Information, Access and Reality: A Response to Harman

Graham Harman at Object Oriented Philosophy has responded negatively to my claim a) that his interpretation of the tool-analysis makes makes idealist assumptions and b) that intentional states such as perceptions and beliefs have real things as their intentional objects:

What is idealism is enemyindustry’s own next sentence: “In contrast, I hold that intentionality brings us into contact with the real with numbing regularity.”

This is idealism, because it holds that the real is convertible into the accessible. It gives no adequate account of the difference between the tree that grows and bears fruit and the tree that I encounter. No matter the level of “numbing regularity” with which I encounter a tree, that encounter is not the tree itself. Until you account for the difference between the two (as I do) then you are an idealist.

Ultimately, I think this is why Meillassoux remains in the Idealist camp, and the same holds even more for the Sellarsian scientistic wing of what used to be called speculative realism. They aren’t realists. They’re partisans of math and science.

Harman’s repudiation of speculative realists like Meillassoux or Brassier who cleave to some form of scientific realism is founded on the claim that insofar as X can be ‘encountered’ or ‘accessed’ or otherwise represented, X is not transcendent enough to be real. For the idealist objectivity is constituted by access conditions; for Harman, reality is constituted by its inaccessibility (to thought, perception or declaration, say) .

This account of the real-non-real distinction has the virtue of clarity, I suppose, but the epistemological problems it generates are considerable.

For example, if the tree encountered is not the tree that grows and bears fruit, the two are different entities.

So how does information about the encountered tree – that it is a cherry tree, not an apple or pear or quince – bear on the real tree? If it does – if my encounter with the encountered tree produces information about its real correlate – I surely encounter the real tree as well.

But since (according to Harman) I cannot encounter the real tree, my dealings with the encountered tree convey no information about the real tree. But then what entitles Harman to say of the real tree that it ‘grows and bears fruit’, or even that it is a tree, or living system of any kind?

If we can glean no information about the real, we are not entitled to apply to it typologies which apply to the things we can glean information about. Object Oriented Ontologists often illustrates their liberality with compendious lists of things to which their position is committed – dogs, black holes, mitochondria, clowns, etc. But if the real is informationally closed, the Object Oriented Ontologist is not entitled to say that these types belong to its domain. We have no informational hook up with the real as opposed to the encountered.

With informational closure the profligacy of OOO is spurious. Without it, OOO is incoherent.

Harman could, at this point, deny that having information generated by the real suffices for access. If we understand ‘information’ in a mathematical sense as an additive quantity representing the reduction of uncertainty produced by an event, he would be right. There is nothing remotely ‘semantic’ about this notion. However, semantic concepts of information – such as those used in causal-covariance theories of representation – are available too. They are afflicted with problems (e.g. how do we get from ‘Event A has information about [causally covaries with] Y’ to ‘A represents Y’?) but they help theorize the capacity of living systems to discriminate features of their environment. In short, semantic-informational access, seems like a candidate for the real, however problematic our theories of it may be.

If OOO was the only workable metaphysics on offer, then perhaps we should, like Kant, learn to live with the thought that the the real is informationally closed as such, that none of our empirical typologies can apply to it. But I think a more nuanced picture is available which retains a conceptual linkage between notions of reality and notions of epistemic independence.

It is hyperbolic to claim that since Y can be discriminated or accessed, it is not real. This is because it is not necessary for Y to be mugged or traduced into something else (Y*) for it to be discriminated via information bearing states of some kind (thus accessed). Informational access, at least, does not require this.

We can (consistent with this picture) deny that Y being real entails its accessibility to S, though (where ‘S’ ranges over those systems, human or otherwise, for which access is at issue).

While we are regularly in contact with some features of real things, there are other equally real features that have eluded us until recently. There may be intrinsic properties of things that elude discrimination in principle. Thus it is not necessary to hold that reality constitutively debars access, only that being real, of itself, does not entail it.

7 thoughts on “Information, Access and Reality: A Response to Harman

  1. Good stuff. The idea that you can define idealism as ‘hold[ing] that the real is convertible into the accessible’ is just bizarre. I don’t see how this is either a necessary or sufficient condition for idealism, but then I think the OOO people say that anything they don’t agree with is idealist.

  2. Thanks for the response, Jon. For a while, I’ve been bothered by the alacrity with which OOO fans advert to the ontology of science to explode the myth that OOO is committed to a metaphysics of common sense, while developing a position that precludes scientific realism – or realism as an attitude towards any theoretical domain. At least Graham Harman is an avowed and consistent anti-naturalist, which I can respect.

  3. The strangest argument along these lines is Levi’s attempt to use Bhaskar’s argument to establish his own version of the withdrawn ‘proper being’ of objects in contrast to their apparent ‘manifestations’. This is because he identifies the former with generative mechanisms, and uses the success of scientific inquiry into such generative mechanisms as a premise in his argument, effectively trying to use the success of science in uncovering the underlying reality of things to prove that this very underlying reality is inaccessible to it. Of course, this is just the gist of the argument one can glean from his blog, and we’ll have to wait for his forthcoming book to see if the full version makes more sense.

  4. Hi Pete.

    I’d also noticed some discussion of Bhaskar on Levi’s site. I’m also interested to see whether he can finesse the kinds of objections raised here. I’m inclined to think, however, that the objection sketched out here is absolutely general.

    Whatever basic epistemic relation the OOO proponent adopts to describe our transaction with withdrawn objects (e.g. translation it will either involve the production of information or it won’t. If it doesn’t then it can’t furnish the conditions for knowledge, if it can, then the thing translated is encountered and is thus not withdrawn.

  5. David, you wrote:
    it will either involve the production of information or it won’t. If it doesn’t then it can’t furnish the conditions for knowledge

    There is another option – information is gleaned from the object in a way which cannot be boiled down to the sort of epistemic relations you are favouring. Or to put it another way, OOO is really a form of mysticism with ‘speculation’ taking the place of revelation.

  6. Jon, I concede that my argument is extremely schematic as things stand.

    I haven’t provided a theory of semantic information or an ontology of information in general – say of the kind of Lucio Floridi is developing (

    More needs to be said about the relationship between purely informational access to objects and other kinds of access (e.g. full blown phenomenological access). I think my strategy here is to argue that phenomenological access and epistemic access both presuppose informational access even if they must satisfy constraints that informational access need not satisfy. So informational access need only be a necessary not sufficient for knowledge in order to be epistemically relevant. I suppose the bottom line is that having a state S bearing semantic information about X grants access to X though S and X are not the same. This is a pretty weak assumption, I think.

    Gleaning information about an object seems to come within the general structure of access relations as far as I can see. I guess the burden of proof is on the OOO proponent to show why gleaning information about X never suffices for access to X. The only way through here is to assume that access is necessarily unmediated – e.g involving direct acquaintance or some such. I think there are lots of good reasons to reject this basically Cartesian model of phenomenology.

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 )

Google photo

You are commenting using your Google 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