Whilst (thanks to Harry Stack Sullivan and Stephen Mitchell) the phrases 'interpersonal psychiatry' and 'relational psychoanalysis' already exist, the term 'relational psychiatry' doesn't yet have an established use. Our question then becomes not 'what does 'relational psychiatry 'mean?' but instead 'what's it reasonable and useful to use these words for?' In what follows I distinguish three different uses. My point in doing this is primarily to urge that they not just be folded up together. There's nothing incoherent about the idea of being relational in all 3 senses. But we become mired in disreputable conceptual confusion as soon as we mistake the reasons for believing in one form of relationality as reasons for believing in another.
1) Relationality as articulating the locus of psychopathology and apt treatment
Consider the difference between these two sets of treatments:
i) Drug treatments, ECT, DBS, neurosurgery, hypnosis, cathartic psychotherapy. Whilst questionable in practice (think e.g. of compliance and expectancy effects), an intelligible assumption is that these treatments work, when they do, in a way we might describe as 'from the inside out'. In play here is the idea of something like a localised underlying fault. The aim is to treat this, release the patient from their 'inner' disturbance, so they'll be able to freely flourish in the world again. What is appealing about this idea of psychopathology and intervention is the way that it promises a simplification of complex clinical pictures and treatments. A whole wealth of symptoms can now be explained in terms of a single underlying cause, and it's by treating the single underlying cause that the whole wealth of symptoms can be quieted. Medications could of course be prescribed not to treat singular underlying biological causes of psychopathology, but to intervene in a psychopathological process of whatever sort. Even so we can readily understand the rationale of using medication to treat a condition by treating its putative underlying neurological cause.
ii) Occupational therapy, family therapy, behaviour therapy, therapeutic communities, open dialogue approaches. An intelligible assumption about these treatments is that they work 'from the outside in'. That is, they aim to re-set the patient's relationships with the physical and personal world by intervening in the context of these relationships. The idea is that practice, or repeated exposure, here will lead to the laying down of healthier dispositions and forms of self-regulation.
In relation to these two visions, a 'relational psychiatry' might be understood as being in the business of reminding us just how autopoietic we are. That is to say, it may remind us that we really do enact - i.e. ongoingly constitute - ourselves in our relationships with our worlds and with others. These enactions may be of both healthy and mentally disturbed casts of mind. Now, talk of causality which presupposes that all causes are of the 'underlying' form, and which thereby ignores the systemic aspect of state formation and maintenance, is rife in such psychiatry as is bioreductionist. And because we habitually think in terms of linear causality, we can miss this circular self-creating aspect of human life. And so we can plump for treatments of type i) when treatments of type ii) - which aim to alter our enactions rather than alter their only putative underlying, inner, causes - may be more effective and less damaging.
Here I want to point out that simply acknowledging that many aspects of human life are permeated with what we could call 'relational causality' doesn't mean that all aspects will be so permeated. For example, certain genetic disorders could inevitably come on line at a certain point in someone's development, inexorably unfolding into a clinical picture. You can't philosophise your way to the irrelevance of linear and 'inside out' causality in human life. All you can do is point out the fallacy of assuming that 'underlying' causes are the only relevant causes in the psychiatric domain. I'd also like to point out that other rationales than the treatment of underlying neurological causes can be offered for why we might use, say, a pharmacological treatment. For example we might think that depression, say, is caused and maintained by a patient's interactions with their environments, and use a medication to help lever them out of one self-maintaining mode of interaction into another, happier, self-maintaining state.
2) Relationality as an anti-monadic philosophy of human existence
Consider next the following two visions of human mindedness:
a) On the first view, to have a mind is to enjoy thoughts and intentions and beliefs and perceptions - and these are all inner entities or inner processes. We infer to them in others by examining others' outer behaviour. We look inwards at our own inner states to find out what they are. These inner states actually cause our outer behaviours and expressions. As well as being hidden away behind our behaviour, and so being inner in that sense, they may also be inner in the literal sense - consisting in states and processes of the brain that's inside our skull. To the extent that the outer world and other people are involved in our inner mental states, this is merely by way of providing these states with content through their causal relations to the inner ('functionalism' and 'identity theory' in the philosophy of mind are examples of this vision).
b) On the second view, to have a mind is to enjoy cognitive (thought-involving) and conative (will-involving) capacities. These are not helpfully thought of as intrinsically 'inner' or 'outer', although we can sometimes hide them (so in that sense alone they can sometimes be called 'inner'). To believe or intend or perceive or think is not to have inner states or processes in play, but to enjoy particular intentional relations to our life worlds. ('Intentional': not 'intended', but rather: 'replete with intentionality or directedness'. When I think of you, you are the 'intentional' object of my thought.) Not being entities, beliefs etc. don't have constitutions and so aren't helpfully thought of as 'identical with' brain states. Instead they're more like sets of dispositions to engage in certain verbal and non-verbal actions. We don't need to look inwards to find out what we ourselves think or intend; in fact we're not usually in the predicament of needing to find that out: instead we can just avow or express, rather than report on or express judgements about, our thoughts and intentions. To have a mind, we can also say, is to essentially be in relation to others. We aren't monads who merely contingently happen to have relations to an 'external' world. Rather, and to borrow the terminology of the existential phenomenologists, our existence is characterised by being-in-the-world and by being-with-others - where the hyphens are designed to show how we're (to now use another one) 'always-already' in relation to a world and to others.
So, might we say that a 'relational psychiatry' is one that is committed to the image of humankind we find in b), and that a non-relational psychiatry has got stuck with a)? Well, we can say whatever we like - but it might not be useful. Daseinsanalysis in the hands of Medard Boss, for example, pitted Heidegger's relational conception of human mindedness against the sometimes rather non-relational vision served up by Freud. But the clinical boons of this reconceptualisation were not vast; reading his work we're treated to long animadversions on Freudian concepts but little by way of novel psychopathological or psychotherapeutic insight. Furthermore, whilst psychoanalytic concepts (transference, defence mechanism, projection, primary process, etc.) often have a non-relational (and objectifying) 'inner mechanism' twang about them, this tells us nothing about the use to which they're put in clinical practice. And it's in that use, rather than in their twang, that their meaning and value lie. (I've critiqued elsewhere John Heaton's far more recent critical evaluation of psychoanalytic theory along similar lines.) In sum, because the idea of a 'non-relational' psychology or psychiatry is, on this understanding of 'relational', basically just a nonsensical muddle about human existence, it's not really clear that talk of a 'relational psychiatry' amounts to anything much either. And those who hope to reap rewards for clinical practice by cutting down the confusions that clinicians can get into in reflective moments, whilst ignoring how the clinicians are actually practicing, are after an easier win than is truly available.
3) Relationality as characterising the treatment medium
Consider finally a third sense of 'relational':
This is the one contained in Stephen Mitchell's notion of 'relational psychoanalysis'. Paramount here is the idea of the clinical relationship as itself (part of) the medium, and not merely the vehicle, of change. Sometimes one finds a clinician defending, say, cognitive therapists against the charge that such therapy ignores the importance of the therapeutic relationship. As Judith Beck writes, for example, cognitive therapy ‘requires a good therapeutic relationship. Therapists do many things to build a strong alliance. For example, they work collaboratively with clients . . . ask for feedback . . . and conduct themselves as genuine, warm, empathic, interested, caring human beings.’ However, as psychoanalytical psychotherapist Jonathan Shedler (caustically yet aptly) responds, ‘This is the kind of relationship I would expect from my hair stylist or real estate broker. From a psychotherapist, I expect something else. [Beck appears] to have no concept that the therapy relationship provides a special window into the patient’s inner world, or a relationship laboratory and sanctuary in which lifelong patterns can be recognized and understood, and new ones created.’ Shedler’s optimism regarding his hair stylist and estate agent perhaps warrants some 'cognitive restructuring', but his point about the therapeutic relationship stands. What a relational psychoanalysis does is use the therapeutic relationship itself as the medium of change - rather than merely form a good relationship which facilitates an independently intelligible therapeutic process. Part of this involves developing such an alliance with the patient as can itself be internalised into a healthier self-relation. Into a self-relation, that is, which intrinsically undoes and wards off such psychopathology as stems from a harsh, unforgiving, lonely-making, self-relation (i.e. stems from the 'superego', 'toxic introjects', 'ego-destructive internal objects', etc). But in the psychoanalytic context one will be expected also to carefully attend to the transference dynamic manifest in that relationship. The closeness of the therapeutic relationship also potentiates the transference in both its negative and positive aspect. Relationships which keep in the polite and friendly zone are not going to allow the patient's underlying fearful expectations to be truly experienced and challenged. When the transference relationship does become the medium of therapeutic change, however, the idealising feelings toward the analyst which guard against unconscious expectations of being hurt or let down or judged can also be made clear. This all makes for a genuine experience - not only of a trust and loving acceptance that can be internalised into a sense of self-worth, but also of profoundly difficult latent expectations of that sort which drive mental illness and which can finally be deconstructed.
As regards a relational psychiatry, I note that although in the UK all trainee psychiatrists are required by the GMC to be able to develop therapeutic alliances, and to have some familiarity with psychoanalytic concepts, they aren't required to have expertise in psychotherapy, let alone in working with or in the transference. Unless we're promoting a radical change in psychiatric practice - in the direction of contemporary psychoanalytic psychotherapy - it won't do to borrow Mitchell's meaning for 'relational' to explicate 'relational psychiatry'.
At this point I should perhaps own that I'm rather 'relationally' minded in all 3 of the above senses of 'relational'. What I want to stress here, however, is the absence of any necessary connections between the 3 above-described forms of relationality. I want to stress this because I think that a clear danger of talk of 'relational psychiatry' could be that the term becomes woolly, and unwarranted support for one of the relational notions above gets drawn from one of the others. Such illegitimate arguments would quickly invalidate the approach.
To spell it out, we may (rightly or wrongly) think that serious psychopathology is sometimes sustained by certain largely unconscious dismal beliefs about self-in-relation-to-other. (Perhaps I have a largely unconscious habit, for example, of expecting others to be dismissive about my concerns - so I tend not to let either myself or the other know about them.) And these beliefs may be considered to be sustained just by default rather than through any activity on the subject's part - in the same way that the reason the colour of the living room curtains remains self-same day to day isn't because anything is actively maintaining it, but rather just because nothing comes along to change it. In this sense, then, we don't here have a relational conception of psychopathology in the sense outlined in 1): the depression isn't here considered as autopoietically or systemically maintained. Perhaps you'd have to really dig out such underlying beliefs with psychoanalysis. Yet this psychoanalytic therapy might well be relational in sense 3): it might require a canny working in the transference before this patient's depression is going to shift. And perhaps if this therapist has got interested in philosophy, his underlying vision of what it is to have beliefs is relational in sense 2). Nothing, however, dictates that this shall be the case: he may work in the transference whilst maintaining a psychopathological theory which deprecates the idea of an ordinary ongoing interpersonal enaction of psychopathology - seeing it instead simply as something which manifests in his patient's life, bodying forth relentlessly from 'underlying' disturbances in self-image - and he may moonlight as a Cartesian philosopher who eschews a conception of human beings as constituted by their relations to the world and to others. Constitution is not necessarily causation: We may maintain the significance of causal relations between self and world whilst deprecating the idea that we are constituted by our world-engagements. Or we may have a fully relational (in sense 2) conception of our existence - i.e. see what we are as constituted by our relationships, rather than as monads whose minds consist in a pure interiority - whilst maintaining the value of a pharmacological treatment which aims to causally alter our mood state. Perhaps, as suggested above, the drug treatment could tip us into a different level of self-sustaining equilibrium (i.e. affect our relational nature in sense 1), and this new manner of world-relation be understood (in sense 2) to genuinely constitute a new form of mindedness.
To end: is there nothing which binds together these different senses of relationality? Well, consider that you're attracted to a monadic (non-relational in sense 2) conception of human existence. You're also not a dualist, let's say, but - despite not thinking that mind consists in immaterial spirit - you're nevertheless attracted to the idea that mind must at least consist in something. Understanding as we all do that the brain is rather important for mental functioning, and not being persuaded by the 'extended mind' theorists (who take mind to consist not only in internal but also in environmental states and in the interactions between these), you tend to think your mental states consist in your brain states. And, unlike most materialists, you're also not a functionalist (i.e. you don't think these states are individuated by their worldly causes and bodily effects), and so you're moved to think them individuable without reference to anything outside your head. With that conception of the mind in play, what might you now think about mental illness? Are you likely to think it non-relational in sense 1) as well? I think many people are intuitively inclined to answer 'yes' to this - but, speaking for myself, and following the logic of the argument, I just don't see it. For surely you could either think that depressed or delusional mental states, which allegedly 'just are' brain states, are self-maintaining underlying causes of the depressed or delusional actions they inspire. Or you could think take these inner brain states to arise and be causally maintained by the interactions in which the person engages. And so even whilst cleaving to the most fully 'materialist' and 'internalist' conception of mind we can imagine, there's nothing here which ought to make you generalise from your non-relationality (in sense 2) about persons to a non-relationality (in sense 1) about psychopathology and treatment. You might well do this, because like many of us you get muddled about what's entailed by what. The relational psychiatrist, however, is not required to follow you in that muddle.
Richard Gipps offers broadly psychoanalytic psychotherapy and is an associate of the Faculty of Philosophy, University of Oxford. He is the web secretary of the International Network for Philosophy and Psychiatry and blogs at Philosophical perspectives on clinical psychology and the pain that breaks.