A very interesting post on the great Imperfect cognitions blog. Wolfram Hinzen takes a linguistic view of delusions. As someone who has ventured into delusions and particularly insight from a discourse analytic point of view, I welcome the linguistic take. Except that I have some problems with what Prof. Hinzen wrote.

Let me start with summarising briefly what I see are the main points of the blog (which is based on a larger paper). The way I understand it is that Hinzen suggests that delusions be seen as propositions which might be assessed on the truth-falseness spectrum. Delusions would be propositions which are false. The author continues suggesting that such propositions must, of course, be seen in the context of their occurrence. For example Hinzen talks about a frame in which propositions are used.

The problem I have is that language doesn’t work like that. People actually don’t set out to make (or is it ‘issue’?) propositions that clinicians will then assess and decide whether the proposition counts for a delusion or not. I am not even certain that you can say ‘utter a proposition’, as the author says. And so, while Hinzen usefully (for him) describes different kinds of propositions, he does not consider questions, promises, warnings. Do they count as delusions? Yes, a question is what we, discourse analysts, call ‘propositional content’ (all that which conveys lexical and grammatical meaning), but it is certainly not a proposition. So, my first question would be: can a question be “delusional”? Would what is implied or presupposed have the same ‘delusional’ status as what is explicit? The author does not stop to consider such possibilities.

But the problem is that people actually do not only simply utter propositions, but they tend to tell stories. And when get to stories, things get way more complicated. I described such problems in my article on insight. So, consider this story (which I take from the article):

I had something like that for example I was sitting at home and I heard behind the wall various voices, and I had an impression at the time that I am [past perspective] being controlled by someone and there are [past perspective] persons behind the wall specifically to control me or in some way, or for example they could [present perspective] have a device with which they control [past perspective] my behaviour, influence [past perspective] my behaviour

Basically, in this story the speaker shifts from what could be seen as having insight to lacking it. When he takes the ‘past perspective’, he looks at his ‘science fiction’ (as he put it) with his eyes at the time, it suggests insight, when he looks at it with the present perspective, it suggests lack of insight. And here we have a question. Is the following snippet:

they could [present perspective] have a device with which they control

in the particular context delusional or not? If we take the ‘propositional’ approach suggested by Hinzen, it is. He clearly constructs his neighbours as people who in the past could have a device controlling him. He doesn’t distance himself from such a proposition. And yet, I would argue that we should actually see not the proposition, but the whole story and not jump to conclusions! Such examples where stories go well beyond ‘propositions’ can be found very easily, some of them in the article I have just quoted.

But there is another problem I would like to raise. I am certain it is clear and obvious both to Prof. Hinzen and to anyone else who came in touch with patients with delusions, and yet, I think, it is worth mentioning.

Prof. Hinzen shows us how to assess delusional propositions, giving, for instance, an example of a delusional proposition:

I am Jesus.

And yet, he does not explain at all how he knows that the person saying this is actually not Jesus. In my view the issue is not trivial at all, as it raises the more general issue of how we assess ‘propositions’. To show my problem, I’ll tell a story.

When I was an intern in a psychiatric hospital (here I described my experience), I met a patient admitted after a suicide attempt, and with the diagnosis of psychotic episode. It was determined that she had tried to kill herself acting on delusions of her husband’s affair (well, meetings for sex) with their neighbour. I enquired how the doctors knew that the husband’s infidelity was a delusion. The doctors told me that they had explained how crucial it was for him to tell them if, indeed, he had been having sex with the neighbour, assuring him that they were uninterested in passing judgement. The man repeatedly denied, leading the doctors to believe that he might actually be telling the truth.  This is how the man’s infidelity became a delusion.

Except, it wasn’t. About a couple of weeks later the man took pity. He saw his distraught wife who continued to be ‘psychotic’. And he finally admitted to be rather economical with the truth. Momentarily, the woman’s delusions vanished.

Let’s first consider the propositions, as Prof. Hinzen suggests. It all boils down to a proposition like this:

My husband has been unfaithful.

The proposition is deceptively simple, until you actually start considering how we might know whether the husband was or was not unfaithful. As hiring a private eye (retrospectively) is unlikely, what we rely on is, realistically, the statements of the husband and, perhaps, some other members of the family. How do you confirm either party’s story?

But that’s not all. The problem with such an approach is that it is hardly reflective how we communicate. And so

My husband has been unfaithful.

is never uttered as a proposition. Moreover, I doubt very much such a statement would be uttered on its own or even in such a way. Can we really imagine a person who just tried to kill herself, simply and calmly saying that her husband had been unfaithful? Somehow, I would expect a few expletives, for example.

Moreover, whatever she did say that was whittled down to the above proposition, would have a number of other functions. For example, it is likely to be a justification of her suicide attempt, probably also an accusation of her husband, or an indictment of her failing relationship. Or perhaps all those things at the same time, together with some others.  If we understand the dramatic contexts and the communicative function of such utterances, we will understand that the truth-conditionality of the statements is not exactly a priority. We will also understand that the husband was not asked to  provide a competing proposition, but an admission of guilt! And this a very different thing altogether.

And here we come to core of the problem. In order to understand delusions, you must understand them not as propositions, but as communicative actions with particular functions and in a particular cultural, social and individual context. I would even suggest that you could even imagine a situation in which

The bastard cheated on me.

could be considered ‘delusional’, despite the fact that the bastard might actually have done. In other words, I would suggest that the ‘uttered proposition’ might in fact be true (surprisingly, what I write here might even make sense in the new definition of delusions in DMS5), and still might count a delusion. It’s all because delusions, I would suggest, are not so much linguistic, but they are discursive. In other words, delusions are situated, always performed here and now with a particular communicative goal. How it happens is for another blog.