As Hilary Putnam makes clear in his seminal paper Brains in a Vat, we obviously cannot be ‘brains in a vat’ (BIV) in the strict sense. We have experiences with brains and vats, and if we were in a situation comparable to the colloquial understanding of the ‘brains in a vat’ paradigm, then those ‘brains’ and ‘vats’ would not actually be brains and vats, because, by definition, ‘brains’ and ‘vats,’ as we typically understand them and can refer to them, are concepts that reside exclusively in our reality. If they were phenomenologically metaphysical, Putnam suggests, then they wouldn’t actually be brains or vats. Putnam is correct in that a brain in a vat would be linguistically challenged to adequately describe its situation, and certainly that it would be unable to know definitively if it actually is a brain in a vat. But Putnam is making more than this mere pedantic argument. He also suggests that a BIV, lacking reference, would be entirely unable to make any accurate claims about its real situation, as it could only reference the experiential reality in which it has direct causal connections, and he suggests therefore that any notion of a BIV-type, matrix-esque situation is self-refuting.
A Brain in a Vat Believes It is Walking, Alexander Wivel, 2008.
I believe Putnam is wrong about this. In this paper, I will argue that Putnam’s theory of causal reference is incorrect, because minds have the capacity to make accurate claims about phenomena and objects with which they are not connected though any direct causal chain. I will argue, in short, that all references we make are indeed dependent upon prior causal experiences, but, contra Putnam, we can nevertheless employ prior causal experiences to develop concepts which enable us to refer to things with which we have no direct casual experiences. This is not entirely dissimilar to Putnam’s causal theory, and I believe avoids the notion of ‘magical reference’ axle around which Putnam gets so wrapped, but still differs significantly from Putnam’s’ construct and would potentially allow BIVs to refer to their situation using meaningful language. Additionally, I will make the point that our inability to reference something does not preclude its existence.
Before diving into the argument, I would like to pursue a brief tangent about my use of the word “metaphysical.” If you are a brain in a vat, those brains and vats are not objectively metaphysical-- they are just regularly physical. Rather, it is your experienced reality that is ‘less than physical’ (infraphysical). However, from your perspective, the brains in a vat would be phenomenologically metaphysical, in that they would be positioned outside of the physics that you perceive and with which you interface, and would therefore, relative to you, be ‘meta’-physical. This is like when an American labels Germans "foreigners." From the subjective vantage point of the American, that label is meaningful and self-referentially accurate, but from the vantage point of Germans, it is incorrect, and in relation to the wider universe, Germans are of course not objectively "foreigners," just as the brains in a vat are not objectively ‘metaphysical.’ Therefore, when I deploy the term ‘metaphysical’ in this paper, I do not intend it as an objective claim but rather as a claim relevant to the subject in question, much like if I said “Michael drives a foreign car.” The car is not objectively foreign, but, for Michael, it is indeed subjectively foreign.
That important clarification aside, we can begin to tackle Putnam’s causal theory of agency. While Putnam acknowledges that we can refer to things which have “very strange causal” relationships, Putnam insists that some traceable causal connection is still required for reference. In fact, in some instances, he mentions that even “weak connections can hardly suffice for reference.” Putnam is wrong about this–As I hope to demonstrate, ‘indirect causal connection’ is sufficient for reference, insofar as we can use prior causal experience to develop concepts which allow us to refer to things with which we have no direct causal chain.
Amongst a series of thought experiments that I will present, we will begin by talking about the future. The future does not yet exist, and even if the universe is deterministic, we lack the ability to treat the universe like a movie that we can condense into a VCR and fast forward– not only is this technologically unfeasible, but it would also create a host of theoretical problems such as what would happen if we attempted to chart a different course than that which the VCR-future detailed. Causality cannot work backwards in time, and although we may become causally linked to the future when it becomes the present, we are not causally linked to the future now, because it doesn’t yet exist. And yes, while we have had experience with anticipated futures becoming reality (e.g., I anticipate the exam will be hard, and it is, thus I have developed an understanding of the concept of a future), we do not have any direct causal connection to the future.
And yet, I can still meaningfully refer to it. I can say “The 2024 Madrid Olympics are going to have a beautiful closing ceremony,” and you could correct me and say “You are wrong, the 2024 Olympics will be in Paris.” In this case, we are referring to something with which we have no prior causal connection; we have concepts of Olympics, we have concepts of the articulated plans for future Olympics, but we are not saying “The IOC has declared their intention to host the 2024 Olympics in Paris.” Instead, we are making the separate claim that the 2024 Olympics (which do not yet exist), will be hosted in Paris in 2024 (which has not yet happened). This claim, of course, could turn out to be wrong, as can claims about things that we have direct causal experience, but it can also be accurate and we can still meaningfully talk about this anticipated future without having a direct causal link to it.
Similarly, Putnam claims that an alien on a foreign planet who has never seen a tree could not possibly refer to trees if he only saw a drawing of a tree, because he has no casual connection with trees. Putnam is wrong about this and the reason he is wrong highlights a problem in his strict causal theory. I agree that the alien seeing the picture of a tree would not necessarily see a representation of a tree. But it is possible that it could, even without having any causal connection to a tree. For example, even having never had any prior exposure to trees, the alien might look at the picture and think: “Wow, I bet that picture is a drawing of an organism from a different planet, it would make sense to have a canopy like that to efficiently perform a photovoltaic process like our solar panels do! I bet it also has rigid cell walls characteristic of [the alien equivalent of mold].” This would be a meaningful and true statement, and although the alien has an extremely limited understanding of trees, he can still talk about them accurately in some contexts, even if he doesn’t know the word “tree” or know much about trees. It is even theoretically possible that the alien might say, “I bet people on a foreign planet live amongst rigid-cell organisms that look like that drawing, and I bet they and call them ‘trees.’” While this scenario may be extremely improbable, it is not strictly impossible, and Putnam’s claim about the necessity of causal connection is simply incorrect.
These concepts can even be dependently arisen. Say, for example, in an incredible coincidence, you and I have the exact same dream one night in which we are at some alien temple complex and an alien there tells us the complex is called the “Glork Complex.” Now, theoretically, I could run into you on the street the next day and say, for no particularly good reason, that: “The Glork Complex looked really cool because it had blue walls.” There is absolutely no shared casual history between our understanding, and yet, although you might be baffled at my statement, I would still be referring to something meaningful for you, and we could talk at length about the nature and appearance of the mutually-intelligible concept of the temple complex, of which we have no shared causal history.
In these examples, there is indeed an important element of causality at work, and I am not trying to do away with importance causality altogether. But it is not the direct chain of causality that Putnam imagines. Instead, in both our references to the future, and the alien surmising correct facts about a tree, minds are able to weave their prior causal experiences into concepts which they can use to accurately refer to things with which they have had no direct casual contact. This is so integral to what we do as humans. Modern astronomers and physicists surmise that there are galaxies moving away from us faster than the speed of light, and thus we literally can never have any casual contact with these galaxies. But, drawing upon the repertoire of concepts that we have developed from prior causal experiences with both our galaxy and physics, we can develop concepts that allow us to talk about these distant galaxies, and make potentially accurate claims about them. I can claim, for instance, that they are host to aliens. I might be right or wrong, but I am still referring to something which we have absolutely no causal connection with and yet I may be correct.
There are certain things, of course, to which we cannot refer because we have no causal background from which to construct intelligible concepts. For example, although electromagnetic wavelengths exist in real life, the color that we ‘see’ when we look at something ‘green’ is an artificial construct of our mind. If an alien saw something ‘green,’ even if it could (unlike certain animals) only process light from the same part of the spectrum as us, the alien might visualize a color entirely unique to his mind– we might see something very different when we both look at something ‘green.’ We would be forever unable to refer to our unique visualizations of the color green to each other, because there are not enough casual elements to construct an intelligible concept. It is similarly impossible to comprehensively refer the color ‘yellow’ to somebody who has been blind their entire life. However, although we cannot fully refer to the color in question, we can build upon prior causal concepts to understand that the other party is seeing a unique color unintelligible to us, and talk about that mysterious color in meaningful (e.g. ‘to me this color is phenomenologically beautiful’), if highly limited, terms. The point being, some casual history is certainly necessary, but I just don’t believe it has to be as direct as Putnam suggests.
The same goes for BIV-type scenarios. The English language might ill-equip us to make accurate and meaningful claims about a BIV scenario, but if we are brains in vats, it is not impossible to refer to them when describing our situation. We can apply concepts from things that we do have prior causal experience with, such as our brains and vats, and develop from them new concepts that allow us to have meaningful conversations about a BIV-type scenario. That’s why I am able to write this paper, and while any comment about I BIV that I make might be definitionally problematic because our language is insufficient, I can still convey an idea to you, and you know roughly what I am referring to. The very fact that you could correct my language and reveal to me why my language is insufficient means I have conveyed the right idea, even if my vocabulary is lacking. For example, we can label the place of the brains in a vat ‘supra-reality,’ and the experiences of the brains ‘infra-reality,’ and posit that we experience an infra-reality-like situation. We can do this without being brains in a vat, and if we are brains are in a vat, we can still do it, because of our ability to scavenge prior causal experience to develop concepts which allow us to refer to things with which we have no direct causal experience. Obviously, such claims are epistemically problematic, and definitionally awkward, just like claims about the future or an alien's claim about a tree, but they can still be meaningful and referential. I have the exact amount of direct causal connection to a hypothetical BIV as I do to a hypothetical galaxy moving too fast away from me to have any causal relation to me. But I can conjecture about the existence of these things, and make potentially accurate claims about them.
This concept theory is really important because it is integral, I believe, to our development of every idea. When the first Greek first thought about Zeus, he created a concept. This concept was certainly contingent upon prior causal experiences (people, thunder, lightning, beards, Mount Olympus, prior stories of Gods, etc.) but he had no direct causal connection to the concept of Zeus– instead, he created something new. Putnam’s theory of strict causal necessity doesn’t leave the door open to the invention of new concepts deriving from things which have no direct causal connection with. I am not claiming that the mind’s processes are ‘magical,’ but the mind certainly has an ability to weave together prior causal experiences to invent new things, without any prior concept, which are useful and can be referred to. These invented concepts may or may not be ‘accurate,’ but they are still meaningful and useful.
Finally, there are all types of things that we can’t refer to that may exist. For example, an alien on another planet might have developed a machine that we, right now, can’t conceptualize and therefore can’t meaningfully refer to. That doesn’t mean such a thing can’t exist. So even if I am wrong and we can’t accurately or sufficiently refer to the metaphysical BIV-type situation, that doesn't mean, as Putnam suggests, that some matrix-esque metaphysical arrangement isn’t occurring or can’t occur. It would indeed be true that we are not dictionary-defined brains in dictionary-defined vats, but our inability to articulate something doesn’t mean there are no metaphysical shenanigans going on. Bacteria are entirely unable to talk about petri dishes. In a very pedantic sense, it is a self-defeating premise for a bacterium in a petri dish to claim it is in a petri dish, because bacteria can’t talk about things or make claims, and if it did talk about being in a petri dish, then it wouldn’t be a bacterium. That’s all true, but nevertheless, petri dishes still exist and bacteria are indeed in them.
The chief objection that I foresee is that I am creating a magical theory of reference to explain how we can refer to things with which we have no prior direct casual connection. I don’t think I’m doing that; I think the mind’s ability to imagine by weaving together prior casual experiences is plausibly a natural process that doesn’t require some ‘spooky’ mechanism. Nevertheless, even if it is a ‘magical theory of reference,’ that in no way invalidates the argument that Putnam’s theory is inadequate. His theory does not explain how we can refer to the future or to a galaxy which can never have any causal experience with, and leads to incorrect claims such as his thought experiment with the tree-ignorant alien.
The second objection I see is that my theory is somewhat incomplete and even slightly vague. To that, I concede fully. I am still wrestling with these issues. Putnam claims it took him several years to develop his theory, and I certainly am not capable of fully flushing out this byzantine topic in the limited time and knowledge that I have. But I hope to have presented some thoughtful and meaty claims that highlight serious theoretical problems in Putnam’s work and which may be worth continued exploration. I do think this issue touches on some very profound issues and warrants further explanation. For example, I think it is very relevant to global theological discourse. People claim to experience ‘revelation’ in a myriad of forms, and perhaps, if that is true, part of the argument for religious relativism is that these experiences can’t be communicated or argued through traditional rational discourse because it is impossible to refer to these potential events and experiences.