Blogging Lovaas, 1

Teaching Individuals with Developmental Delays: Basic Intervention Techniques
O. Ivar Lovaas, Pro-Ed, 2003

Preface

Individuals are evaluated on many dimensions, including emotional development, social skills, educational achievements, and language skills. Within behavioural psychology, the branch of psychology that forms the foundation of this teaching manual, such dimensions are referred to as behaviours. Behavioural psychology thus assesses such observable aggregates as intellectual behaviours, emotional behaviours, social behaviours, educational behaviours, language behaviours, aggressive behaviours, occupational behaviours, and self-help behaviours.

Insofar as behaviours can be observed, they can also be separated and objectively measured.

The very first paragraph of the text contains in four lines an almost complete list of the sleight-of-hand effects that characterise behaviourism in general and ABA in particular.

Individuals are evaluated on many dimensions, including emotional development, social skills, educational achievements, and language skills.

This is a statement of commonsense observation, like that dialogue in Mark Twain.
"Do you believe in adult baptism?"
"Believe in it? Hell, I've seen it done!"

Like that joke, Lovaas' line relies on an identification of what is seen with what is claimed. Individuals are evaluated - yes, there are tests, we have seen them done, we have undertaken them ourselves as tester and testee. In another sense, though, the words "Individuals are evaluated" are a claim, as would be "Individuals are transfigured" or "Individuals are infected" or "Individuals are weighed" - that is, there is not only a behavioural claim but a truth claim, amounting to a claim that there is a link, or even an identification, between the evaluation process and the evaluation findings, on the one hand, and "emotional development, social skills, educational achievements, and language skills" on the other.

Common sense - which I do not regard as determinative, but which needs to be mentioned - would probably say that yes, there was presumptively a connection between our evaluations and something real in that individual, between educational tests and educational achievements; as a culture and a state we rely on innumerable tests every year to sort out exactly that. We'd also probably say, though, that the transmission line between the external assessment and the internal education is loose and complicated, and that there are many possible hazards that would throw the assessment off.

Lovaas's list, in any case, includes a blend of internal mental states and external manifestations. Emotions are unobservable; social skills only exist when manifested; educational achievements can be mental states or manifestations; language skills we'll come to later. That blending is a step on the way to an identification between mental states and behaviour.

The next step comes. "Within behavioural psychology, the branch of psychology that forms the foundation of this teaching manual, such dimensions are referred to as behaviours." Again, the words 'are referred to' are intended to carry the burden of 'are'; these dimensions are behaviours, are wholly behaviours, are whatever their involvement with mental states not in any meaningful sense mental states. To underline here, things that are not behaviours, or are at least not wholly behaviours,are to be referred to - are to be identified with - behaviours, as if I was to say "This dog will be referred to as a cat" and expected to command agreement.

So then to another list: "Behavioural psychology thus assesses such observable aggregates as intellectual behaviours, emotional behaviours, social behaviours, educational behaviours, language behaviours, aggressive behaviours, occupational behaviours, and self-help behaviours." Here the hybrids - the 'observable aggregates' - are approaching the oxymoronic. What is 'intellectual behaviour'? The intellect is internal, behaviour is external, philosophers have struggled for centuries to link the two, sometimes through the pineal gland, but there appears to have been a shortcut. Occupational behaviours, though, are unquestionably behaviours; one's job description is not under any philosophy internal. Aggressive behaviour, on the other hand, is a descriptor of a medieval humour, reading back from the thrown punch to a socially transgressive attitude that is then assumed to be an internal category. But then - 'observable aggregates'? You can observe a behaviour, you can't observe an aggregation; that's a construct, a hypothesis.

I'm not just spinning my wheels here, being ironic; classical behaviourism did have a means of resolving just this difficulty. In full-court-press behaviourism there was only behaviour, no aggression, no intellect,no emotion. There was only stimulus and response, and no resort to any explanation that involved the black box. I think that was the wrong answer - spectacularly so - but it was consistent, which modern behaviourism isn't. Modern behaviourism has abandoned the underlying theory without changing any of its own behaviours; its prescriptions are identical now and then, which is why the Lovaas method has been able to bridge the eras.

But we haven't finished yet."Insofar as behaviours can be observed, they can also be separated and objectively measured." That 'insofar' is another linguistic card trick. The sentence could mean either "To the extent that [intellectual] behaviours can be observed, they can be measured"; which is a mild claim, particularly if one sets the extent to zero. It also carries the burden of "Because [intellectual] behaviours can be observed, they can be measured" and we have been carried further into accepting the ability of behavioural psychology to speak authoritatively about that which it cannot know.

And even if it is only behaviour, does that mean we must accept that it can be separated and objectively measured? I'm still unconvinced. How does one separate a single behaviour? When does it start, when does it stop? We are certainly going to get practical examples further in to the book, but the theoretical issues have not yet been dealt with satisfactorily.

And measured? Again, some doubts remain. If one has separated behaviours, they can be counted; but can they be measured, a term which implies comparison, even numerical comparison? Heights ands weights can be treated mathematically, multiplied, divided, subtracted and added; behaviours would seem to have only the less-than< and greater-than> operators.

To be fair, that last Lovaas sentence was from another paragraph.

OK, you don't go to Lovaas for theory, that was Skinner; this beginning may simply be a pious invocation of the old Skinnerian gods. Even so, it's hardly satisfactory.