What it is to be something: order and semiosis

by Arlyn Culwick


This paper proposes a semiosic approach to the problem of how something might come from nothing, and uses it to illuminate the problem of how order or directionality might occur. I approach the problem with the question: “in what respect is there something if it is not also something to something else?” For example, consider a collection of objects, none of which can have any kind of relation to each other. In what respect could they comprise a universe? – in no respect, since there is no principle of unity (not even spatiality) by which the objects can be one system. From the perspective of what it is to be a system, there is no difference between there being no objects at all and each object being nothing to any other object: in either case the outcome is identical: no unity, thus no universe.

Now suppose that for anything to be an object, it would have to be a system of interactions in its own right. (This appears to be the case for the actual universe.) Since the above system is one in which there are no relations, it could contain no objects, since being an object requires having relations. In other words, there is “nothing”.

This scenario will be used to make a case for the primacy of relation or “form” over “matter”. In addition it will be used as a limit-case to discuss the status of order and states of randomness or chaos. Of particular significance is the distinction between cases of no interactions and of indeterminate interactions (i.e. randomness), which sheds light on conditions for the possibility of randomness in the first place. The notion of the often-presumed antecedence of chaos to order in the natural world will be addressed. Lastly it will be argued that a fully reductionistic universe is untenable under these considerations, and that “things” (including energy) cannot be physically fundamental.


Download this paper as a PDF

Presented on 18 July at the Ian Ramsey Centre‘s 2014 conference on Evolution and Historical Explanation in Oxford.



Note to the reader:

This is a short paper intended to promote discussion and further thinking, rather than to present an argument exhaustively. In order to facilitate this, it is written with an absolute minimum of technical terminology, and in a simple, direct style.

If I were to advance an ethics of scholarly presentation, I would struggle to reconcile the merit of facilitating maximum engagement with a topic and the merit of perfect scholarly rigour and completeness. The two are at least partially opposed and must be held in tension. The former is favoured here, and to that end I hope to minimise the need for background knowledge of the subject and make use of thought experiments wherever possible, so that for the most part the reader simply needs to think with me, to test out my observations themselves, and either come to the same conclusions, or disagree.




One of the oldest conundrums in philosophy is the question of how it might be possible for something to come from nothing. Related to this is the question of why there is a universe in the first place, rather than nothing at all. In this paper I propose a resolution to these questions by taking a perspective informed by mind-independent sign-action or semiosis.

For obvious perceptual reasons it is reasonable to believe that the universe is made of “stuff.” It is also sensible enough to extend this hypothesis and suppose that the universe is fundamentally made of “stuff.” Such has been the broad course of a lot of Western philosophical history, if not under the guise of “matter” then under a matter/form hylomorphism.[i] In the twentieth century, the physical sciences challenged this notion to some extent with the assertion that energy is physically fundamental, with field theory, waves, and fundamental “particles” that are anything but material in the normal sense of the word. However this has not generally displaced the deep-seated intuition that in order for there to be something, it does not suffice to just be a bare attribute, or a mere relation, or an abstraction or form. No, such things are thought to depend on “stuff” under some rationale – that is, attributes need subjects, relations depend on their terms’ existence, and abstractions require the objects from which they’re abstracted. In every case, the former entities depend for their existence or instantiation upon “matter,” broadly construed.

Not unrelated is the question of the place and priority of order and chaos or randomness. This and the above are frequently intertwined, often with consequences as to whether order or chaos is given physical or metaphysical fundamentality. Did the gods create order out of primordial chaos? Did God create the universe out of nothing at all or impart order? Does matter emerge from sub-quantum indeterminacy? Alternatively, perhaps randomness thrives in “pockets” of reality where interactions are underdetermined; or perhaps deterministic systems generate infinitely complex (yet still deterministic) “chaos.” What is our narrative about chaos and order? A narrative will be proposed here, again entailed by the mind-independent action of signs.

The progressive erosion of “matter” by advances in physics

The physical sciences experienced a profound advance when Isaac Newton proposed a theory that specified only rules of interaction between bodies in space, rather than anything directly concerning their materiality. As Ralph Austin Powell observes,[ii] under an Aristotelian-Thomistic framework Newton’s laws specify only formal causes of a particular type, that of extrinsic formal specification,[1] rather than causes associated with materiality like agent or efficient causality. Extrinsic formal causality later proves empirically useful for Einstein, when, in his theory of general relativity, he substitutes gravitation – an agent cause – with space-time curvature, an extrinsic formal cause.[iii] These two cases are examples of an often-unrecognised, but crushingly total, triumph of science on behalf of philosophical realism over nominalism. It is a matter of mind-independent physical relations, not substances, turning out to be (a) what are available to empirical observation and expressible in theoretical discourse, and (b) turning out to be what the physical world is constituted by.

This latter point – that the universe appears to be constituted by relations – needs bearing out. I do not suppose that the physical world is only constituted by relations, since this would be to go beyond the bounds of what is empirically observable and theoretically describable (namely, extrinsic formal specification); but I do assert that scientific discovery reveals truths about the universe.[2] Neither do I suppose that it is necessary to explain relative being under an Aristotelian-Thomistic rationale; what is important here is the nature of what is described, not the terminology used. In fact, that a minor and peripheral scholastic causal mode, originally intended to explain mind-dependent action, turns out to be this significant in the history of scientific theory may well suggest a rewriting of causal schemes along non-Thomistic lines. Nevertheless, the following regress seems to occur in physical sciences in which “matter” is reduced to relations:

  • A theory – say, gas laws – explains (and predicts the behaviour of) an object or system of objects in terms of relations between aspects, components, or attributes it has (in this case, molecules).
  • Since relations necessarily have terms, the objects or systems embodying these terms are empirically investigated, resulting in a theory that describes their internal relations (e.g. molecules are in fact atoms interacting).
  • These new relations’ terms (atoms) are investigated, resulting, of course, in a further theory that describes the behaviour and constitution of the object in question in terms of further relations between new terms, ad infinitum.
  • Thus, gases are relations between molecules, which are relations between atoms, which are relations between electrons, protons, and/or neutrons, which are relations between quarks, leptons, and bosons.

The latter are called “fundamental” particles not because they are known to be indivisible “building blocks” of reality, but merely because it is unknown whether they have parts in relation. It is possible that at smaller scales there are no further parts or “particles” (that is entities identifiable by energy level, spin, etc.), but this is not to say that there would be nothing; there would be something – perhaps simply a vague fuzz of relations between relations, with no term having intrinsic specification of any kind. Such a scenario is theorised under the name “quantum foam” to be interspersed by particles (and their complimentary antiparticles) momentarily popping into existence and then annihilating themselves, giving spacetime a granular texture that otherwise would be “smooth”, that is to say, having nothing to cause enough disturbance to quantise out a particle. This scenario suggests thoroughgoing fundamental relationality that lacks the stability to sustain a mechanism to produce “stuff” reliably.

Thus, the picture suggested by the progress of physics is of a world both fundamentally and pervasively relational, with each term (or object) turning out to be simply another set of relations. It would be relations “all the way down,” to a point where no mechanism is present to cause discrete entities to exist. At this point, would there simply be relations and no “stuff”? Is this possible without relations having terms (as they must, necessarily)? Can matter (under some exotic guise) somehow still be a genuine fundamental property of the universe?

Being something to something else

It does appear as if “matter” dissolves into purely relative being. Yet relations require terms, or else there would be nothing being related, thus no relation. This is a conundrum, and to resolve it, I propose to consider it in the light of a core idea undergirding sign-action: what it would mean for something to be something to something else:

Imagine a collection of objects, none of which can have any kind of relation to each other.[3] In what respect could they comprise a system? – in no respect, since there is no means (not even spatiality) by which the objects can be related. There would be nothing over and above their existence by which they could comprise a system.

Universes necessarily are systems

Now a universe is necessarily a system, however scantily related its components might be, or else there would be nothing to make it an entity in its own right over and above its components. If one were to argue the contrary, the only basis by which a universe could be anything would be by force of describing it as such in language or in thought. As such, however, it would be a mere nominalism. Without a mind-independent basis intrinsic to the universe in question, there would be no way in which a universe could be anything in its own right.

Therefore from the perspective of being a universe and not just a relationless “collection”[4] there is no difference between there being no objects at all and each object being nothing to any other object: in either case the outcome is identical: since a universe requires some way of relating, there would be no way to be a universe. Therefore a universe is impossible if objects can have no relation to each other.

Objects necessarily are systems too

Now suppose that for anything to be an object, it would have to be a system of interactions in its own right. This appears to be the case for the actual universe, as illustrated just now. Since the above scenario is one in which there are no relations, it follows that it could contain no objects, since being an object requires having relations. Therefore, this scenario can feature neither objects nor a universe. In other words, the scenario contains “nothing” at all.

Relative nothing and absolute nothing

It must be clarified that although I denote a very high grade of nothingness by the term “nothing”, I do not mean nothing abolutely. After all, it is perfectly conceivable that something exists beyond the human capacity to imagine its being, involving in itself neither relation nor matter, nor a universe. However such an entity would not have parts; it would not have attributes; it would not be an object; it would not be a relation. There is no content of any kind by which one might think of such an entity, other than in terms of the bare property of existence. There are no grounds to rule out the existence of such an entity. Therefore the above scenario only reveals that there can be “nothing” in terms of the capacity for human thought to imagine how things can be, rather than nothing absolutely. As such, this is the most basic content to the term “nothing” that can be employed concerning the universe, since it cannot be ascertained whether absolute nothingness obtains. As such it is an acceptable grade of nothingness to address the question of why there is something (that is, a universe) at all.

Objects and objecthood

Qualifications aside, this scenario sheds light on the question of whether relations or their terms (that is, objects) are physically fundamental. As scientific progress implies, any given object evidently reduces to relations all the way down. In the actual universe, at tiny scales there might be no “particles”, but logically there will always be terms of relations, even if these are just more relations functioning as terms. So although “objects” may not always be present, “objecthood”, that is, whatever functions as a term of a relation, logically cannot be absent from any physical situation.

The fundamental primacy of relations over objecthood

However this does not make relations and objecthood, or form and matter, hylomorphic or jointly fundamental. On the contrary, it establishes relations as physically fundamental. For the scenario implies two things:

Firstly, since objects are always constituted by relations, objecthood is always a kind of construct or composite. In contrast, a relation is not constituted by its terms, but is instead a purely simple entity. For example, a relation between two objects, say, that of “being-to-the-left,” is a physical attribute of the universe with no component parts; it is a bare specification of physical spatial relation and is not made up of anything simpler. The same applies to attributes, otherwise known as “transcendental relations.” For example, considered in itself (as opposed to being considered as an attribute of something), the blueness of a blue shirt is a simple quality with no parts. As such, relations can be simple whereas objects are always composites of relations. Therefore, relations are more physically fundamental.

Secondly, relation is the defining feature of what it is to be a universe. By definition a universe is a kind of whole, a unity of parts; therefore what defines a universe is not its parts, nor the sum of its parts, but the fact of its parts being related. Put this way, a universe just is these relations. Therefore in a second sense – as the condition for universehood – relation is fundamental to the universe.

Redefining matter

In this light, matter takes on a doubly subordinate status, far from the hylomorphism of Aristotle or the atomism of (most) modern science. It is first fundamentally subordinate to form (that is, relation), and furthermore is empirically only one of several conditions of objecthood, since at tiny scales the sense of the term “matter” is lost when applied to even fundamental particles – never mind to waves, fields, or energy.

Under the perspective developed here then, matter reduces simply to “objecthood,” that is, to being the term of a relation, regardless of the scale or fundamentality of the relation. Matter is thus simply how relations are constructed or signified when functioning as terms of another relation. As such, matter is an entirely relative phenomenon: it is relations functioning as a term of a further relation, which thus functions to specify them as an object.[5]


In the light of the above, order and chaos take on clear roles, fundamentally speaking. Since a universe has, as its fundamental organising principle, things only being something if they are something to something else, then necessarily every entity in the universe must participate in this organising principle. Such a principle is the minimal order required to be a universe. But this entails that universes (and all their parts) fundamentally have order. Therefore, for a universe, order is more fundamental than chaos (of either the deterministic or absolute sort).

Randomness as a species of order

However this is a distinction between there being interactions and no interactions (or nothingness), not of there being determinate interactions and indeterminate interactions, the latter of which is what is meant by randomness. A case of indeterminate interactions requires that interactions are possible – specifically that something must be something to something else – no matter how vague or changeable or minimally determined the relation might be. Thus true randomness is a species of order.

Determinacy and freedom

Furthermore it appears that there is nothing to prevent the existence or even the prevalence of true randomness in a universe. It is one thing to be related in some respect, it is quite another thing to be related in every respect, in which case the two things would simply be one thing. Now if things are not related in every respect, then there is at least one respect in which they do not determine each other. And if they do not determine each other in a given respect, then, contrary to the common view that order equates to determinism and thereby rules out the possibility of freedom, their relations in this respect are “free” or “accidental”. This is to say that their interactions in this specific respect would be truly random (i.e. meaningless, or “nothing”) to each other,[6] or at the very least either partly or vaguely (indeterminately) related. Thus, random interactions in the universe would be characterised by entities relating to each other, but in a limited way, where they are in some respects not related, such as in Brownian motion where entities interact but nonetheless their velocity and direction remain chaotic.

A non-reductionistic universe

The model I have been developing reveals randomness to be a sort of minimal species of order, and establishes a specific kind of order – that of being something to something else – as basic to universes. However this does not imply that the universe is thereby reductionistic, that is, that every relation is simply specified deterministically by simpler relations all the way down. This is because there is no necessity that entities determine each other fully; entities can relate in limited respects. If this is the case, then interactions between entities of a given type do not reduce to interactions of entities constituting them, since the larger entities are partially constituted by indeterminate interactions that have the freedom to introduce new attributes to them unrelated to the interactions of their component parts. As such, a minimum requirement is met to satisfy the condition of being non-reductionistic (though indeed there are other – and significant – additional features that make the universe non-reductionistic which cannot be developed here).


The theses in this paper stem from a simple notion: that something, in order to be anything, must be something to something else. This founding idea gives rise to what is known as sign-action or semiosis, an account of the process of being or becoming something to something else. In this paper I have embodied some of its most elementary aspects without formally introducing its terminology, which is a pity, but time is short. Nonetheless a new perspective on reality has been proposed. Due to its fundamental relativity it might be described as “second-personal physics”[7] as opposed to the traditional “first-person perspective” implicated in treating objects as things in themselves. This perspective is distinct from an Aristotelian and an atomistic one; it establishes the primacy of relative being over absolute being; it has a clear notion of physical causality as extrinsic formal specification – rather than the current mess that is philosophy of science; and it yields an account of non-reductionistic freedom. I hope that this is an idea that will be found worthy of being seriously entertained.

Autobiographical statement

Arlyn Culwick is an independent researcher in analytical philosophy based in Oxford. His research interests include the action of signs, the ontology of relations, teleology, autopoetic systems and the nature of significance. These topics combine in a causal model of sign action or “semiosis” with sound empirical and novel metaphysical implications.



[1]   That is, caused extrinsically to the entity in question, rather than intrinsically as the original Aristotelian causes operate. Additionally the causal action is to “specify” or delineate an attribute in the affected entity rather than to exemplify it, as would be the case with a blueprint. Hence “specificative extrinsic formal cause”. For more information on this besides Powell’s article referred to above, see John Poinsot’s Tractatus de Signis (1632); details given in the reference list below.

[2]   In other words, humanity knows about the physical world thanks to empirical investigation. There may be other aspects of the world that we can’t know about, and so it would be naïve to suppose that empirical investigation yields the whole picture. However this is not to cast doubt on the truth of scientific discoveries. It is just to be aware that truth and the whole truth are distinct.

[3]   For the sake of clarity: I do not think that you will be able to visualise a collection of objects that can have no relation to each other. However by a simple process of abstraction it is not difficult to consider such a scenario. As soon as you think of anything by which the objects might be related, rule it out: By their being related in space? No. By their having shapes? No. By having a surface? No. By having the attribute of being Turkish? No. And so forth…

[4]   It is a linguistic and mental fact that by thinking of things as being a “collection” they’re thereby related in thought. However this is to be taken strictly as an artefact of thought, with no bearing on the scenario itself.

[5]   For an introduction to what this amounts to – and it is in fact the process of sign action itself – see the paper I delivered three days ago at the previous conference. Secondly I want to emphasise that although objecthood is simply being the term of a relation, this does not mean that the relation is merely extrinsic to the object and thus that objects do not have intrinsic natures. On the contrary, the relation is what makes an object an object; it is its intrinsic nature. Secondly objects need not be (and are not) objects in terms of only one relation, but generally, and especially for the domain of classical physics, are multiply specified.

[6]   I do not mean to suggest this as a solution to the problem of “free will”, even though it is often sought for as if it would solve the problem. After all, to be undetermined is to have no reason to act, thus undermining what it means to will. Therefore if human will is free, it would necessarily be in a different sense, such as that of being able to do what one wants to do.

[7]   Credit – and thanks – goes to Susan Eastman for coining this term in a conversation this week.


[i]   Aristotle, Metaphysics, W. D. Ross (trans), 350 BCE, http://classics.mit.edu/Aristotle/metaphysics.9.ix.html (accessed 2014-07-18)

[ii]   (Powell, 1989, p. 186)

[iii] ibid.