Imagine you are at a dinner party and someone approaches with a glass of wine in his hand. You have never met him before. His hair is bleached from the sun, he's deeply tanned, and he is wearing loose fitting, light coloured clothes.

As he joins your group, he steps right on top of what you were saying and starts prattling about the weather in the Bahamas the last time he sailed his yacht down there at this time of the year. He was so bored with the place, having been there several times, and he much prefers the coast of France anyway.

"They're really two completely different places you know." You don't know whether to laugh or scream, and you're sizing the man up to be an obnoxious, narcissistic, waste of time. That is a diagnosis. You don't have to be a psychologist to diagnose people; we do it with each other all the time.

However, when a professional does it in a professional context, for a clinical reason, there is something more than a conversation over a glass of wine going on. It can come in the course of therapy, and it can take place with formal assessment.

What people do over time in therapy results in a felt sense of the other for each one. That is informative, and it can be diagnostic. Recently, for instance, while meeting with a couple, the woman complained that she was always having to remind the man and follow up after him. I listened and began to wonder if the man had an attention deficit that had never been assessed in childhood. I offered dialogically my curiosity and then painted a word picture of a person with adult ADD; the woman's eyes brightened and she said, "That's him!" While understanding a dynamic as an executive dysfunction instead of a lack of caring did not remove the annoying behaviour, it did put a different set of clothes on it.

Whenever I get a referral of a child from a parent who comes with voluminous documentation about what is going on with the child, I always insist on meeting with the child him or herself before interacting with the data from schools and other clinicians, and that is in order to get a "feel" for that child myself. Each person is different, and I want to see what it is like to be with that person. The contact provides first-person experience that contextualises any observations made by others.

Sometimes the need for formal assessment procedures emerges from the therapeutic context, because a clinician realises that something is going on for which he or she would like to find out more. In the course of a dialogical approach, this curiosity and interest in what the results of psychological testing might show can be shared with the client. Conversely, when the process begins with the need for assessment, the results of testing can become quite important therapeutically. For instance, when I do assessments that are formal psychological evaluations to begin with (i.e., referrals from another clinician, an agency, or a department of community mental health requesting a certain kind of psychological evaluation), I always meet with the subject (and parents or social workers) afterwards in order to go over the results, which is not just to satisfy ethical propriety but actually to get a sense of the person's reaction. That reaction becomes part of the assessment because hearing those results affects the client and must be monitored for its significance. There is a counseling piece that accompanies the process of giving feedback on the testing. In that sense, all assessment is, to me, part of an advancing therapeutic relationship.

All this accords with what Lynelle Yingling observed in her source book on the Global Assessment of Relational Functioning scale (a resource available for clinicians to diagnose family systems). She indicated that in assessing, one could choose the outsider, the insider, or the insider-outsider approach.

Outsider evaluation is described as the use of a clinical rating scale by an observer/therapist/team or researcher. Insider methods use client self-report about his or her experience of system functioning. Insider-outsider methods use a more collaborative process whereby the client and therapist develop criteria delineating desired change in a well-defined manner.

Since I tend to work more dialogically, in accord with a gestalt therapy approach, I find myself often working that third option.

Now, if I were at the dinner party, and I had Ms. Yingling's options in mind, I might choose from the three options. What might each look like?

Outsider: The tanned man interrupts me while joining a group of us at a dinner party, and I immediately think of an article I read in a professional journal about how narcissists behave in social gatherings, plus I start checking off the criteria in the Diagnostic and Statistical Manual of Mental Disorders under "narcissistic personality disorder".

Insider: The tanned man interrupts me while joining a group of us at a dinner party, and I say to him, "Can you give us three reasons why we should listen to you instead of continuing with our conversation?"

Insider-Outside: The tanned man interrupts me while joining a group of us at a dinner party, and I say to him, "Wow! I'd love to hear what you have to say, but I wonder if you might hold that for just a moment while we finish what we were just talking about; can you do that?"

Now, frankly, I might also be inclined to use a fourth option that I shall call the scorched earth approach. The tanned man interrupts me while joining a group of us at a dinner party, and I say, "You presumptuous BRUTE. Go away!"

Well, on second thought. Maybe not.