How do we know that we know what we’re talking about? It is the ability for our language to refer that enables us to talk about things, about actual objects in the external world. A language that refers picks out things in the world and grounds our speech in reality. But what exactly is it for language to refer? In this short essay I will propose some conditions for reference, and question whether we meet these conditions with a familiar thought experiment.
The first condition for reference is intention. In order to refer to something, I must be trying to refer. Saying words or producing symbols at random is not sufficient to say that someone is referring. Suppose I buy an enormous bag of lentils. I carry them home but, unbeknownst to me, there is a hole in the bottom of the bag, and they slowly spill out, leaving a trail of lentils behind me. My route home from the lentil emporium is convoluted, and by the time I am home I have left a complex trail of lentils which, when viewed from above, spells out “Isaac Newton.” I did not mean for my lentil trail to refer to Isaac Newton; I didn’t intend to leave a lentil trail at all. There was no trace of Isaac Newton in my mind. It is clear that in this peculiar case I did not refer. There are myriad other cases, situated at all points of the whimsical-realistic spectrum, that illustrate this same point: intention is necessary for reference.
The second condition is causal relation. My words can refer to something only if they stand in the relevant causal relation to it, that is, if the thing was causally efficacious in my coming to say the words. Information about Isaac Newton trickled its way down through time to me. The appropriate causal chain is in place for me to be able to refer to Isaac Newton.
Imagine a reclusive painter, ignorant of scientific history. He doesn’t read books or talk to anyone, but spends his life in his shed painting portraits of people from his imagination. He has never heard of Isaac Newton the polymath. One day he paints a portrait of a man in 17th Century garb, with long hair and chiselled features, which looks exactly like Isaac Newton. The painter decides, on a whim, to title the portrait “Isaac Newton”, because he likes the way it sounds. When the painter says “Isaac Newton”, he refers not to the real Isaac Newton, but to a painting (or an imaginary person depicted in a painting). It is a painting which looks exactly like Isaac Newton, and so the image he has in his mind as he speaks is just the same as the image someone referring to the real Isaac Newton would have, but nevertheless he is not talking about the real Isaac Newton, but about a picture. The reclusive painter cannot refer to the real Isaac Newton, because he does not stand in the appropriate causal relation to him. The real Isaac Newton played no part in the painter’s speech.
The third condition for reference is resemblance. In order to refer, the image I have in my mind when I am trying to refer must roughly match the object. If I have in my mind an image of a large grey animal with flappy ears and a long trunk whenever I say the words “Isaac Newton,” then I am mistaken about what the words mean, and have failed to refer.
I hope I have demonstrated that there are three conditions for reference: intention, causal relation and resemblance. The question remains whether these conditions are met and whether our language refers. Putnam discusses a popular thought experiment which shows that we cannot be sure that we meet all of the conditions for reference.
Imagine that you are a brain suspended in a vat of fluid, wired up to a powerful computer. The computer titillates your brain such that you have sensory experiences of cars and trees and philosophical journals just as if they were really there. The brain has the same perceptual experience that you do now, even though there are no objects in front of it. If we are not brains in vats, then objects in the external world are causally efficacious in our coming to talk about them, but this is not the case for the brain-in-a-vat (BiV). The BiV does not stand in the relevant causal relations to external objects to be able to refer to them. The BiV has the same intentions and mental images when talking “about” objects as we who interact with the external world do, but the BiV is like the reclusive painter, and has no actual experience of the objects that it seems to be referring to. Because the BiV does not stand in the correct causal relation to the objects in the external world, it cannot be said to refer to them.
Putnam claims that our language does refer, so we know that we are not BiVs – but do we know that our language refers? The language of the BiVs (vat-English, as Putnam calls it) is identical in all respects to English except that it does not refer. The BiVs have the same perceptual experiences that we do in the external world, so if we have a ‘feeling’ of referring, they would have a feeling just the same. Although English and vat-English are different, they are indistinguishable from the point of view of the subject, and so if we were BiVs, we would not be able to tell just by looking at our language. It follows that we do not know whether our language refers to anything in the external world. It’s quite possible that you don’t know what you’re talking about, and neither do I.
Sarah Ayliff is a 3rd year Philosophy student at the university of York.
Putnam, H. (1981). Reason, Truth and History. Cambridge: Cambridge University Press