Harman: The reason for this dialogue is that we are both realist philosophers working in a subfield, continental philosophy, that has never been sympathetic to realism. In this tradition the usual procedure, following in the footsteps of Edmund Husserl and Martin Heidegger, is to treat realism as a pseudo-problem. It is said that we “always already find ourselves outside in the world” in the mere act of intending an object before the mind. And since the phenomenologist “brackets” the question of whether or not a given phenomenon really exists, then even when dealing with illusory objects we are already outside in the world. For many years I was inclined toward phenomenology myself, and thus am well aware of the various self-deceptions this otherwise admirable school employs in evading the question of how the phenomenal and the real must differ. I saw this again recently in an anti-realist article by Dan Zahavi, the phenomenology gatekeeper of my generation (Zahavi 2016).
Until quite recently, almost no philosopher who was continentally trained saw anything of value in a realist position. Indeed, in our first correspondence some years ago, you stated accurately that “for decades admitting that one was a realist was equivalent to acknowledging one was a child molester” (DeLanda 2007). The situation in analytic philosophy has always been different. To some extent analytic philosophy was launched explicitly by G.E. Moore and Bertrand Russell as a realist reaction against the neo-Hegelian British Idealism of F.H. Bradley and others (Soames 2014). There are also plenty of anti-realist analytic philosophers, of course, but my point is this: analytic philosophers have always taken realism seriously in a way that continental philosophers generally still do not.
Now, I know that you actually came to philosophy from computers by way of analytic philosophy, when you found yourself wanting to understand programming languages. But you eventually found your chief influence in a continental philosopher, Gilles Deleuze, and despite the precision and clarity of your writing you are better known by continentals than by analytic philosophers. My own career has been less unorthodox than yours, though still atypical. I attended two rather continental graduate programs in philosophy, and was driven to a realist position by finding in my dissertation that Heidegger only makes sense if pushed in a radically realist direction. That dissertation became my first book, Tool-Being (Harman 2002). In a remarkable coincidence, that was the same year that you published your landmark realist interpretation of Deleuze – though even your fans often accept everything but your realism (DeLanda 2002). I can't remember at the moment if your previous books had also declared realism explicitly (DeLanda 1991; 2007). But at least we have a specific date, 2002, as the beginning of a prominent realist current in continental philosophy, flowing from multiple directions. By 2007, there was the Speculative Realist movement, featuring me and three colleagues, which broke apart two years later due to personal and philosophical disagreements (Brassier et al. 2007). In 2011 came the New Realism group of such authors as Maurizio Ferraris (2015) in Turin and Markus Gabriel (2015) in Bonn. I have since learned that Ferraris was on the realist bandwagon as early as 1991, which led directly to a break with his mentor Gianni Vattimo, one of the most vociferous continental anti-realists of recent decades (Vattimo 1991).
DeLanda: Let me add a few remarks to your historical capsule. I was an unapologetic realist after 1991, the year in which my book about warfare was published. The space of the battlefield, although it is decidedly a cultural space, is inhabited by metallic projectiles, shrapnel, shock waves, and fire. All those lethal objects affect human soldiers, leaving corpses and mutilated bodies behind, regardless of whether the soldiers believe the objects exist or not. And for similar reasons my book on the history of the millennium, focusing on matter and energy flows, famines and epidemics, was also unambiguously realist. To take just one example, bacteria and viruses were objectively affecting our bodies centuries before we formed any beliefs about them. On the other hand, I did not offer an argument in those books for the position that all coherent materialisms must be forms of realism. I just took for granted that if human history had been so deeply affected by the material culture of weapons and battles, of vaccines and quarantines, of matter and energy flows in industry and trade, then a belief in a mind-independent world followed logically.
Harman: That brings us to the present moment. On the one hand, we have your philosophy, which is both ardently realist and ardently materialist. Then there is my own position, which is ardently realist while rejecting materialism as a form of either upward or downward reductionism, depending on whether it takes a scientific/Marxist or social constructionist form (Harman 2010a). But there is also a third position that is sometimes confused with both your philosophy and mine, which might be described as “materialism without realism.” Here we find Karen Barad, a materialist who also argues that objects have no reality apart from their interactions with the mind; she bases this on Niels Bohr's Copenhagen Interpretation of quantum theory (Barad 2007). Although Barad calls her philosophy “agential realism,” there is nothing realist about it, since she grants reality no autonomy from the human mind, or at least not from human practices.
Along with Barad's philosophy, we find another, perhaps even more prominent form of materialism without realism. This can be found in the two most influential continental philosophers living today: Alain Badiou and Slavoj Žižek. Despite their differences, Badiou and Žižek are close allies, and it is easy to see why. First, both sympathize with German Idealism's elimination of the Kantian thing-in-itself. And second, both sympathize with Lacan's elimination of the Freudian unconscious as a hidden psychic force in favor of the unconscious as an immanent disruption of consciousness: slips of the tongue as gaps in language itself, the encounter with a traumatic kernel that resists symbolization, and so forth. The German Idealist and the Lacanian moves are clearly anti-realist in both letter and spirit, and this is precisely where most continental philosophers are positioned today. Žižek even goes so far as to say: “The true formula of materialism is not that there is some noumenal reality beyond our distorting perception of it. The only consistent materialist position is that the world does not exist” (Žižek & Daly 2004: 97).
Let me ask you two questions. First, would you like to disagree with or add to anything I have said about the current state of realism? And second, would you agree that Badiou and Žižek are steering philosophy in the wrong direction as regards materialism's relation to realism?
DeLanda: You are correct that I have been using the terms “realism” and “materialism” as if they were interchangeable when they are not. All (coherent) materialists must be realists, but not all realists must be materialists. A devout Christian is surely a realist about heaven and hell, since he would not accept that my disbelief in those transcendent spaces in any way impinges on their actual existence. Yet, such a Christian realist would clearly not be a materialist. My definition of “materialism” is as a form of realism that rejects any entity that transcends the material world. The reason I used the term “realism” in Intensive Science was that it is a term that is much more “in your face” than “materialism,” and because the latter term has been debased in the last few decades beyond recognition. (I also use the term “neo-materialism” for similar reasons.) But I should stop using the terms “realism” and “materialism” as if they were synonymous.
Now, moving on to the debasement of the term “materialism.” The problem with people like Žižek is that they use this term as short for either “dialectical materialism” or “historical materialism.” So the statement you just quoted can be paraphrased as “the only consistent Marxist position is that the world does not exist.” This revised statement is arguably still false: there are plenty of contemporary Marxists who would disagree with it, and certainly the Engels of “The Dialectics of Nature” would condemn the statement as pure bourgeois ideology. But it at least makes sense, since it would point to Žižek's belief that after poststructuralism (most adherents of which were anti-realists), Marxism itself had to change to absorb the new ideas put forward by these thinkers. To me, all this implies (as I have argued for a long time) that Marxism itself is exhausted and that we need to create a brand new leftist political economy based on neo-materialist ideas. You have a more nuanced take on Žižek's strategy. Do you care to share it?
Harman: Žižek for me is a fascinating but sometimes frustrating figure. As an entertainer he's nearly unparalleled in the history of philosophy. Giordano Bruno was probably the last philosopher as funny as Žižek, and before that maybe Diogenes the Cynic. Admittedly, some people don't find him funny at all: all the sex jokes become overwhelming for some readers, and occasionally he crosses the line of decent behavior (the child pornography reference near the beginning of The Parallax View being perhaps the worst instance). Nonetheless, I love Žižek's animal spirits and his tendency to make blunt statements and stick his neck out, unlike his evasive and dithering postmodernist forerunners. I also think he's probably underrated as a philosopher, though of course I agree completely that his statements about how one must now be an idealist to be a materialist are absurd. But I mentioned earlier that he can be frustrating, and what frustrates me most is that he spends very little time on the crucial issue we're discussing right now: realism and anti-realism. He simply takes it for granted that Hegel knocked Kant flat on his back on the question of the things-in-themselves, and Lacan (an extreme idealist whose “Real” has little of reality about it and functions primarily as a trauma for humans) simply bolstered Žižek's confidence in his idealism. I don't get the sense that he has ever really worked through the arguments about realism fully for himself. Above all, he simply seems annoyed by the realist option, and usually just repeats the Hegel–Lacan arguments on this topic.
DeLanda: Žižek as entertainer, or, perhaps more accurately, as social commentator, does have something to offer. There are many social phenomena that are either too insignificant (like the different shapes of toilets across European countries) or too complex and poorly understood to be tackled by sociologists or anthropologists. In these cases, witty social commentary is perhaps the best that we can have, while we wait for a serious theoretical treatment to come along.
Harman: A related point, one that Žižek (1989) raises himself in connection with his discussion of Alfred Sohn-Rethel (1978), concerns another “materialist” but definitely anti-realist use that people have made of Marx in recent years. Whenever I make the realist point that objects must have reality independent of the various networks and social systems in which they are involved, a few people inevitably accuse me of “commodity fetishism.” Yet this seems to be a stunning and rather basic misreading of Marx. “Commodity fetishism” is the first topic of discussion in Das Kapital, where it is clearly a theory of value, not an ontology. In other words, Marx is complaining about people who think that pepper or shirts have value in themselves apart from the labor that went into producing them. Fair enough. But he is not making the anti-realist ontological claim that nothing exists independently of labor. In fact, Marx (1992) gives us at least three examples of economic goods that are nonetheless not commodities: air and water freely used by everyone, items bartered in a tribal system, and even corn-rents delivered to feudal lords. But what we have now are pseudo-Marxists like Andrew Cole (2015) and Alexander Galloway (2012) who misinterpret Marx as an anti-realist metaphysician of culture.
DeLanda: I know what you mean. I once had the misfortune of having an exchange with Galloway in which he dismissed the objectivity of some scientific classifications (e.g., the Periodic Table of the Elements) by simply quoting Foucault quoting Borges. The novelist had come up with a brilliant but nonsensical classification that Foucault used as an example of how arbitrary classifications can be. But Foucault never wrote about any real scientific field: psychiatry, early clinical medicine, grammar, early economics, criminology etc. are all mere discourses, not fields that reliably produce knowledge. Moreover, their subject matter is human beings and that makes them bring institutional norms and practices of control that further distort the discourse. But as much as I despise those who can use a novelist's remarks as an argument against objectivity, I have even more contempt for those who appeal to the worse parts of science – such as Barad. The idea that the consciousness of the observer determines the actual state of an electron is a myth. It was floated in scientific circles (by von Neumann?) as a funny idea to convey the flavor of the uncertainty principle to non-specialists, but it has no basis whatsoever in experimental science, any more than Schrödinger's cat does. In fact, with the exception of a few clueless quantum physicists, no one in the scientific community believes that myth. Hence, humanities departments are the only parts of the academy in which that myth has flourished, and they are, of course, packed with idealist professors.
But what intrigues me at this point is your own position. You mentioned social constructivism as one reason not to embrace materialism, but social constructivists are all linguistic idealists. Marxists are (or used to be) materialists, but theirs is a special brand in which a priori schemes of synthesis (the negation of the negation) form the core of their position. That makes Marxism a bad form of materialism, and certainly does not count as an argument against a materialist position per se. Finally, you mention scientists and suggest that, for them, materialism implies reductionism. This is also wrong. First of all, most scientists are not materialists but empiricists. That is, they believe in the mind-independent existence of objects and events that can be directly perceived (all others are just theoretical posits). Thus, a causal relation is not for them an objective relation in which one event produces another event, whether there is anyone to observe it or not, but the observed constant conjunction of two events. Russell, whom you mentioned before, was not a realist, but an empiricist, as are many of the most famous names in analytic philosophy, like Quine, Goodman, and van Fraassen. Second, the only true reductionists in science are physicists. Chemists have always known that a compound substance (say, water) could not be reduced to its component parts (oxygen and hydrogen). The latter are fuels that excite fire, the former puts out fire. In chemical textbooks from the beginning of the eighteenth century it is already part of the codified consensus that compounds have properties that are novel relative to those of their components, hence that cannot be reduced to them. And, of course, it is chemists, not physicists, who are the real experts on matter.
Harman: Russell is not enough of a realist for me either. I would agree that he is more of an empiricist, but not everyone has as much at stake as you (or Roy Bhaskar) do in distinguishing between realism and empiricism (Bhaskar 1997). At any rate, there are plenty of people in analytic philosophy who interpret Russell no less than Moore as part of a realist reaction to British Idealism.
As for the relation of scientists and social constructionists to materialism, here's how I look at it. “Materialism” in its original form meant a reduction that explained things away in terms of some ultimate physical element or particle. This was true of the pre-Socratics, many Enlightenment thinkers, and even twentieth-century scientists such as Ernest Rutherford, who admitted to imagining atoms as hard red billiard balls.
But there's another way to reduce things: upward rather than downward. This is what idealists and social constructionists do, by saying that nothing is hidden behind the given. I'd put your empiricists on this side as well. Now, why on earth would anyone call this second group “materialist”? Don't ask me, but many of them do call themselves that. I already gave the examples of Barad and Žižek, who don't seem to be the least bit materialist, but presumably want to link themselves to the Left/Enlightenment political prestige of the term, despite the fact that both give the human subject an outsized role as making up 50 percent of the cosmos. This is anathema to traditional materialism, in which the human being is ultimately just another piece of matter, whereas Barad gives the observer the almost magical power of co-creating the universe, and Žižek holds that the thinking subject is so important and unique that it had to be created through some sort of “ontological catastrophe.”
Levi Bryant is just as puzzled by this second claim to materialism as I am. As he nicely puts it: “Materialism has become a terme d'art which has little to do with anything material. Materialism has come to mean simply that something is historical, socially constructed, involves cultural practices, and is contingent … We wonder where the materialism in materialism is” (Bryant 2014: 2).
DeLanda: I agree, but this boils down to the confusion between “materialism” and “Marxism” that I just mentioned (as if Marxists had a monopoly on materialist ideas), with the socially constructed part (which is of phenomenological origin) being the souped-up version of commodity fetishism and bourgeois ideology. The old Left is frankly embarrassing at this point. Hence the need for labels like “neo-materialism” to signal a break with that increasingly irrelevant tradition.
Harman: But I want to press you about your own reason for speaking of materialism. You said that the weakness of realism is that it allows, say, for devout Christians to be realists about Heaven and Hell, whereas materialism would close off that option. All right, but in order to reject belief in Heaven and Hell, is it really necessary to reject belief in anything immaterial? I think a more plausible reason for rejecting the afterlife would simply be to say something like: “No one has ever returned with credible reports of a Heaven or Hell, and I find it absurd that, even if an omnipotent being existed – and why should I believe this in the first place? – then this God would reward or punish souls in the manner of a pre-school teacher.” Wouldn't this be a more prudent reason to reject the afterlife? That way, you could still keep open the possibility of non-material entities. For example, is it utterly impossible that ghosts or spirits exist in some form? William James thought it was quite possible, and kept an open mind about it. What about non-material beings living in other dimensions parallel to ours? Admittedly, there is no compelling evidence for such entities, but what about everyday objects such as tires, which do need matter to exist, but can easily discard particular material elements, and for this reason are best conceived as forms having some sort of loose relationship with matter? There are also the fictional creations of novelists, such as Charles Bovary or Anna Karenina, who aren't material in any bona fide sense.
In short, is it necessary to justify one's skepticism about religious and other supernatural entities by recourse to the rather sweeping theory that nothing exists except whatever is made of physical matter? I think this is a dubious proposition in ontological terms, and, historically speaking, materialism has often led to premature decisions about what should and should not count as real.
DeLanda: Well, first of all, the term “materialism” should not imply only the mind-independent existence of matter, but also energy and physical information (material patterns and forms). It is a shame that a term that includes all three items would be so unwieldy: material-energetic-informationism. But of course we can simply forget about the name of the concept and focus on its definition. Ghosts and disembodied spirits are a separate subject, and no, I do not believe in the Cartesian dualism that their non-material substance would imply. My subjective observers are always embodied and situated. On the other hand, a very important concept in my ontology is that of an emergent property: a property of a whole that is produced by ongoing interactions between its parts. Emergent properties can be quite subtle: they can appear, for example, in computer simulations (e.g., gliders in cellular automata) that exist above the level of programming languages, which in turn exist above the level of computer hardware. Strictly speaking, only the latter is “material,” the next layer being made out of symbols and syntax, and the uppermost layer constituted by patterns or forms interacting with each other as patterns or forms. Since very few people would consider the software running in their computers as something immaterial, I use the analogy quite often to give mathematics a material status, since it is also about mechanical procedures (functions) and data structures. In a similar way, the analogy with simulations can give novels and their characters a less mystifying status.
Now let me ask you this. Neo-materialism involves a rejection of entities that transcend the world of patterned matter-energy. But this includes not only angels and demons, ghosts and spirits, but also Aristotelian essences. What's your position on this?
Harman: I'm much more fond of Aristotle than you are, and this seems to have something to do with our differing views on essence – clearly not one of your own favorite terms, but one of my own unfashionable favorites. For instance, I remember a passage you wrote about how what differentiates two atoms of the same chemical element is the different genesis of each in the core of some star.
DeLanda: Well, we realists must all respect Aristotle; after all, he was for 2500 years the most influential realist philosopher in the world. But the mind-independent entities that are legitimate members of my ontology are all historical; they must have a date of birth (even if we can't establish it) and a date of death (even potentially). Atoms meet that condition; they are born through nucleosynthesis in some star, and have a definite half-life (a given maximum duration for their existence). Aristotelian essences are, by definition, ahistorical, untouched by corruption and decay, as he would say.
Harman: I wonder if you're being too hard on Aristotle. Unlike Plato, Aristotle treats individual things as the primary substances, and these individual things for him are generally destructible. In fact, Aristotle is (as far as I can tell) the first Western philosopher not to identify the substantial with the indestructible, a definite step forward. It seems to me that there's only a problem with the Aristotelian essence if we identify it with the species, and at least in the Metaphysics I don't think it can be read that way.
DeLanda: I disagree. Aristotle's realist ontology includes genera and species (constituting the very nature of a thing, that is, the essence) as well as singular individuals. Thus, Corsicus is metaphysically characterized by his animality (genus), his rationality (species) and his being musical (individual). But Aristotle clearly states that only the first two are necessary, the third is accidental, and that metaphysics as a science can only be concerned with that which is necessary. Hence, although concrete historical individuals are part of his ontology, they are not the kind of thing we can philosophize about, since we can only have a priori knowledge of genera and species. Personally, I do not have any use for a priori knowledge.
Harman: He also says in the Metaphysics that no individual can be defined, because definitions are made of universals whereas individuals are always concrete. At first glance this seems to support your point, since it might be taken to suggest that definitions can only be given of the super-individual, so that philosophy would be unable to work in the realm of individuals. But I read this passage differently. Corsicus and Socrates are not the same individual, and the difference between them is clearly not just one of accidents. Aristotle says that a substance is that which can sustain different qualities at different times: Socrates can be happy or sad, but remains Socrates in both cases. It wouldn't be very Aristotelian to say that the human being can be either Corsicus or Socrates, but is still a human being in both cases. That would sound more like Plato, because “human being” could certainly be a Platonic form, but it would be a stretch to call “human being” an Aristotelian primary substance, since the primary substance would have to be either Corsicus or Socrates individually. And Corsicus and Socrates must therefore each have an essence, an unchanging innate character, because both are capable of supporting shifting accidents such as “happy” or “sad.”
DeLanda: Perhaps, but I do not think Aristotle would agree that the difference between Corsicus and Socrates would be a topic that can be approached philosophically, because there is not a necessary connection between Socrates being Socrates and Corsicus Corsicus, in the same way that there is between Socrates and being a rational animal.
Harman: Let's turn to another issue. You say that mind-independent entities are “defined by the historical processes that created them.” My question would be: which historical processes? All of them? Obviously you can't hold that absolutely everything that happens to an entity defines it, because then your entities would be completely defined by intrinsic relations, a notion you otherwise reject. Not everything that happens to an entity leaves traces on it, because entities are only equipped to register some impacts and not others. Think of all the environmental information around us that we miss simply because we don't have the sensory organs of moths, bats, or dogs. Surely it's the same way with atoms formed inside stars. But this means that what defines entities is not the entirety of their morphogenesis. And furthermore, once an entity is generated, it takes on a life of its own.
DeLanda: You are correct that if the processes leading to the emergence of a given entity are not fully specified we fail to give its historical origin any explanatory value. This was the problem with the first generation of emergentist philosophers – like Samuel Alexander (1920) – who believed that individual entities possess irreducible properties but that the emergence of those properties could not be explained and had to be accepted as a brute fact (they had to be accepted with natural piety). But if we reject this, if we strive to discover mechanisms of emergence (however hypothetical), then the details of the historical process can be specified, and the factors that make a difference in the birth of an entity (the significant factors) can be separated from the insignificant ones. But I detect another objection in what you just said. We both agree that objects must have a historical origin, so the only disagreement is over how much of a trace their birth has left on their identity. This, of course, varies from one kind of object to another. Atoms are born in stars, but we find little or no trace of that in their current nature. Humans (and other animals) do keep traces about their history (birth and experiences) and these memories (and scars) surely affect who they are today. Most other things fall somewhere in between. Species certainly bear traces of their evolutionary past, traces left in their genomes that can be used to track their genealogical connections with each other. But even non-organic objects bear witness to their origins; while a single atom of uranium may not contain information about its past, a large chunk of uranium does, in the statistical distribution of isotopes that it contains – a signature so unique that we can trace samples of actual uranium to the nuclear reactor that first produced them.
Harman: I like this distinction between atoms of uranium and chunks of uranium. But I also have reservations about your strong statement on a priori knowledge. Certainly I agree, if you mean purely deductive knowledge. One of my objections to Quentin Meillassoux (2008) concerns his tendency to think that all philosophical truth must be deducible by purely rational means. However, it also seems to me that the big steps forward in philosophy have not been a posteriori, but have resulted from a priori analysis. Leibniz's relational interpretation of space and time, for instance, was dreamed up in an armchair. Einstein finally put the pieces together for physics, but even Einstein did it with thought experiments, not physical ones. Heidegger's tool-analysis was a dramatic breakthrough in twentieth-century philosophy, but it derives from close intellectual attention to what Husserl claimed to be doing, not from any sort of experimentation.
DeLanda: Sure, thought experiments (what Dennett calls “intuition pumps”) are important in both philosophy and science, and they can lead to the development of powerful insights. One recurrent problem with thought experiments (of the Twin Earth kind) is that the situations are underdescribed at crucial points, a problem that compromises the insights that may be derived from them (Dennett 2013). This is not a problem in computer simulations because you can run them over and over again, methodically changing parameters, and keeping statistics of the different outcomes. You simply cannot do that with thought experiments. But even if we disregard this, the hypotheses generated by thought experiments must eventually be tested so we can support them with evidence. Leibniz may have arrived at a conception of relational space a priori, but we still have to establish if that is in fact the real nature of space. Evidence may never be conclusive or capable of compelling assent, but it is nevertheless indispensable for me. My rejection of a priori knowledge follows a familiar line of argument. Analytic a priori statements like “all bachelors are unmarried man” are true without evidence, but they are trivial or insignificant. Kant famously tried to fix this by introducing the synthetic a priori. The problem is that his three examples of synthetic a priori (Euclidean geometry, Aristotelian logic, and Newtonian physics) all turned out to be a posteriori a hundred years later (once many other geometries, logics, and physical theories had been introduced.) Yet even after all that, Husserl continued to believe in the synthetic a priori, even postulating a special intuition (eidetic) to grasp the exact essences involved in this kind of knowledge. Now, my rejection of any kind of a priori has not made me very popular among philosophers. Indeed, many critics of my book on Deleuze complain that I rely too much on science (that is, a posteriori knowledge) to defend his ideas.
Harman: I'd like to hear you say more about what you think the difference is between philosophy and science. Unlike the critics you mention, I loved the scientific flavor of your Intensive Science and Virtual Philosophy (DeLanda 2002). But do you really think that all philosophy needs to proceed in this way?
DeLanda: I think that philosophy and science have a symbiotic relationship. Scientists are good at producing models of different kinds (mathematical equations, chemical formulas) and at performing causal interventions in reality to produce laboratory phenomena and measure their properties and dispositions. But when it comes to weaving their results into a coherent ontology, their overspecialization, and their tendency to look at physics as providing an exemplar for all scientific practice, leads them astray. Even in questions of epistemology they tend to come up short, being unable to decide if science is descriptive or explanatory. The end result is that physicists end up with a world-view that is based on rhetorical assertions like: “Science is the discovery of eternal and immutable laws through unbiased observation.” Philosophy, on the other hand, can play a synthetic role, bringing rigor to a realist ontology and clarity to epistemology, but only as long as philosophers' long infatuation with physics and its rhetoric of eternal laws is left behind. Unlike specialized scientists, materialist philosophers must take into account all the cognitive products (concepts, statements, problems, explanations, classifications) of as many different scientific fields as possible, and provide a coherent synthesis.
Having said that, there is another point of contact between my philosophy and science, specifically mathematics. I said above that my ontology contains only unique historical entities. I use the term “individual” for these entities, but this needs to be clarified. Today, the terms “individual” and “person” are used as synonyms, but the former can also be used to qualify a variety of other entities. We use it for organisms (individual plants and animals) but it can also be used to qualify communities, organizations, cities, and countries. In all cases, it denotes an entity defined by its history (including its present history). But one can object that this ontology leaves unexplained why there are similarities between individual entities: why are individual hydrogen atoms, for example, similar to each other and dissimilar to helium atoms? This cannot be explained merely by history and must be complemented with an account of why some historical processes regularly yield similar products. At this point, a scientistic philosopher would say, the reason for the regularity is that those processes obey the laws of nature. But as far as I am concerned, to introduce eternal and immutable laws into an ontology is no different from bringing general essences into it. So what else can we do?
Since the eighteenth century, when Leonhard Euler invented the calculus of variations (an algorithm that finds the singularities or special points in differential equations), physicists understood that the systems studied by classical physics have a tendency to be in the state defined by a singularity, typically, a state characterized by a minimum value of some property, as in expressions like “Water seeks the lowest level.” (More generally, all classical systems seek the point at which the difference between potential and kinetic energy is at a minimum.) The regularities in the phenomena studied by classical physics can be explained by the existence of singularities of the minimum or maximum type. Euler himself saw singularities as having an explanatory value that was complementary to that of causes, singularities being what explains the long-term tendencies of the processes that produce individual entities. He phrased this as the need to use both efficient and final causes in the explanation of physical phenomena. Since the time of Euler, other mathematicians (e.g., Henri Poincaré) have discovered singularities that are more versatile than the original ones, but that also explain long-term tendencies to be in a steady state (e.g., homeostasis), to periodically oscillate (e.g., metabolic cycles) or to display more complex behaviors. Hence, my ontology includes both efficient and final causes, although the latter must be seen as involving only a local teleology that lacks absolute necessity; a given process may have several singularities available at once, and which one is presently governing its behavior is a contingent fact. Your ontology, on the other hand, includes efficient and formal causes. Can you elaborate on this?
Harman: In the first place I am attracted to formal causes because of my suspicions about the existence of matter. Where is this matter supposedly located? Where on earth can we find formless matter? Since there is no such thing, we ought to pay attention to forms. As you noted yourself in A New Philosophy of Society, assemblages are surprisingly robust to changes in their components, and I would say the same about forms more generally. Whenever someone appeals to the “materiality” of some situation or intellectual problem, it always turns out to have something to do with forms. This will sound puzzling to some readers, because the recent tendency has been to link form with abstract intellectual models that fail to take the concrete details of a situation into account. But my use of “form” refers to the medieval substantial forms, which were organized and structured and not directly knowable.
The substantial forms were exterminated by Descartes, of course, since he wanted to treat physical realities solely in terms of tangible properties such as position and movement. But they have taken on new life in the work of later thinkers who treat the background conditions of things as more important than their explicitly visible content. Heidegger is an obvious example, though the media theory of Marshall McLuhan would be another (McLuhan & McLuhan 1988; Harman 2013b).
And again, I return to the deficiencies of the concept of matter. Is there any point to this notion other than as a feeble explanation of the difference between real and imaginary things? It is well known that this difference, which parallels the one between essence and existence, has theological roots in the idea of God as the Creator who gives existence to previously conceived essences. I think a better approach would be to assume that there is some difference between the forms of 100 real crowns and 100 imaginary ones, to use Kant's example.
DeLanda: Well, you may be right that there is no matter out there without form, but the opposite is also true: where can we find forms without matter? I believe that once we introduce physical information (that is, pattern without semantic content) into materialism, this problem disappears. But the question of how much form a particular piece of matter has becomes a contingent matter. Atoms, for example, as we understand them today are wave-like entities without any geometrical properties whatsoever. Waves do, of course, have form, but this form is shared with light, electro-magnetism, and other phenomena that were traditionally thought of as immaterial or at least incorporeal. When two atoms form a molecule, on the other hand, geometrical form emerges (angles between bonds, length of bonds) though molecules are still intensely vibratory and do not resemble a stick-and-ball model. Going in the opposite direction, a gas at very intense temperatures becomes a plasma, its atoms losing their identity and becoming an almost formless matter. In short, form (as specified by physical information) may always be there, but how well defined this form is cannot be known a priori.
Let me add a related point. The final causes I postulate (e.g., singularities defining tendencies) act by stabilizing a particular form. Because the singularity corresponds to the state that is “preferred” under particular conditions, it defines the state in which the object will tend to be even as it is buffeted by noise and fluctuations. Thus, a crystal of table salt adopts the form of a cube because this is the form that minimizes bonding energy for its particular components (sodium, chlorine). The cube emerges (every time salt crystals are born) and endures in time, because the singularity defines the most stable state. In an account like mine, in which all objects are historical, stability and endurance are key. Many unstable objects may come into existence and then rapidly disappear, and their lack of an enduring identity makes them invisible not only to human beings, but also “invisible” to other objects, at least those that need a certain amount a time to complete an interaction, a required duration made impossible by the instability of the object. So I do have an account of how final causes relate to the genesis of form. Now let me ask you: how are formal causes supposed to act to confer form upon matter? Since there is no matter without form, the action of formal causes must have been there from the start, but how exactly?
Harman: The problem I have in answering this question is that I do not accept any concept of matter without form, but do accept form without matter. A clear example would be a fictional work, which is filled with entities having a definite form but which do not consist of matter at all. Of course, one could always object that the paper on which the fiction is printed is made of matter, or that the brain of the author is made of matter. But my first response to this would be that we are talking about the fiction itself, not its components or its causal forerunners, and the fiction itself is clearly an immaterial form. My second response would be that even if we agree to look at the substrata of the fiction, where is the matter in these substrata? Paper is not just a pre-Socratic apeiron or amorphous blob, but has a definite formal structure, and the same holds for the molecules and atoms in the paper, and for the brain of the author as well.
DeLanda: Let me just interject that in the case of novels I would not consider the material substratum to be paper. Characters exist several levels of emergence above that. What would need to be done in this particular case is first to give a materialist theory of language, in which the material substrata are either pulses of air shaped with our tongues and palates, or physical inscriptions. Above this basic layer (which together with certain non-linguistic practices can already handle reference) another one develops through a progressive differentiation of simple monolithic (non-recombinable) words, a level of semantic content. Above this, a level of syntax emerges (as the differentiated words become recombinable). I gave a rather detailed account of this in one chapter of my Philosophy and Simulation (2011). Once these emergent layers are in place, we can use them to create yet another level: stories, true or fictional, with characters whose identity is specified using syntax and semantics (as well as narrative conventions.)
Harman: OK, but at what point do we reach this thing called matter? I see it as merely a relative term for the form that happens to lie beneath whatever form we are currently considering. You just gave some excellent examples from science of forms that are more or less well defined, but it seems to me that even the less well-defined forms still have form. For instance, you said: “a gas at very intense temperatures becomes a plasma, its atoms losing their identity and becoming an almost formless matter.” Why “an almost formless matter” and not “an utterly formless matter”? I think the reason is that you already agree with me that there is no such thing as formless matter, but you are still interested in being able to distinguish between more and less organized forms as well as more and less stable forms. I would agree with you that not all forms are equally organized or stable, and would only insist that some degree of form must always be present. But allow me to ask the following: why do you feel the need to retain terms such as “material” and “materialist” at all? I suspect the reason is because materiality is your ultimate principle of the real, as with your earlier examples of Heaven and Hell. You want to say that Heaven and Hell are forms without matter (i.e., they do not exist) but you also want to say that a coal mine, for example, is a form inscribed in matter (i.e., it does exist). But even though it is correct to say that all real and unreal things have form, I do not see why we should say that the real ones have matter and the unreal ones do not, because I am still not sure what this matter is supposed to be.
DeLanda: You ask, “At what point do we reach this thing called matter?” The problem is that there is no straight answer to this question because what matter is is objectively dependent on physical scale. The world may be mind-independent but it is not scale-independent. Take the example of a lake with well-defined geographical borders. Now, two differently scaled organisms inhabiting this lake can be presented with an entirely different reality. A small bacterium, whose weight is negligible relative to the viscosity of the water, will be presented with a medium that allows it to move around only if it keeps its motor (its flagellum) on all the time; a large fish, on the other hand, can throw its much larger weight around, and can therefore swim using a thrust and glide maneuver. To put this in more subjective terms: one and the same object, the lake, will appear much more viscous to one organism, and much less viscous to another. Less subjectively, one and the same lake will provide differently scaled organisms with different opportunities for locomotion. The lesson from this is that once you embody and situate the observer, you thereby give it a scale, and that determines in part how reality will objectively present itself to him.
Now to return to your question: at the human scale matter is best exemplified by macro-quantities of chemical substances, such as the sulfuric acid contained in this glass jar. This is the “matter” humans have interacted with for millennia, from flint to bronze, and from soda to potash. Because I believe in the irreducibility of emergent properties, I do not believe that once we learned about atoms and molecules these macro-substances became a mere appearance. The causal powers of these macro-substances are different from those of their constituent atoms or molecules: a single molecule of sulfuric acid does not burn your flesh; a large population of them does. Hence matter does not refer to “the form that happens to lie beneath whatever form we are currently considering.”
It follows that my “principle of the real,” to use your expression, is efficient causality. The sulfuric acid in this jar exists for me (as a distinct type of material object) not because of the form that I perceive (or even the form I extract by viewing the jar from different angles), but by the experience (real or imagined) of the causal power of the acid to burn my flesh. But if one does not believe in matter, then what exactly are the entities that enter into causal relations? Can we attribute causal powers to forms without a material substratum? And what about all the properties and dispositions that are attributed to material and energetic systems by scientists? Are temperature or pressure properties of forms? Or are we willing to simply dismiss all properties used in science because they do not conform to our ontology?
So, to summarize, I only believe in entities that are immanent to patterned matter-energy, including the kind of local final causes represented by singularities, but I do not believe in anything that transcends that. (The terms “transcendent” and “immanent” must be used carefully because they are ontology-dependent; for idealists, “immanent” means that which does not go beyond subjective experience, while “transcendent” means that which goes beyond it.) But thanks to the concept of emergence, being committed only to matter-energy does not constrain the variety of forms that the latter can sustain, and the regularities in these forms can be explained immanently using only efficient and final causes (singularities.) How do the formal causes you advocate operate, how do they endow matter with form?
Harman: The answer is that I do not think formal causes confer form on matter, because I do not think matter exists. Only form exists, though for me it comes in two varieties: the real (which exists in its own right) and the sensual (which exists only as the correlate of some real entity that encounters it).
So I would rephrase the problem as being not one of how form shapes matter, but of how one real form can influence another, given that no two real things can make contact because of their mutual withdrawal. And the solution for OOO (Triple O, as object-oriented ontology is usually abbreviated) is that they (i.e., real things) only make contact indirectly, through a sensual form that is the mediator between them. By the same token, two sensual forms cannot make direct contact either, but require a real form as their mediator. A good analogy would be magnets, where two poles of the same kind repel, whereas two poles of the opposite kind snap together immediately.
DeLanda: I understand, but your solution to the operation of formal causes (a sensual form mediating between two real forms) needs a lot more elaboration, since in one understanding of “sensual form” this would seem to imply that humanly perceived forms are needed to mediate between two causally interacting real forms, which would reduce your position to idealism. Hence, the term “sensual form” must refer not only to what we humans perceive, but also to what all other non-human entities can “perceive” of each other. Since this is a very peculiar way of conceiving what the term “sensual form” refers to, let's postpone the discussion until we have introduced enough background.
Harman: You have it right: I am not an idealist, but hold instead that non-human entities are also mediated in their interactions by sensual forms. It can be compared to Kant's doctrine of the noumenal, but without restricting the noumenal (as Kant does) to a sad byproduct of human finitude. No direct contact is possible between any two real objects or any two sensual objects.
DeLanda: Before we continue drawing out the differences between neo-materialism and OOO, let's focus on what we have in common: our shared rejection of anti-realism.