real patterns all the way down

After hearing and reading about Ladyman and Ross for some time now, with opinions ranging from volatile dislike to euphoric endorsement, I’ve finally taken the time to read Every Thing Must Go and come to my own conclusions. I’ll use this post to sketch my reading of the arguments of ETMG and offer some thoughts and questions at the end. As usual I’m open to feedback and am always curious to hear of alternative readings.

In many ways ETMG is a book I should be naturally predisposed to read. With my own reading of Hume’s thought as compatible with realism (or hyper-realism as I would call it) I should be interested to see how they incorporate Humean verificationism, along with Peircean verificationism (another philosopher I greatly admire), into their arguments for realism. Ladyman and Ross also draw heavily on current work and research in physics, a discipline I initially pursued and still take great interest in; and they use their readings of physics as a mathematics of real patterns to legitimize work in dynamic systems theory (a systems theory properly understood [more on this below], which I draw on heavily in my Philosophy at the Edge of Chaos book. Finally, by their own admission Ladyman and Ross (though I shouldn’t forget the et. al. of Spurrett and Collier) see the arguments of their book as an alternative to the neo-scholasticism of contemporary analytic metaphysics. I agree wholeheartedly with this objective, though I may have an inner neo-scholastic in me for I find some of Ted Sider’s arguments helpful and think Peter Unger’s essay “The Problem of the Many” is a classic, albeit likely a classic in neo-scholastic analytic metaphysics by Ladyman and Ross’s lights. Despite this, I’m generally in agreement with their criticisms of contemporary analytic metaphysics and thus, with this and all the other points mentioned above I had some very good reasons to read this book.

I should begin with Ladyman and Ross’s critique of contemporary analytic metaphysics, but in order to understand this critique it seems important to me first to get a handle on the overall thesis of EMTG. As I read it, the primary objective of EMTG is to steer successfully between the Scylla of Humean antirealism (which is exemplified in its most sophisticated form for Ladyman and Ross in the work of Bas van Frassen) and the Charybdis of traditional scientific realism. Their project is thus one of an attempted rapprochement between antirealism and realism, or it is an effort of ‘consilience’, to cite a term they use repeatedly throughout the book. The conceptual workhorse of this effort is the concept of ‘real patterns’, which they get from Daniel Dennett’s classic 1991 Journal of Philosophy essay of the same name. Ladyman and Ross are quite forthright in affirming the importance of this concept. As they put it,

…to put matters as simply and crudely as possible, it’s real patterns all the way down. This slogan could summarize the whole conclusion of this book… (228)

With this concept in hand, Ladyman and Ross tackle the subject of scientific realism and argue that with their version of realism, what they call ‘ontic structural realism’, they can avoid some of the very compelling arguments against traditional views of scientific realism, arguments that could be seen as motivating van Frassen’s antirealist verificationism. At the same time, however, Ladyman and Ross argue that van Frassen’s antirealism is inadequate of one is ‘to circumscribe the observable in a principled way…[and] it is necessary to endorse some modal facts that are theory-independent’ (107). In their efforts at consilience between realism and antirealism, the concept of real patterns plays a crucial role by enabling, Ladyman and Ross argue, one to maintain a realist position that is immune to its most common criticisms while at the same time giving due emphasis to a Peircean verificationism that is allows for modal facts – namely, real patterns – it’s real patterns all the way down. Before commenting on the general thesis and objectives of EMTG, I want to look at the arguments a little more closely. First I’ll look at the role ‘real patterns’ play in Ladyman and Ross’s understanding of ontic structural realism. In the next section I’ll see how real patterns are incorporated into their version of verificationism. And finally I’ll close with some questions and comments, which will serve more as an invitation for further feedback and comment than as an opportunity for me to pronounce an authoritative verdict regarding the merits EMTG. For me this is an exercise in the ethics and practice of charitable reading.


With respect to science, realism is most commonly used in order to account for the success and failure of scientific theories and research. The reason the theory of the ether failed is that there is no ether, and, conversely, the reason Pateur’s theory of microorganisms succeeded was that there really are microorganisms. More generally, the scientific realist will argue that the reason we know more about the natural universe today than we did in Newton’s or Aristotle’s day is that we are being ineluctably drawn towards the true opinion of reality by the real itself. Peirce is a realist in this sense, for he too will argue in “How to Make Our Ideas Clear” that ‘the opinion which is fated to be ultimately agreed to by all who investigate, is what we mean by truth, and the object represented in this opinion is the real’.

There are problems, however, with this understanding of realism. There are several criticisms that have been made of the traditional conception of scientific realism, or what has also been called naive realism, but ‘the most compelling form of argument against standard scientific realism’, Ladyman and Ross argue, hinges on the understanding of theory change (83). Put simply, if a theory at one point relies upon there being a particular entity such as ether, phlogiston, etc., the advent of a new theory that dismisses such entities will entail a change in the ontology of what is real – ether is no longer real. The problem for the scientific realists that follows from this is that, as Ladyman and Ross put it, we have very solid ‘grounds not for mere agnosticism but for the positive belief that many central theoretical terms of our best contemporary science will be regarded as non-referring by future science.’ Ladyman and Ross’s way around this problem is to reject the notion that the reality that is referred to by ‘our best contemporary science’ consists of self-individuating elements and objects such as ether, phlogiston, electrons, and so on, but rather refers to a structural pattern of relationships–namely, real patterns. Ladyman and Ross state their solution as follows:

Our solution to this problem is to give up the attempt to learn about the nature of unobservable entities from science. The metaphysical import of successful scientific theories consists in their giving correct descriptions of the structure of the world. (92)

This does not mean that science is to be restricted solely to what is actually observable, or to what has actually been observed. This would be van Frassen’s solution, but as we saw above Ladyman and Ross find that even for van Frassen an adequate circumscribing of the observable entails modal possibilities, a modal structure, and it is this modal structure of the world that our best science (mathematics) represents. For example, in the search for exoplanets astronomers use instrumentation to detect subtle changes in the brightness of a star – a slight dimming – with the theory that this darkening is the result of a planet that passes between the observational perspective and the star; and they will use other measurements to detect slight wobbles that are the result of the gravitational pull of planets. In both cases we could say that the observation of the exoplanet is dependent upon a particular theory; however, for Ladyman and Ross a stronger modal claim is implied, namely what are scientific theories represent is not simply what is actually observed but what might possibly be observed; in other words, they represent a structure of modal relations that can be tracked and which allows for the possibility of verification. Given better instrumentation, or a long distance voyage, one would actually see the exoplanet. Now what is primary here for Ladyman and Ross are not the entities that either are observed or observable (e.g., exoplanets), but the modal structure of the phenomena themselves. This is what ultimately enables them to address the problem of theory change. Although the theory of the ether might have been abandoned, the theory that replaces it continues to project forward the structural relations between phenomena that the ether theory accounted for. For Ladyman and Ross, therefore, what is fundament are the structural relations themselves, and relations without relata; or it is the real patterns without a fundamental layer of entities upon which these patterns supervene. In short, it’s real patterns all the way down.

…there are mind-independent phenomena (both possible and actual), but these relations are not supervenient on the properties of unobservable objects and the external relations between them. Rather, this structure is ontologically basic. (128)

This basically sums up Ladyman and Ross’s arguments for their version of realism. As an aside, and in anticipation of some of my own comments, Ladyman and Ross claim that their approach is amenable to a Platonic reading. As they put it, ‘To say that all there is are relations and no relata, is therefore to follow Plato and say that the world of appearances is illusory.’ (152). In other words, for Ladyman and Ross the world of objects, of self-individuated things is illusory, or it is an emergent phenomena of relations which is fundamental – again, it’s relations, real patterns, all the way down. There is a comparison to be made here with Deleuze (I can’t resist, I know). In his “Method of Dramatization” talk, Alexis Philonenko asked Deleuze whether, in comparing Deleuze to Maïmon, there is any role for illusion in Deleuze’s schema. Deleuze responds:

It seems to me we have the means to penetrate the sub-representational, to reach all the way to the roots of spatio-temporal dynamisms, and all the way to the Ideas actualized in them: the elements and ideal events, the relations and singularities are perfectly determinable. (DI 115)

The comparison with Plato should be apparent with Deleuze’s use of the term ‘Ideas’. Similarly, for Ladyman and Ross the ‘world-structure’ or the universal real pattern as they will also discuss this, corresponds to the Platonic Idea. The world of objects as self-individuated elements, as relata, are thus the illusory manifestations of this Idea, much as for Deleuze, as he will put it immediately after stating what is cited above, that ‘The illusion only comes afterward, from the direction of constituted extensions and the qualities that fill these extensions’. But there is a further question is: what does it mean for these relations and singularities to be perfectly determinable? For Ladyman and Ross it means to be represented by our best science. ‘The “world-structure,” Ladyman and Ross argue, ‘just is and exists indpendently of us and we represent it mathematico-physically via our theories.’ (158) For Deleuze, however, the Ideas are not represented; they are too dynamic, to untamed for that. Deleuze will explicitly characterize his understanding of Plato:

If we think of the Plato from the later dialectic, where the Ideas are something like multiplicities that must be traversed by questions such as how? how much? in which case?, then yes, everything Ive said has something Platonic about it. If you’re thinking of the Plato who favors a simplicity of the essence or a ipseity of the Idea, then no. (DI 116)

I suppose a case could be made that Ladyman and Ross understand the ‘world-structure’ as ‘something like multiplicities’ and not as ‘a simplicity of the essence’, but once this Idea is represented we no longer have the dynamisms of multiplicities. Ladyman and Ross appear to acknowledge this themselves when they admit that mathematical representations will continually encounter untamed phenomena that requires further domestication. In any event, it’s time to move on to discuss Ladyman and Ross’s version of verificationism.


On the one hand it might seem that verificationism of any sort would undermine the type of realism Ladyman and Ross affirm. If the ‘world-structure’ just is and exists ‘independently of us,’ whereby independently of us is taken to mean independent of any givenness to us, or hence as independent of our access or accessibility to it, then tying real patterns to verification would indeed seem to undermine their position. Ladyman and Ross are aware of these potential difficulties and attempt to head them off, first by emphasizing that they endorse a Peircean verificationism, by which they mean that for a pattern to be real,

it must be such that a community of inquirers who wished to maximize their stock of true beliefs would continue to be motivated to track the pattern notwithstanding any shifts in practical, commercial, or ideological preferences that are not justified by new evidence bearing on the epistemic redundancy or non-redundancy of the pattern. (36)

Ladyman and Ross will later follow Dennett and argue that ‘there are (presumably) real patterns in lifeless parts of the universe that no actual observer will ever reach, and further real patterns whose data points are before our eyes right now, but which no computer we can instantiate or design will ever marshal the energy to compact.’ To resolve the apparent inconsistency between claiming a real pattern is real only if it motivates a community of inquirers to continue to track it and this subsequent claim that there are, presumably, untrackable patterns, Ladyman and Ross will make a distinction between the universal real pattern, the Platonic Idea, which is real independent of what we take it to be though inseparable from the tracking of fundamental physics, and the real patterns of different ontic scales – what they will call the scale relativity of ontology, or rainforest realism. For example, my ability to track a song across different versions, such as the song My Way as sung by Sid Vicious and Frank Sinatra, entails tracking it at a scale that need not consider the scale and physics of sound waves, though the latter, and especially fundamental physics, sets the constraint within which the relative ontologies are able to track real patterns. To give on other example, economies and institutions can be tracked as real patterns at their relative scale and independently of the biologies and psychologies of the individuals, which would involve another scale.

The question to address now is what it means to track a real pattern, for the answer to this question gets to the heart of Ladyman and Ross’s understanding of verificationism. The first important component of their answer is the concept of ‘compressibility’. A basic component of real patterns, and hence of reality since it’s real patterns all the way down, is that they entail the compression of a larger data set, a compression that allows for the tracking and projectibility of the pattern. My ability to recognize the song as sung by Sid Vicious or Frank Sinatra is an example of tracking a real pattern that compresses the data such that it can be tracked repeatedly and in different contexts and circumstances. Compressibility, however, does not mean that any person needs to be capable of compressing the data into a trackable pattern. They are clear on this point:

As scientific realists we understand the foregoing as referring to compressibility by any physically possible observer/computer–that is to say, given our verificationism, compressible period, rather than compressibility by people. (221)

That said, some trackable patterns are real and others are not. I can follow the patterns of the stock charts, or I may look for patterns in the stars and create elaborate astrological charts; or I may seek to resurrect the ancient Roman practice of reading entrails. In each of these cases one may seek a pattern in order to project future possibilities – when to buy or sell; whether another person is compatible; or when to initiate a military campaign. But Ladyman and Ross would claim that we are right to admit that these patterns, even if they are there, are not real. What differentiates between real and unreal patterns, according to Ladyman and Ross, is ‘that a pattern, to be real, must be projectible from a perspective that physics tells us could be physically occupied.’ (236). In other words, if a pattern is projectible in terms of our best contemporary science, namely our best ‘mathematico-physical’ representations of the world-structure, then it is real; if not, it is not. Since the reading of entrails has long since been tracked by our best science, it has long since ceased to be considered a real pattern.

This brings me, finally, to Ladyman and Ross’s reminder to readers ‘of the deflationary nature of our commitment to modal structure. We claim that science has provided evidence that some structural properties are properties of the whole universe. The evidence for all generalizations of this sort is testable.’ (303). As I read this deflationary account, to say that a pattern is real is simply to say that it can be tracked by the universal real pattern, or by the modal world-structure as represented by our best contemporary science. The task of a metaphysics naturalized, therefore, is to unify the various scaled ontologies with respect to fundamental physics. It is for this reason that philosophy and philosophers, on Ladyman and Ross’s understanding, are to serve as ‘naturalistic philosophical under-labourers to science…’ (242). The efforts of these underlabourers are testable, open to verification and likely to change as we continue to track the universal real pattern, but the reality of patterns is nothing less than their being projectible by our best contemporary science. Nothing more need or can be said.


I have gone on for far too long but I want to close with a few comments and questions. My main task in this post was to present a fair reading of what Ladyman and Ross are up to. Up to this point I have neither been attempting to disavow or embrace their arguments, but have just been attempting to understand them. With this work out of the way I’ll list a few questions/problems that I have with ETMG.

1. As alluded to earlier, I am not persuaded by their deflationary account. If the reality of the world-structure is forever to exceed our mathematico-physical representations, as Ladyman and Ross imply, then why accept the deflationary account that reality is nothing less than what our mathematico-phisical representations tell us is real? I’m more inclined, as I’ve argued in numerous other posts, to embrace a Deleuzo-Humean hyper-realism.

2. I’m also not sure that the account of verificationism Ladyman and Ross argue for has escaped the antirealist circle (or the correlationist circle as Meillassoux might argue). Although real patterns need to be compressible from a perspective that is not necessarily a perspective occupied by people, it is nonetheless a perspective given to a conceivable subject, even a non-human one. Now one can take this in a panpsychic direction, or even a Whiteheadean direction, and argue that there is indeed a possible, nonhuman perspective that can be taken, but Ladyman and Ross do not go in this direction and it seems it would be incompatible with the stress they place upon mathematico-physical representations.

3. And finally, although Ladyman and Ross argue that there are nothing but real patterns all the way down and that individuated objects and entities are merely emergent phenomena of the patterns themselves, it seems possible to me to embrace the claim that real patterns are sufficiently identifiable to be considered things or objects. Just as Spinoza argued, and as I touch on in earlier posts, that a singular individual remains singular as long as it maintains its pattern of motion and rest, it seems to me that real patterns can be considered singular things themselves and hence a mereology of real patterns becomes possible, and perhaps as well their critique of analytic metaphysics for relying too heavily upon self-individuating elements is overstated.

There are probably other things that I will think of, but this is what comes to mind first.

6 thoughts on “real patterns all the way down

  1. “if a pattern is projectible in terms of our best contemporary science…” Here L&R might have reproduced the problem they were trying to avoid by going relations rather than relata. I’m thinking specifically not so much of patterns in entrails, but diffraction patterns supposedly produced by a single photon entering one or both or whatever in the double slit experiment. From what “physical space” is this occurring? Quantum theory (official version) has a taboo on giving a definite answer to this at present.

    If patterns that can be projected from somewhere in space are what their view depends on, then they are not using contemporary physics.

    The notion of “somewhere” in physical space at all is also undermined by the easily reproducible phenomenon of nonlocality, which Schrödinger calls the one salient feature of quantum theory. If they accept nonlocal patterns, then they should give Rupert Sheldrake’s morphic resonance a spin. But then most contemporary scientists think that’s worse than entrails…

    • Thanks for this Tim. I think you may be right about their solution via real patterns reproducing the problem they hoped to solve. I don’t think they would be particularly bothered by the example of nonlocality with respect to a projection from somewhere for they do discuss Bell’s inequality theorem which sets forth a fairly sophisticated argument on behalf of nonlocality itself – hence, on their reading, even nonlocality is projectible by our best contemporary science. However, where I do think you might be on to something is with the purpose L&R made the move to relations without relata in the first place–namely, as I see it this move was pivotal to their effort to avoid the problems that beset scientific realism in accounting for the ontology of entities as a result of theory change – now there is ether, now there is not, etc. For L&R, however, there is sufficient continuity of phenomena to be tracked and represented by theories such that the move to the mathematico-physical representations of this continuity can avoid these problems. But when L&R make statements such as the following,

      Thus it appears overwhelmingly likely that some kind of mathematical structure that resists domestication is going to be ineliminable in the representation of the world in fundamental physics.

      I find that there position becomes difficult to maintain and wonder why they need to maintain the deflationary view, for it seems that even the real patterns that are tracked and projectible in terms of our best contemprary science will entail, by L&R’s own admission, an ineliminable ‘mathematical structure that resists domestication,’ and hence a structure that is not projectible unless and until it is domesticated by a mathematico-physical representation (which is why I call for a Deleuzean logic of sense rather than a logic of representation to address these issues, but this entails a nondeflationary account of reality). This raises again the whole problem of differentiating between those patterns that are real and those that are not (which you bring up with the point about entrails). In the process of avoiding the problems with traditional scientific realism, it appears that L&R have brought on themselves another problem, what you might call the problem of untamed and untamable reality.

    • Thanks for this David. I am more committed to my critique of their deflationary realism than to their possible correlationism. In essence, I don’t dispute the necessity of mathematico-physical representations, and if this entails a form of correlationism then so be it, for in this case much if not all of modern science would be correlationist. My bringing up of correlationism was more in the service of my main point, which is that L&R’s understanding of science is not sufficiently realist, and it is this insufficient realism associated with correlationism that motivates the critiques of Meillassoux, Harman, and others. For my purposes I would go after L&R’s insufficient realism by arguing directly for hyper-realism rather than by way of rejecting correlationism (though admittedly I do not do that in this post, but attempt to sketch how this would look in other posts). Related to this I disagree with L&R’s claim that philosophers ought simply to be under-labourers to science. Philosophy as I understand it creates concepts that take advantage of a logic of sense that is irreducible to the agenda of scientists who seek, as L&R readily admit, to put forth the best mathematico-physical representations of real patterns – hence they follow a logic of representation rather than a logic of sense, and the latter does justice to realism on my view whereas the former, if it is to be consistent, must be committed to a deflationary realism.

  2. Pingback: Dewey…the most important philosopher of the twentieth century | Aberrant Monism

  3. Pingback: siding with history | Aberrant Monism

Leave a Reply

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

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s