is, there is structural retention as theories change

    As a physicist, I don’t think that there is such a clear-cut relationship between “knowledge of reality” and “structure of reality”. One can arguably define “knowledge” as a set of experimental data we have collected so far. This is cumulative by its very nature, but there is no structural retention to speak of, as it accumulates. The data itself contains no structure at all — it is a job of a theory (a mathematical model) to “guess” the structure and reproduce the experimental data from that structure. But this process is wildly model-dependent, and the same knowledge (i.e. data) can be reinterpreted in terms of vastly different structures (i.e. predicted by vastly different theories).

    The fact that a new theory usually contains the old theory as its approximation is not a manifestation of some necessary structural retention. It is merely a convenience, one that establishes that the new model correctly predicts all the data that the old model did, and then some. A typical example is the inverse-square law of gravity from the points of view of Newtonian mechanics versus general relativity — same law (i.e. data), completely different ontological structures describing it.

    My point is that structure is theory-dependent, while knowledge about reality (the data) isn’t. In that sense, “reality” does ontologically exist, stipulated by data, but its structure exists only epistemologically, stipulated by our theories.

    Or am I missing something? 🙂

    Liked by 1 person

  20. Tell that to my desk (an object), on which I comfortably set my laptop (another object).

    I guess what I’d say to your desk is that everyday talk is a lot different than theoretical metaphysical talk.

    Of course there are objects in the realm of everyday language. And it’s not hard, given our sensory apparatus, to understand why we tend to conceptualize in terms of objects. But that doesn’t mean that everyday objects need to have some sort of fundamental status in a metaphysical theory — any more than amino acids need to have a fundamental place in a quantum mechanical theory.

    I hope your desk and I understand one another. At the very least, it’s unlikely to make a glib reply ;).

    Liked by 1 person

  21. Hi Philip,

    Even if my Worm program was compiled to memristors (the latest “soft hardware” technology), I don’t think that it would be “wormy” enough compared to its being compiled to biomolecular assemblies.

    But what is ‘wormy’? From our point of view they are soft squishy things but that is a perception of them.

    Suppose you gained a sharper perception and greater cognitive power to deal with the new information. Now you can view an entire worm cell by cell.

    Your cognitive power and perception increases more, you see the whole worm as an assembly of molecules. Then by atoms, then by subatomic particles.

    Finally you are viewing an entire worm as a collection of quarks.

    Now you are seeing the worm as an assembly of little packets of properties like charge, mass and spin, none of which really make sense except in relation the the structure around it. Indeed at this level it may not even be so easy to distinguish between the worm and it’s surroundings.

    Now, at this level – is there the categorical distinction that you are describing?

    What is the distinction here between the simulation and the real?

    Like

  22. Despite the complexities of the technical language and the theories they manifest, I believe I get the general drift of the paper. It appears to support a position I’ve been reaching recently (although I was long favorable to it previously.)

    When I first read of Structural Realism here, I confess I was attracted to it, as a middle ground between Nominalism (and the relativism that leads to) and either metaphysical or epistemological Realism (and the ungrounded assumptions these sometimes depend on). But my Buddhist and Pragmatist commitments have won out, which leads me to accept Nominalism, with all its risks, rather than adopt any kind of Realism, however attractive, and the weak foundations it likely rests upon. And this essay seems to confirm my doubts about Realism.

    Basically, although any realism can be used to build a coherent picture of the world, its foundations will prove to be incoherent on analysis. This actually has to do with the problem of induction – we only experience the world of our senses in particulars, universals are generated in the mind.

    What exists in the mind has real existence – in our minds – and frequently function as useful correlates in modelling and negotiating the world around us.

    But the fact remains that we do not experience any ‘treeness’ in a tree – we experience this specific thing in front of us, with all its differing qualities, many of which we can measure, some of which are strictly (as Aristotle would say) accidental. This thing that splintered during a strike of … what we call lightning…. and then twisted and regenerated in order to survive – it has qualities we can measure, but the measurements we take of the twisting never evidence the effort to survive.

    The incoherence of any Realism is that it presumes that there is some universal treeness in the thing before us – but it has no way of demonstrating this. When it tries to do so, it lapses into conceptualizations that it tries to deny. So there’s ‘treeness’ and then there’s this thing in front of us, and then there’s the concept ‘tree’ which can be defined and studied by classifying similarities between individual things of like characteristics. But none of this ever successfully links into a metaphysical truth about the next similar individual entity we experience. The best we can say of it is that it is similar.

    Maybe that’s enough; indeed, maybe that’s all we have.

    What Arenhart and Bueno are arguing is that OSR quavers between committing itself to the existence of objects it denies, and a Nominalist structuralism it doesn’t want. They suggest that the first position ought to be abandoned, and the second accepted. I believe they are on the right track. However, I disagree with the additional implication that Naturalism is weakened by this. It’s not surprising that the notion of the “pragmatic” recurs again and again in the text. Pragmatism isn’t a final solution, but it is functionally useful position to take until a satisfying solution comes along.

    Liked by 1 person

  23. >Ontic Structural Realism is a version of realism about science according to which by positing the existence of structures, understood as basic components of reality, one can resolve central difficulties faced by standard versions of scien- tific realism.

    From this first sentence I see that the paper seeks to help justify scientific realism. Thus before getting into the technicals of the article itself I must ask, “What is my initial perception of scientific realism?” Wikipedia gives me the impression that this realism concerns both how well science might potentially occur, and how well it is indeed occurring. Furthermore I’m told that it may be broken up to mean, for example, “I actually feel such realism for physics much more than I do for psychology.” I’ll now present my initial ontological view of scientific realism — perhaps OSR will change it somewhat once I do give this article a good read:

    The human effectively takes what it thinks it knows (evidence), and checks to see if this is consistent with various associated models of reality. In fact today we have a scientific community which has used this process to develop a vast array of accepted understandings, and apparently in recent years these understandings have permitted us to become quite powerful. But contra scientific realism, shouldn’t we always remain relative idiots when compared against the true complexities of reality itself? Are we not, after all, just a minor aspect of it? For this reason I currently claim scientific realism only to an associated relative degree.

    Is Ontic Structural Realism, or “positing the existence of structures, understood as basic components of reality…” going to make me more scientific realist? Actually I am already far more optimistic than most regarding fields such as philosophy and our mental/behavioral sciences. But might “positing the existence of structures” suggest the potential for a relative idiot (like the human I think), to glean something more than relatively good understandings? Yes I am willing to open my currently pessimistic mind.

    Like

  24. Hi all,

    I and others may be guilty of equivocation on the meaning of “object”.

    1) Everyday things such as tables and chairs as well as molecules, atoms, quarks and so on.

    2) Nodes in relational structures. These are much more abstract, having no intrinsic properties and existing only insofar as they tie the relations together (just as the relations exist only insofar as they tie the nodes together).

    Massimo,

    > whether the structures referred to in the first one are mathematical

    To me, all pure structure is mathematical.

    > you have used physical structures to represent the relations, and non-physical objects to represent the physical ones

    If the objects I’m talking about are mere nodes in a relational structure, then they are no more physical than the relations. That’s what I was trying to illustrate by comparing them to vertices in a matchstick-structure.

    > how can we need objects and yet do away with them as the ontological basis for reality?

    In light of the distinction I drew above, we need nodes to build relational structures. We don’t need physical, substantial objects with their own properties and ontology independent of relational structures.

    > there are only relations, without relata.

    I interpret that as meaning that the relata are devoid of content and exist only with respect to the relations, being essentially abstract objects such as the vertices in a matchstick structure. That has to be what they mean, because if you didn’t have relata of any kind then there could be no structure at all, as you cannot have unbounded relations.

    > Which would mean that being realist about those structures is a bad bet.

    I would be surprised if OSRists are committed to realism about any specific structures (e.g. the electron). I think there is a mathematical structure isomorphic to reality and am realist about *that*, whatever it may be. I am realist about the electron only insofar as I suspect it forms part of the underlying structure of reality. I assumed this was the OSRist position, because whatever structures we think we know about now could fall apart in the next round of experimentation at the LHC.

    Hi Jonas,

    > one must specify which mathematics one is talking about

    On plenitudinous Platonism, all consistent systems of mathematics exist. I believe in all abstract objects that can be precisely defined, irrespective of the axioms used to define them. Some structures can be defined in multiple systems, e.g. the Mandelbrot set does not require one to take a position on the axiom of choice.

    > one should be able to specify which structures one is being realist about and what exactly they are.

    As I said above, I think there is a mathematical structure which is isomorphic to reality. I don’t know what that structure is. I don’t think this is so different to most physicists who believe there is some final theory of everything we are asymptotically approaching with successive breakthroughs. One can believe that there is such a theory without knowing what it is, exactly.

    But whatever the structure is, I am a realist about that and I think that structure is not only isomorphic but identical to our universe.

    Like

  25. Marko: “Metaphysics is (and will always remain) underdetermined by science. But I also don’t think that this represents any breaking news to anyone”

    Massimo: “You’d be surprised. Both Lawrence Krauss and Neil DeGrasse Tyson would disagree. Stephen Hawking too…”

    I think you’re misunderstanding their position. They would agree (I hope) that theory is under-determined by data. That’s pretty standard, and is why science adopts Occam’s razor. Rather, they would dismiss metaphysical positions that aren’t supported by evidence. They have a point. You say:

    … don’t we want to know “what’s really out there”?

    We indeed do. But the only way we can find out what’s really out there is by evidence. In other words, we can only find out what’s really out there to the extent that it is determined by science.

    My own view is that there is something “real” about reality, in the sense that there is something “causal” about reality, where I identify being “real” with being capable of causation (sorry DM).

    But, the best we can do is to develop models about that reality. Those models (mathematical and physical concepts) are not in themselves “reality”, but are concepts about reality.

    The issue of what is ultimately causal (and thus what is the basic ontic reality) is then to be discerned as best we can from those scientific models of reality. But, given that we know that we don’t have “final” versions of fundamental physics, and given that science is inevitably provisional, I don’t think we can currently get very far in deciding what is the most basic ontic “unit”.

    Our current understanding of quantum mechanics starts with quantum fields (mathematical constructs about variation in time and space), but we also know that our theory of quantum mechanics is inconsistent with our theory of time and space. Therefore we know that we don’t understand what are the basic ontic “units” of fundamental physics.

    I don’t think we can currently say anything more, though the path forward is finding better theories of fundamental physics. I think that is pretty much what Marko said also.

    Can metaphysics get to an understanding of that fundamental ontology where science cannot? Well, that would be rather a first.

    Like

  26. I got Ladyman and Ross’s book after hearing the rationally speaking exercise, and concluded that neither my philosophy or my physics were up to making a worthwhile judgement – echoes of recent discussions about the relationship between expert and non-expert. In fact, I mentioned OSR to a former physics professor in a discussion group I belong to, and got equally blank looks and shrugs.

    I must say that I have some sympathy for DM’s point; in fact, I’d be hard put to it to explain why there aren’t strong elements of Platonism and perhaps idealism in OSR. I certainly feel in need of some help; something between the level of Ladyman’s explanation on Rationally Speaking and that of their book and this paper. Not much to ask, surely ;-). If any feels inclined to help me out…

    This reminds me of the time I was asked to fill in teaching some basic physics to some 12-14 years old children with learning delay, mainly through deafness. I could manage so far as the syllabus went, but worried about my lack of knowledge in depth. For instance, I could deal with forms of energy, but children are inclined to ask things like, “But what is energy, really”. I was a little reassured to find Richard Feynman saying that nobody knows that!

    Liked by 1 person

  27. Marko Vojinovic: “Metaphysics is (and will always remain) underdetermined by science.”

    But as a mainly theoretical programmer (not a physicist), I wonder:
    Programming languages have been developed for domains that may be completely “artificial” (continua, hyperbolic spaces, unbounded nondeterminism, Turing jumps, … *) and for which the “hardware” may never exist or be made. Is this doing science?

    * http://ucnc15.wordpress.fos.auckland.ac.nz/

    Like

  28. Jonas,

    thanks for chiming in during the discussion, your comments are very helpful!

    pete,

    “I wish I could read the entire thing, as I’m a huge fan of structural realism in general”

    See the link provided by Labnut above.

    Marko,

    “Remember 🙂 — us physicists are notorious for not swaying to authority figures, Hawking or otherwise”

    Glad to hear it… 😉

    “did you quote those three names because they are physicists, or because they are popularizers of physics? As popularizers, they seem to do a fairly equally good/bad job. As physicists, they are worlds apart from each other. So I’m a bit confused — how did you actually come up with them as examples?”

    I agree with your assessment. The names came up because those people have made public comments about philosophy and metaphysics. As you say, it isn’t at all clear how much those comments reflect any consensus within the physics community.

    “to be honest — no. I thought that such a question was long gone, defeated by things like the Matrix, brains-in-vats, solipsism, Boltzmann brains, vivid dreams of a butterfly, etc… I’d say that these “alternatives” render any attempt at a conclusive answer completely hopeless”

    Ah, I see. I consider all that stuff only as useful reminders to adopt a position of epistemic humility, not as a signal that we should stop searching.

    “an experienced scientist will instead be acutely aware of the shortcomings of those theories, and will never dare to seriously espouse any of them”

    Hmm, again, the very same people mentioned above seem to be serious exceptions. And so is, for instance, Brian Greene about string theory and so forth.

    “I am not aware of any cases where a known-to-be-wrong metaphysics can be considered suitable for any purpose whatsoever”

    Yes, I take your point. I keep being univalent about the whole metaphysical project anyway (see my recent essay here at SciSal: https://goo.gl/wk3os6).

    Aravis,

    “how would you describe the difference between scientific and metaphysical realism?”

    Not sure what you mean here: realism is a metaphysical position, and indeed antirealists in philosophy of science typically defend their take by saying that it is metaphysically more conservative.

    Asher,

    “I guess what I’d say to your desk is that everyday talk is a lot different than theoretical metaphysical talk.”

    Indeed, but you yourself used everyday examples (human body, liver). In general I think that regardless of the level of analysis one’s metaphysics needs to be consistent, or one is in trouble.

    “hat doesn’t mean that everyday objects need to have some sort of fundamental status in a metaphysical theory — any more than amino acids need to have a fundamental place in a quantum mechanical theory”

    Interesting analogy, but I don’t buy it. Quantum mechanical entities are fundamental to amino acids, not the other way around. But OSR says that there is *nothing* (physical) fundamental to everything else. That’s hard to swallow and requires detailed justification, which OSR supporters have, so far, failed to provide.

    ej,

    “When I first read of Structural Realism here, I confess I was attracted to it, as a middle ground between Nominalism (and the relativism that leads to) and either metaphysical or epistemological Realism”

    Funny, that was my own early reaction to OSR!

    “But my Buddhist and Pragmatist commitments have won out, which leads me to accept Nominalism, with all its risks, rather than adopt any kind of Realism”

    Ah ah! My Stoic commitments have me lean the opposite way, toward realism! (Though to be fair, in this they are greatly helped by my other professional side, as a scientist.)

    “the fact remains that we do not experience any ‘treeness’ in a tree – we experience this specific thing in front of us, with all its differing qualities, many of which we can measure, some of which are strictly (as Aristotle would say) accidental”

    True enough, but I don’t see why any of that doesn’t license the rather obvious inference that there is indeed a tree, the causal nexus of all those qualities.

    “The incoherence of any Realism is that it presumes that there is some universal treeness in the thing before us – but it has no way of demonstrating this”

    Not sure why you say that. Realism doesn’t imply universalism.

    “none of this ever successfully links into a metaphysical truth about the next similar individual entity we experience. The best we can say of it is that it is similar.”

    Good enough for me, and for science.

    DM,

    “To me, all pure structure is mathematical.”

    Yeah, I know. And I still can’t make any sense of that position.

    “If the objects I’m talking about are mere nodes in a relational structure, then they are no more physical than the relations.”

    See above.

    “We don’t need physical, substantial objects with their own properties and ontology independent of relational structures.”

    And yet all of science seems to tell us that the world is made of physical objects…

    “I interpret that as meaning that the relata are devoid of content and exist only with respect to the relations”

    But the claim is deeper: that at bottom there are only relations. Relations between what?

    “I would be surprised if OSRists are committed to realism about any specific structures (e.g. the electron)”

    Electrons are objects, not structures. And yes, OSR is committed to realism about structures, as it says right in the name.

    “because whatever structures we think we know about now could fall apart in the next round of experimentation at the LHC.”

    Precisely the authors’ point, which I think is damning for OSR.

    Coel,

    “Rather, they would dismiss metaphysical positions that aren’t supported by evidence.”

    You mean like the multiverse, or string theory? 😉

    “But the only way we can find out what’s really out there is by evidence.”

    Suitably combined with theorizing. We don’t just read reality off data.

    “I identify being “real” with being capable of causation”

    Agreed.

    “Those models (mathematical and physical concepts) are not in themselves “reality”, but are concepts about reality.”

    Again, agreed, though I’m increasingly less sure of why this is germane to the discussion at hand.

    “given that we know that we don’t have “final” versions of fundamental physics, and given that science is inevitably provisional, I don’t think we can currently get very far in deciding what is the most basic ontic “unit””

    Well, to be fair to Ladyman and Ross, they state right at the beginning of Every Thing Must Go that what they are trying to do is to glean metaphysics from the best available physics. They then compare all current alternative physical theories (e.g., string, loop quantum gravity, etc.) and make the bet that whatever they all have in common is most likely to survive the next round, and therefore to be a better candidate for ontic basicality.

    “Can metaphysics get to an understanding of that fundamental ontology where science cannot? Well, that would be rather a first.”

    Depends on what you mean by “metaphysics.” If you mean the sort of a priori analytical stuff I criticized in the essay linked to above, I think you are right. But that’s not what OSR’s are trying to do (that’s why L&R use terms like “scientific” or “naturalized” metaphysics). And of course I think that theoretical science engages in quite a bit of metaphysics, even without acknowledging it.

    gwarner,

    I hear you. I don’t know of any intermediate level treatment of OSR, but perhaps our authors do.

    Liked by 1 person

  29. Robin,

    I think the problem is in going from “collection of quarks” to ” packets of properties”. Quarks are elements of the (physical) substrate. Properties are elements — (identifier, number) pairs — in a language about quarks. The mistake is saying the substrate and the language are the same thing.

    Like

  30. Marko,
    thanks!

    “Objects cannot be completely eliminated from CT, but there is a clear improvement over ST, in my opinion. Was this point just amiss, or deliberately omitted as off-topic for the paper? Or did I miss to read the relevant paragraph?”

    As I mentioned previously, the main target of the paper was eliminativist OSR, so that we took this fact you are mentioning as bad news for this version of OSR (even if it is an improvement)!

    “Do you have in mind any particular difficulties?”

    Yes, Epistemic structural realism (ESR) is generally wedded to a syntactical approach to scientific theories, which many find troublesome for a lot of reasons. Also, ESR is not always clear as to whether one must be “agnostic” or “atheist” about the underlying objects: is it the case that their qualities are not known to us and we can’t know whther we are able to know them, or can we make sure that we don’t know their qualities. Of course, friends of ESR would disagree that the problems are really a trouble, but there are a lot of discussion about it and no agreement is in view.

    “My point is that structure is theory-dependent, while knowledge about reality (the data) isn’t. In that sense, “reality” does ontologically exist, stipulated by data, but its structure exists only epistemologically, stipulated by our theories.”

    I see your point, but I think structural realists would not agree with that! In fact, most philosophers would insist that data only make sense in the light of a theory.

    Philosopher Eric,
    “Is Ontic Structural Realism, or “positing the existence of structures, understood as basic components of reality…” going to make me more scientific realist? Actually I am already far more optimistic than most regarding fields such as philosophy and our mental/behavioral sciences. But might “positing the existence of structures” suggest the potential for a relative idiot (like the human I think), to glean something more than relatively good understandings? Yes I am willing to open my currently pessimistic mind.”

    In fact, OSR is trying to bridge the gap between epistemology and metaphysics: what we know (through our scientific theories) is what there is (structure). How well that works is something to be discussed, of course. But OSR believes that we may be better off eliminating objects than adopting, as ESR, objects whose natures cannot be known.

    Hi DM

    “On plenitudinous Platonism, all consistent systems of mathematics exist. I believe in all abstract objects that can be precisely defined, irrespective of the axioms used to define them”

    I think I see your reasons for that. However, besides other problems with Platonism, you will have to explain why inconsistent structures do not exist (like those inside paraconsistent set theories). Also, in this Plantonic heaven one should be able to see that the Continuum Hypothesis, for instance, is true or false. However, it is true in some set theories, and falso in others, and independent in others. How a Platonist takes that fact?

    Coel

    “Can metaphysics get to an understanding of that fundamental ontology where science cannot? Well, that would be rather a first.”

    As Massimo mentioned, the point of OSR is to get the structure of reality (its ontology, according to OSR) through science, by looking at our best scientific theories. The friends of OSR believe to be engaged in some kind of naturalized metaphysics, although Ladyman and French have different understanding on how that activity is supposed to be pursued.

    Gwarner,

    “I must say that I have some sympathy for DM’s point; in fact, I’d be hard put to it to explain why there aren’t strong elements of Platonism and perhaps idealism in OSR.”

    In fact, it is hard to see that. Some friends of OSR say that structure is not to be identified with its mathematical representation. So, it seems, it is a kind of Platonic object that may have distinct formal representations. Really, I just can’t see an easy answer here, and your worry seems, to me at least, to be justified.

    Like

  31. In general I think that regardless of the level of analysis one’s metaphysics needs to be consistent, or one is in trouble.

    Right. So if we dispose of objects as fundamental metaphysical entities, we’d want to account for how they arise after that and why they seem so fundamental to us.

    but you yourself used everyday examples (human body, liver)

    Metaphors/analogies/isomorphisms are ways of expanding the map conceptually, but eventually we set them aside for a rigorous formulation.

    I keep being reminded of your reaction to Smolin’s discussion of invented vs. discovered. Something about the way he framed the issue (expanding what was previously seen as a dichotomy) lit up a lightbulb for you. I expect it’s much the same with objects in metaphysics. Perhaps L&R understand what structure is but aren’t explaining it in a way that is lighting up the lightbulb for some people. The mind returns to its desks and laptops like Samuel Johnson to a rock.

    The table and the laptop are just physical processes that are relatively stable over time and exhibit relative causal containment.

    But OSR says that there is *nothing* (physical) fundamental to everything else

    This is the real issue, not your desk or laptop. I would say that what they’re calling relations (and what I call processes or Coel calls “capable of causation” or DM calls mathematical structure) are the basis of physicality. Does that make structure/relations/processes non-physical themselves? Proto-physical? Meta-physical? Does it really matter if we understand how the physicality of physical theories arise from structure/relations/processes?

    If we’re asking something like, “How can something physical arise from something non-physical?” in this context, we’re thinking about it wrong.

    I don’t agree with OSR, by the way. I think an object/relation formulation leads to exactly the sorts of confusions that the paper expresses wrt objects.

    Like

  32. Massimo; Thanks – On your (and DM’s) point that because “whatever structures we think we know about now could fall apart …. “which I think is damning for OSR.” I thought Ladyman and Ross tried to acknowledge but maybe defuse that point by making it clear that they were aiming at a defeasible metaphysics based on best current science, implying that they felt there could be enough flexibility to cope with at least some theory change?
    Having said that I don’t fully get the philosophy or physics, I’m going to explore what I think I get, in a somewhat roundabout way, hoping that this will help me, at least.

    I listened to Sean Carroll on the Rationally Speaking podcast and the Naturalism meeting videos. Sean says that he disagrees with OSR, because he doesn’t thinks that it is structure that is basic; it’s the single quantum wave function that constitutes everything that is. That is why he is a physicalist but not a materialist. Not sure how that distinction pans out in practice. I suspect it’s to do with the rainforest realism that James Ladyman discussed; this preference for the belief that what actual exists comes in a complex system of relationships, not the single “desert landscape of Quine and Sean Carroll? Especially, as Massimo discussed with Jim Holt at NECSS; according to some interpretations of quantum theory consciousness collapses the wave function and define the nature of reality. Now, at worst this seems in danger of serious woo-woo, at best, it shows why I was worried about idealism.

    Now, I liked what Sean Carroll said about how his single-wave function view dissolves issues about reductionism. If I understand him, he’s saying that the common mistake is to look to break reality down into smaller and smaller chunk. Whereas it’s really a matter of deciding which level of resolution we choose to examine and explain; fundamental physics, special sciences, social sciences, humanities and so on.

    I think Massimo will sympathise with that, and it ties into some criticisms by John Searle (no, not the Chinese Room again!) of what he sees as persistent philosophical mistakes in materialism that have confused the mind body- issue and made it seem harder than it may really be. Oversimplifying wildly and not arguing his; he says that materialism has been stuck on false oppositions and forms of reduction and a misunderstanding about the subjectivity of consciousness (it’s ontological, not epistemically subjective). But the end point is that consciousness really exists and isn’t the problem for naturalism that materialists have general thought it was. In effect, it just is the physical functioning of the brain (or any other causally sufficient entity) under a different aspect.

    And if we put that together with Sean Carroll’s view, we just might find even more aspects of reductionism becoming a non-problem. If I’m tentative about that, I’m even more so about how OSR’s rainforest diversity might fit in, or not.

    Like

  33. The aim of science is to find what reality is.
    In the last few centuries, what science found, and mathematics confirmed, in its practice, is that all theories are local.

    Instead the authors of the paper explicitly state that they are not interested by Category Theory, because they want to choose between it and Set Theory. They want to impose just a single foundation for the OSR.

    They also want to impose a structure on Categorical Objects (whereas the entire aim of CT is that objects have no structure).

    The authors seem unaware of the way category theory is used: small categories, n-categories, Grothendieck categories. Those categories are different, but used, as needed.

    Mathematics has become a theory of local theories, structured as needed for the work at hand.

    This philosophical revolution has been long in coming. It can be tracked back all the way to the Twelfth Century philosopher Abelard, and his work, “Sic Et Non”, the “Yes And No”. There, Abelard showed that, depending upon the point of view one took, the initial axioms one proceeded from, one would arrive at diametrically opposite conclusions in 158 questions of theology.

    Later Buridan showed and wrote that the geocentric and heliocentric system could not be distinguished.

    The old approach to science (inspired by the theology of monotheism) was to have just one explanation for everything.

    This is was thrown out by Descartes. Descartes made a tennis ball analogy to deduce the correct laws of light refraction (erroneously called Snell’s laws, in the Anglosphere, as Descartes elucidated them first). Descartes knew that this was just an analogy, a trick. Somehow, thought Descartes, light had enough of the properties of tennis balls, that the reasoning could work.

    Huygens later explained the change of direction of light upon entering a transparent medium of higher refractive index by hypothesizing that light was a wave, and that it went more slowly in such a medium. , just like a regiment of soldiers would turn, if, crossing a line one arrived on at an angle, those on the side would slow down.

    Huygens knew light was not made of soldiers. It was just a model, based on a guess. This general method of guessing models, and then using logic within the model, was brought ever more physical laws, and relations between them.

    By 1900 Poincare’ discovered the truth that time was local. Thus space was local (light is used to measure space by looking at the time it takes for it to go round trip).

    Poincare’ concluded that space-time was “relative”. Poincare’ also pointed out that the constancy of the speed of light, because it was always found to be, was a physical law. The “Principle of Relativity” ought to rest on that, what was found to be true, instead of a priori reasoning.

    However the paper at hand asserts explicitly, in its rejection of CT, that a priori reasoning ought to be imposed not just on reality and physics, but on mathematical practice itself.

    Liked by 1 person

  34. Interesting analogy, but I don’t buy it. Quantum mechanical entities are fundamental to amino acids, not the other way around

    Yeah — isn’t that exactly what they’re saying? Structure is fundamental to objects but not the other way around?

    Okay – I’m remembering now why commenting is unsatisfying. I think I’ll go back to just lurking :).

    Like

  35. Here on Main Street one wonders why the various schools of speculative realism and object-oriented ontology (Meillassoux, Harman, etc) about which one hears so much, don’t seem even to exist for SciSal. Another continental fad. Yet Ladyman is an important reference point in the work of this school. The point is to undermine the marriage of first philosophy and scientism and to foreground aesthetics and eco-ethics.

    Like

  36. Massimo,

    “‘The best we can say of it is that it is similar.’
    Good enough for me, and for science.”

    True; which I accept Pragmatistically.

    Does Realism necessitate universalism? The forms of Realism we have today are considerably more sophisticated than that debated by Medieval scholars. That debate largely hinged on the question of whether universals were naturally acquired (somehow embedded in the things themselves), or intellectual constructs for understanding (hence always artificial). Despite our greater sophistication, I still hear echoes of that debate in our contemporary discussions.

    Classical Realism committed itself to the principle Adequatio res et intellectus, that is, the ability of the essence of a thing to impress itself through our senses on our intellects which then needed adjusting to the thing for adequate knowledge of the thing. (The essence is thus what is known, and guarantees the universality of what can be said of the thing – the treeness of the tree that can be found in all trees.) The Nominalists replied that the particularity of existing things denied the possibility of finding any essence in them, that such ‘essence’ was purely a matter of intellection – a conceptual compounding of the signs from various individual instantiations.

    Once the early empiricists introduced the notion of the mind composing reality out of sense data and recurrent regularities, Classical Realism came to an end – as Peirce complained, most modern philosophers hold some form and some degree of Nominalism. (Peirce considered himself a Classical Realist, and thought that Adequatio could be salvaged, largely through semiotics; but he was wrong; the chain of signification only finds a resting place in action. Thus Pragmatism: we have adequate grasp on the thing when can do something effectively in response to it – turning the presumed tree into a chair, extracting its chemical components, writing poems about it, reforestation, etc.)

    Still, Realism continues to haunt us. First, as you’ve noted elsewhere, scientists and mathematicians often need to function as if assuming some form of Realism regarding objects of study.

    Second (and this I don’t think we’ve discussed much), coherent explanations of the world necessarily deploy Realist language; we have to discuss objects as though they exist even when we want to deny them any necessary existence. In higher order explanations such that professionals debate, this presents little problem; professionals are trained to understand the game in which the language is deployed.

    But we also have to teach interested high school and undergraduate students using Realistic language, just to provide them with a well grounded picture of the world. They then need to to find their own way to alternative explanations independent of such grounding. (Which explains the kinds of philosophical debates we find among precocious undergraduates.)

    I find this problematic. Our natural epistemic bias is toward Realism; this seems embedded in our language. Yet “tree” first signifies a concept – useful; malleable; shared with like minds.

    But I hold no metaphysical certainty about the thing that grows outside my window. Rather disappointingly, it doesn’t seem to care.

    Liked by 1 person

  37. I’m a total mainstreet amateur but my comments are:

    1. In regard to:

    “Metaphysics is (and will always remain) underdetermined by science. But I also don’t think that this represents any breaking news to anyone.”

    I would agree with this based on what little I’ve read about metaphysics so far. But, I think a different way of thinking could make a difference. Metaphysics is supposedly the study of being and existence. The universe “be”s and exists. Physics is the study of the universe. So, logically the laws of physics should be derivable from the principles of metaphysics. To me, this means that a metaphysician could use his/her knowledge of fundamental existent entities to build a model of the existent universe and eventually make testable predictions. This metaphysics-to-physics approach would make metaphysics more relevant to science and vice versa, and I think it would make faster progress towards a fundamental understanding of our universe than either metaphysics or physics alone. No one ever agrees with this, but to me, it seems to be straightforward reasoning.

    2. In regard to

    “But OSR says that there is *nothing* (physical) fundamental to everything else. That’s hard to swallow and requires detailed justification…”

    For me, it’s almost impossible to swallow.

    3. In regard to simulating a worm, this is actually being done, and it seems kind of neat!

    4. I sometimes think we spend too much time arguing about whether the fundamental unit of existence is an object, a mathematical structure, a relationship, etc. Whatever it’s called, it seems like there is some existent entity at the base. Maybe, instead of arguing over what it is, we could try to figure out the properties of a generic existent entity and use these properties to build a model of the existent universe, as mentioned in 1, above. We can argue forever without convincing each other. Maybe, it might be worth skipping over the exact name of this entity?

    Thanks for an interesting article and discussion and for listening to my points.

    Like

  38. n short summary, OSR is wrong because
    • simulation ≠ assembly
    • language ≠ substrate
    • blueprint ≠ building
    OSR confuses between the two sides in these pairs. What’s the alternative to OSR? I would say a physicalism in which the distinctions in the above pairs are recognized.

    in a just-published essay in Scientific American co-written by Victor Stenger (his last), “Stenger argues for the validity of philosophy in the context of modern theoretical physics.” In his view, many who communicate to the public promote platonic realism (“the belief that the objects within the models of theoretical physics constitute elements of reality”), whether they are aware of it are not. It takes into account the second inequality above.

    “Physicists are philosophers, too”
    http://www.scientificamerican.com/article/physicists-are-philosophers-too/

    Liked by 2 people

  39. Hi Massimo,

    Me: To me, all pure structure is mathematical.

    You: And I still can’t make any sense of that position.

    I don’t think that this position is particularly controversial. I think you misunderstood me. I’m talking about pure structure, i.e. form and not substance.

    > And yet all of science seems to tell us that the world is made of physical objects…

    Science doesn’t really have much at all to say on metaphysical questions like this.

    > Relations between what?

    Nothing of substance! Abstract vertices only.

    > Electrons are objects, not structures.

    They can be both. An atom is both an object and a structure. If the claim is that an electron is a primitive, elemental object with no structure, how do you know? They certainly do have various properties, such as spin, charge, mass and so on. If OSR is right, these properties arise out of relations in a structure.

    > OSR is committed to realism about structures, as it says right in the name.

    Yes, it is (I believe) realist about the structure of fundamental physics. It is (or ought to be) agnostic about what that structure actually is, because that has yet to be discovered by science (if it ever will be).

    Hi Jonas,

    > You will have to explain why inconsistent structures do not exist (like those inside paraconsistent set theories)

    Naively, we might have thought that the idea of parallel lines which diverge or converge is inconsistent. By redefining terms and axioms, however, it is possible to create hyperbolic or elliptical geometries where we can make sense of these ideas. As a plenitudinous Platonist, I think those geometries are just as valid.

    I think paraconsistent logic (and presumably set theory) is doing something similar. It takes what would be inconsistent in classical logic and forms a way of dealing with it coherently (and indeed consistently). As such, because we can be consistent in our dealing with these apparent contradictions, I regard these approaches as perfectly valid and consider any constructs that can be formulated within them to be as real as any in mathematics.

    An example of something I don’t regard as existing is something like a largest prime number or a square circle. Perhaps you could change the axioms to allow such things to be consistent, and that would be OK.

    If we consider a mathematical system which has contradictory axioms and we have no defined strategy for dealing with these contradictions, then I don’t think we can sensibly discuss any structures formed from them because the properties of the structures are not going to be fixed, depending on how we manipulate the axioms to reach different conclusions.

    > also, in this Plantonic heaven

    I personally really dislike this kind of imagery (similarly Platonic realm and so on)..

    > However, it is true in some set theories, and falso in others, and independent in others.

    By regarding the different set theories as distinct structures. It is true in some and not in others. Similarly, the “four corner hypothesis” is true in a square and the “five corner hypothesis” is true in a pentagon.

    Like

  40. Massimo

    There often seems to be a lack of support at the intermediate level about any complex area of expertise; there are lots of introductory materials and support structures for the beginner, and a thriving community of expert practitioners at the top end, working at a level that’s difficult to access unless you have already reached that level. I think the unhappy valley in between is usually bridged, if at all, by the learner being immersed in a community that will coach, scaffold and generally chivvy the learner over the chasm, working on some kind of apprenticeship model. For those who aren’t embedded in the relevant community, this can be a near insuperable chasm. And of course, the high end experts are too busy doing their stuff, and mentoring their own local group of upcoming professionals, to pay attention to the wider group of intermediate level mainstreet onlookers. Plus, I suppose, it’s a much smaller potential audience for books; there are a lot more absolute beginners out there. And in the world of “publish or perish”, efforts to help the intermediates don’t count.

    Aravis and Massimo

    Aravis wrote: “Massimo, how would you describe the difference between scientific and metaphysical realism? I have some thoughts forming on this article, but they depend, in part, on how this distinction is cashed out.”

    That’s something I’d very much like to hear the two of you discuss.

    Like

  41. Massimo wrote:

    Aravis,

    “how would you describe the difference between scientific and metaphysical realism?”

    Not sure what you mean here: realism is a metaphysical position, and indeed antirealists in philosophy of science typically defend their take by saying that it is metaphysically more conservative.

    —————————————————————————————————————-

    Yeah, I wasn’t very specific was I?

    So, what am I getting at? Realism, with respect to ontology? Realism, with respect to theoretical entities? Realism in the form of a certain view about the truth of scientific statements?

    What I’m wondering is the extent to which arguments against metaphysical realism are also arguments against scientific realism. While I read the article, I kept thinking about Goodman’s “Ways of Worldmaking,” Quine’s “Ontological Relativity,” and Davidson’s “On the Very Idea of a Conceptual Scheme.” But I’m not sure I understand the concept of “structure” that the author is using well enough to tell whether these essays and books are applicable to the question at hand.

    Like

  42. Massimo:
    “Besides, don’t we want to know “what’s really out there”?
    Marko:
    “Well, to be honest — no. I thought that such a question was long gone, defeated by things like the Matrix, brains-in-vats, solipsism, Boltzmann brains, vivid dreams of a butterfly, etc… I’d say that these “alternatives” render any attempt at a conclusive answer completely hopeless.”

    I agree that such questions lowers the certainty that we believe certain things about the external world. However it is not necessarily the case that we can not “know” certain things about the external world. It is unclear whether the possibility that things like vivid dreams, matrixes and us being a brain in a vat makes it impossible for us to know things that would depend on these possibilities being untrue.

    This deals with questions of how we define “knowledge” that we didn’t even delve into when we dealt with skepticism. (Lots of interesting questions are raised by that topic.) But suffice it to say that if knowledge is some form of justified true belief then it *may* be we know things about the external world even though we might not know, we know. That is, the justifications might be there even though we have reason to question whether they are there.

    “Instead, we should focus on the epistemological question — how is the “out-there” best described by math (or some other suitable language)?”

    But then aren’t you agreeing with Massimo?

    It seems to me that a scientist could punt on the whole reality question if he confined himself on how we can satisfy certain desires. I desire to live longer – this seems to work for that. I want a device that does this – this seems to work for that. I want a food additive that tastes like this – this seems to work. From that angle science doesn’t really care what reality is like it only matters that it seems to predicably yield results when certain conditions are met. It doesn’t matter if we are in a matrix or dreaming because, whatever reality it is us “thinking things” are acting in, doing certain things seems to yield predictable products that achieve our aims.

    But it seems science often has another goal. Not just getting what we want but understanding/describing what is out there.

    Liked by 1 person

  43. Aravis,

    ah, I see. So I think of scientific realism as a subset of metaphysical realism. It is about the ontological status of unobservable entities (e.g., electrons, quarks, strings) which play crucial roles in scientific theories. The realist says the science compels us to accept the ontological reality of such entities. The antirealist says we need not make any such commitment (it leads to “inflationary metaphysics,” to use van Fraassen’s terminology), and we can simply be agnostic about the ontology while accepting the empirical usefulness of postulating such entities “as if” they really existed. (Think of the usefulness of epicycles, before we discovered there is no such thing.)

    As for “structures,” I think the only way to make sense of the notion here is to think in terms of mathematical structures, hence the close link with Tegmark’s mathematical universe hypothesis and with mathematical Platonism. Part of the problem, as the authors point out, is that OSR supporters haven’t really been able to provide a good definition of structures, or at the least not one that does the heavy ontological work they want to get done.

    Liked by 3 people

  44. > It is about the ontological status of unobservable entities (e.g., electrons, quarks, strings)

    I understand from what you write that metaphysics has a “theory of observation” and can say what is observable and what is not. What is this theory and on what is it based? I’m curious. In particular because “structures” (whatever they may be) seem to be just as unobservable as electrons etc.

    Like

  45. Patrice Ayme

    “Instead the authors of the paper explicitly state that they are not interested by Category Theory, because they want to choose between it and Set Theory. They want to impose just a single foundation for the OSR.

    They also want to impose a structure on Categorical Objects (whereas the entire aim of CT is that objects have no structure). ”

    I don’t take the paper as making the confusions you point to. As mentioned in a previous comment, the whole poit of the argument concerning CT is that it is committed to objects (as you have noticed), being thus inadequate to represent the point of view of eliminativist OSR. That simple.

    Philip Trift

    “Physicists are philosophers, too”

    Thanks for the link! 🙂

    DM

    thanks for your comments, I think I can now figure out how your position accomodates the difficulties pointed to. Of course, it will still have to face some of the typical difficulties with Platonism, but that is another issue.

    “I think paraconsistent logic (and presumably set theory) is doing something similar. It takes what would be inconsistent in classical logic and forms a way of dealing with it coherently (and indeed consistently). ”

    Totally agree!

    Aravis,

    “But I’m not sure I understand the concept of “structure” that the author is using well enough to tell whether these essays and books are applicable to the question at hand.”

    Agree completely with Massimo here. Perhaps that is the great problem for OSR: the concepts of structure we can understand are unsuitable for the purposes of OSR, and the concept of structure suitable for the purposes of OSR are not understood!

    Aravis and Massimo (on realism X metaphysics)

    Massimo said:
    “So I think of scientific realism as a subset of metaphysical realism. It is about the ontological status of unobservable entities (e.g., electrons, quarks, strings) which play crucial roles in scientific theories.”

    Agree completely. However, it seems that sometimes a distinction is in order: when one accepts for instance, that electrons exists due to our best scientific theories, is one also committed with expliciting the “metaphysical nature” of the electron? For instance, must one also determine whether the electron is an individual object, a bundle of tropes or something else, and go on into such metaphysical subtleties? In a more general fashion, does the committment with a kind of unobservable entity requires that one also explicitate the metaphysical nature of such entity? It seems to me that the scientific realist does not need to enter into such considerations, which however, are relevant for the motivations of OSR (due to metaphysical underdetermination, as mentioned in the paper). What do you think?

    Liked by 1 person

  46. Jonas,

    Thanks for your responses! 🙂

    That is, your denial to give importance to this issue of reality is based on a philosophy of a kind many naturalists believe hopeless, so that science is seen as an alternative (and that is the curious point).

    I agree with that in general. My specific point was that doing science is an epistemological undertaking, while doing metaphysics is an ontological one. I proposed we keep doing the former and ditch musing over the latter. It is quite unlikely that we will ever guess the “correct” metaphysics, and all other “incorrect” attempts are, well, incorrect, and therefore useless. In contrast, science can be useful despite being ultimately incorrect, due to its epistemological nature on one hand, and “approximativeness” on the other (see my reply to Massimo).

    I see your point, but I think structural realists would not agree with that! In fact, most philosophers would insist that data only make sense in the light of a theory.

    I completely agree, that is also why I don’t subscribe to OSR. I also acknowledge the difficulties of ESR you pointed to. That said, *having* the data (the bare numbers) and *making sense* of the data (constructing scientific laws) are two different things. The former can be collected by brute observations, the latter requires a theory. That’s why computers can collect, remember, manipulate and display a bunch of numbers, but realizing that those numbers represent a JPEG photo of a cute little cat is beyond their capability.

    Occam’s beard,

    Coel said:

    “The difference between a physical model and a metaphysical position is that the former, despite being incorrect[…]

    Just a short note on attributions — you quoted me there, not Coel. 🙂

    Massimo,

    The names came up because those people have made public comments about philosophy and metaphysics.

    Well, there’s the rub — when a high-end physicist goes on to make a public lecture, nobody in the audience attempts to challenge what they say, so they can promote OSR uncritically. But if you actually engage them and make them defend their position, when push comes to shove most of them will actually backpedal away from OSR and admit that — although they would *like* OSR to be true — they cannot really seriously claim it is (Brian Greene included 🙂 ).

    Try it with some physicist, I’m curious how far they can go defending OSR against a serious philosopher. 🙂

    I consider all that stuff only as useful reminders to adopt a position of epistemic humility, not as a signal that we should stop searching.

    Fair enough, it’s your glass-half-full versus my glass-half-empty. 🙂

    Everyone,

    Metaphysics is (and will always remain) underdetermined by science. But I also don’t think that this represents any breaking news to anyone.

    I’m amazed by the number of times this was quoted. 🙂 Did I really say something so profound there, or controversial, or what is going on? Of course, I don’t mind, I’m just surprised…

    Like

  47. Jonas:
    Thanks for answering. You say:” As mentioned in a previous comment, the whole poit of the argument concerning CT is that it is committed to objects (as you have noticed), being thus inadequate to represent the point of view of eliminativist OSR. That simple.”

    Category Theory consists in the outstanding, and perplexing, meta-discovery that general theorems about very general structures were often the simplest way to prove particular theorems in particular mathematical structures (this effort was mostly led by Alexandre Grothendieck). I have studied Category Theory for many years. What is important in CT are the diagrams, and the theorems one can get from them. Any Category Theory is exclusively about structure.

    Category Theory is based upon the observation that mathematics is all about relations. “Objects” in Category Theory just anchor the relations (aka “morphisms”), they have no structure whatsoever. If one defines an “object” as something with a property, as philosophers do, the “objects” of Category Theory are the morphisms. Agreed, that’s confusing at first sight.

    (Some mathematicians were angry at Category Theory for decades, spiting it as “Abstract Nonsense”. But now various Category Theories have brought so many spectacular results, they are not controversial anymore.)

    Even the simplest traditional “objects” do not escape relations. Pointing at an object establishes a visual relation. The “properties” of an “object” are themselves sets of relations.

    All what exists is defined by relations. Existence, ontic or epistemic, is all about relations, in all ways.

    More related content will be published on my site.

    Liked by 1 person

  48. I’ve spent a good bit of time on Wikipedia and the SEP trying to make sense of the terms and ideas associated with this article. Some of the comments provided here do fortunately seem a bit more enlightening. I’ve found Robin Herbert, Marko Vojinovic, and Coel particularly helpful. The main thing that I’ve been taught here however, is that the premise behind Massimo’s coming book does remain solid — this traditional kind of philosophy should never bring us generally accepted understandings of reality. In fact when I was a highly philosophical college kid, I reasoned that perhaps a formal education in the field does tend to place even humanity’s most gifted thinkers, on a road to failure (or at least regarding the development of a community with generally accepted understandings of reality).

    Rather than continue to tax precious grey matter with ontic and epistemic structural reality, I would hope to convince some to take up my own very simple perspective: When compared against reality itself, the human is surely just an idiot. Whether “science,” “philosophy,” or even “deciding what to have for lunch,” there is apparently just one essential process that we can use to figure things out. Here we take what we think we know (evidence), and use this data to evaluate various models of reality. Should I order the tuna sandwich? Perhaps the evidence suggests that I should not. The essential difference between ordering lunch and science, is that science has developed a community with a vast array of generally accepted understandings of reality. (Apparently philosophy both has not and will not.)

    The reason that I am here, however, is because I believe that our mental and behavioral sciences remain quite primitive today, and specifically because there are certain philosophical elements of reality which they will need to formally grasp. Thus I’m trying to promote this message, as well as my own theory from which to potentially found our mental and behavioral sciences. Here we would attempt to determine what we effectively are (ie: the biological nature of good/bad), rather than what we would like ourselves to be (ie: moral).

    Occam’s Beard:
    “Descartes’ cogito could then be, “amo, ergo sum”, I love, therefore I am. Feeling goes as deep as thinking. With respect to metaphysics, when I make a judgement and say “it seems to me”, isn’t my subjective preference at least partly how it feels to me?”

    I’d say that your ideas aren’t nearly so unkempt as your title suggests! I identify the conscious processor as “thought,” and sensations of how something feels are indeed included as such.

    Like

  49. Patrice,

    I’m a bit surprised by the interest generated by category nowadays. I’ve always had the feeling that the basic idea behind category theory actually is quite old and shouldn’t come as a surprise, although CT takes these ideas to new levels, and Grothendieck took them to new levels again – A tout seigneur tout honneur.

    Even before I went the the university, I had a course of “analytical geometry” in which geometrical properties like length weren’t introduced as intrinsic properties of some space, but as things that were invariant under certain transformations (“morphisms” if you wish). The transformations (the “relation”) came first and were the main character in the play, the length just turned out to be something that was invariant under the chosen transformations and therefore could be called a “property” relative to these transformations.

    Relative to other relations you’d have other properties like “parallel transport” but not length and so on. The space had no intrinsic geometrical properties – it had properties relative to transformations. (If you choose the transformations general enough, you’re left with almost no geometrical properties – at least that’s what you think, because mathematics loves to play tricks on you.)

    This approach goes back as far as the Erlanger Programm (1872), I think.

    Like

  50. Hi Marko,

    I’m amazed by the number of times this was quoted. 🙂 Did I really say something so profound there, or controversial, or what is going on?

    Perhaps people were just surprised to read a scientist saying it.:) There seems to be a debate at the moment between the scientists who seem to talk as though metaphysics is completely determined by science and those who say that it is not determined by science at all (Instrumentalism for example).

    Like

Comments are closed.