function improves and as our tools to investigate brain malfunction develop so
the range of psychiatric disorders known to have an organic basis is likely to
increase ( Rutter, 1986). It also cannot be doubted that child psychiatric disorder
has important associations with neurodevelopmental immaturities, so-called "neu-
rological soft signs" ( Shaffer et al., 1983), and with developmental delays in
functions such as language ( Howlin & Rutter, 1987). In these instances the con-
ceptual difficulty stems from the uncertainty as to what the "soft signs" or de-
velopmental delays reflect ( Shafer et al., 1983; 1986). As such, they cannot be
taken as unambiguous indicators of organic brain dysfunction, although some-
times this will be the case. A further difficulty, moreover, with respect to con-
sideration of hyperkinetic or attention deficit syndromes, is that the
neurodevelopmental associations are by no means confined to this type of be-
havioural problem. Thus, Shaffer and his colleagues ( 1985) found that neurolog-
ical soft signs were most strongly associated with anxiety disorders in adolescence.
Accordingly, one of the key features of the MBD concept that must be rejected
is the idea that brain damage or dysfunction gives rise to a uniform behavioural
picture. It does not ( Rutter, 1981). Indeed, the presence of brain damage or dys-
function is associated with an increase in the risk for a wide range of psychiatric
disorders of which attention deficit disorders are by no means the most prominent.

It should be added that the idea that MBD syndromes were largely explicable
on the basis of perinatal complications must also be rejected. There are, at best,
only very weak associations between such complications and any of the features
said to characterize MBD and the presence of reproductive abnormalities does
not serve to delineate any type of behavioural syndrome ( Nichols & Chen, 1981).
To an important extent, this negative finding is a function of the fact that in most
cases perinatal complications do not in fact give rise to brain damage and it is
possible that conclusions may need to be modified once there has been a more
adequate follow up into middle childhood of children in whom perinatal compli-
cations have been associated with brain damage as identified by modern imaging
techniques ( Stewart, 1983; Stewart et al., 1987). However, it seems dubious
whether this will give rise to a specific association with attention deficit or hyper-
kinetic disorders. All in all, biological findings fail to establish the nosological
validity of an MBD syndrome ( Ferguson & Rapoport, 1983; Taylor, 1986b). How,
then, is this dismissal of the MBD notion to be reconciled with the claim a few
years ago by Gillberg and his colleagues ( 1982, 1983) that "about a third of chil-
dren diagnosed according to strict criteria as suffering from 'minimal brain dys-
function' show generalized hyperkinesis" (p.245)? The claim derives from a
systematic and large scale epidemiological study. The flaw is that similar problems
in attention were used to define both MBD and the hyperkinetic syndrome. Ac-
cordingly, the association was to a large extent tautological. As used by these
investigators, MBD was defined in part by the very attentional deficits that it
is supposed to explain. That point is recognized by Gillberg (this volume). His
data are of very considerable interest in their own right but they do not help with
syndrome definition. Of course, there is nothing wrong with an hypothesis that

-2-