Degree of compulsion-to-use, or 'addictiveness', has long been seen as varying between different substances. There is irrefutable evidence of differences in pharmacological action between individual drugs, as well as at the broader level of categorisations such as CNS stimulants and depressants. A number of texts are available which provide accounts of the pharmacological action of different psychotropic substances, and no purpose is served here in simply relaying information which is readily available elsewhere. However, given the current concern about cocaine and its derivative, crack, it seems appropriate to use this as an example in the context of the current discussion, with the proviso that the basis of the argument is the same for other drugs also.
It is widely held that crack cocaine has uniquely addictive properties, and this idea is reiterated not only in the popular press, but also in the more straight-laced media. For example, the 'Observer' newspaper (Browne, 1988) carried a full-page feature on crack in which it was reported that according to unidentified 'experts', three out of four first-time users became 'instantly addicted'. From this standpoint, the article goes on to attribute a variety of other behaviours, which clearly have an economic basis, to the action of the drug itself. For example:
'Because of its volatile effect on the user and its addictive qualities, crack leads to a significant increase in violence, muggings, burglaries, theft and other crimes.'
In fact, most theft of whatever type stems from the desire of people wanting something, but not having the economic resources to realise their ambition. This is true of cocaine, but it also true of colour-television sets, motor-yachts, and bars of chocolate. The fact that poor people who want drugs steal in order to get them hardly amounts to a major insight into the motivators of theft.
However, more problematic is the assertion that cocaine has the capacity to be instantly addictive. The available evidence reveals that cocaine and crack -have rapid, but short-lived, pharmacological effects of a fairly dramatic nature, these varying (as with all drugs) as a function of the purity of the drug and the mode of administration. Other things being equal, injection produces the most rapid hit, but with cocaine the effect lasts for about 20 minutes via this route, whereas the effect from snorting lasts longer but is said to be less intense and 'smoother'. Crack is relatively new, and consists of freebase cocaine contaminated with varying amounts of sodium carbonate used in its preparation.(see note) Because of its recency, the precise pharmacology of crack is less well documented, but verbal reports suggest the hit is very rapid due to increased volatility, almost analogous to injection with the parent substance, cocaine.
Cocaine and crack produce increases in certain neurotransmitters in parts of the brain, neurotransmitters being chemical 'messengers' that assist in the transmission of impulses between nerve cells. Cocaine causes increases in the levels of monoamines, notably dopamine, a chemical transmitter that is to varying degrees lacking in sufferers from Parkinson's disease but which is often present to excess in schizophrenics, where its causal influence is a matter for some debate. However, cocaine is a monoamine oxidase inhibitor (MAOI), which means to say that it achieves its effect of increasing monoamine levels NOT by stimulating their production but by inhibiting their re-uptake by those cells that normally carry out this function. In other words, it inhibits reabsorption, rather than stimulating production. The detailed pharmacology has been revealed in recent studies by Bozarth (1990) and by White (1990) in studies of rats.
Such pharmacological studies almost certainly reveal the mechanism underlying cocaine's pleasurable effects, and consequently why it is used for recreational purposes. They also enable comparisons with other drugs to be made, and provide support for statements about the relative intensity and differing quality of different-drug experiences. Clark (1990) for example, showed that even rats could discriminate between the subjective effects of different drugs, using an operant (lever-pressing) paradigm.
From this point, however, the mythology of addiction takes on a curious turn. The next inference stems from the preference for a particular form of explanation rather than from the inductive logic of science, and asserts that the person 'has to have' the drug, and 'has no choice' but to use it. There is, of course, nothing in the pharmacology itself that justifies such an inference, which signals a change in preferred philosophy rather than a change in fact. The problem of whether any behaviour is mechanistic or volitional hinges around definitions and beliefs. At the end of the day, any mechanistic (logical positivistic) interpretation requires a pharmacological/physiological basis for all action, so the specification of cocaine use as being non-volitional because it has a demonstrable pharmacology fails to distinguish it from any other behaviour and any other pharmacology. Drug taking cannot be seen as more determined since there is no behaviour which is less determined, unless the determinist is, paradoxically, prepared to allow metaphysics to enter the equation.
It is interesting to ask why it is that on the basis of evidence which is inappropriate to the purpose (namely the discovery that cocaine has, unsurprisingly, a knowable pharmacology), the non-logical conclusion is derived that 'therefore' cocaine-related behaviour is quite different in fundamental principle from 'ordinary' behaviour with respect to philosophical imponderables like 'free will' and 'volition'. The reasons why this happens will be discussed in detail in a later chapter.
However, the suggestion that drug-related behaviour is in some sense automatic is supported by another body of evidence which is less easily dismissed. A number of workers have performed variants of an experiment in which some small animal, normally a rat, has a cannula inserted into its brain through which tiny amounts of drug can be administered to selected sites. In a paper by Bozarth (op cit) originally entitled 'The Pre-eminence of Animal Studies in Comparative Substance Use', but subsequently published under a more modest heading (1990) a number of studies are described in which rats, surgically prepared as described above, are placed in a Skinner-type box. The lever-pressing response delivers measured amounts of drug to locations in the rat's brain. In these circumstances, the rat shows a pattern of lever pressing which, from the point of view of the observer, appears to be to be compulsive and automatic (see chapter 3). The rat presses the lever to obtain drug in preference to all other behaviours including eating and if left to its own devices the rat will eventually starve.
The question that now arises is the extent to which this situation serves as a model for the behaviour of human beings, seeking out their drug of choice despite the possible negative consequences of this behavioural focus. The standard rebuttal to claims that animal studies can actually shed light on human action is usually some type of comparative argument, the essential gist of which is that animals are basically deficient in ways that people believe are important; and consequently one cannot make inferences from one to the other willy-nilly. However, whatever the virtues of this argument, the fact that it can be trotted out whenever necessary as a way of defending against anyone who would argue for phylogenetic continuity in behaviour, makes it somewhat unsatisfactory. In the present circumstances, a more satisfactory argument concerns the demand characteristics of the experimental situation.
Rather than claiming that humans are not like rats, it seems equally reasonable to suggest that people in the same situation as the rat would behave in very much the same way. Studies of deprivation and confinement do indeed reveal fundamental changes to what would be described as normal human behaviour, among which is the increased significance that becomes attached to events that would normally be regarded as trivial (Watson 1978). Furthermore, the real analogy is not from the rat in the Skinner box to the drug user in the street; but from the drug user in the street to the rat in his/her accustomed environment. It is almost certain that given a normal environment in which to explore and wander, and a means of ingesting the drug which makes sense to the rat, the hapless creature will show more variety of behaviour than is revealed in the Skinner box. A telling experiment was performed by Johnson and Johnson (1972) with Siamese fighting fish. The fish, placed in a type of aquatic Skinner box, had been previously shown to display 'automatic' aggression whenever a suitable sign-stimulus (e.g. another fish) was introduced. The fish appeared to seek out such aggressive encounters, to find them highly rewarding, and to do little else besides. However, in truth there was little else to do. When Johnson and Johnson introduced a variety of stimuli into the environment, the fish worked harder to obtain sight of these novelties. What looked like an automatic and inevitable drive towards aggression turned into something more like curiosity and exploration when a more varied set of alternatives was offered.
The crucial study in the present context would thus involve observing whether rats, taught to use some drug, would seek out that drug to the exclusion of all else when offered an environment comparable in variety and richness to that normally available to rats. Though the precise study has not been carried out, logic suggests that the answer would almost certainly be 'No'. Consequently, with respect to the reasons why people and animals use drugs, the animal studies show little except that ethologically senseless environments produce ethologically ridiculous behaviours.
However, although no study has been performed which compares rats in boxes to rats in a natural environment, and the effect of this manipulation upon drug self-administration through an ethologically sensible route, two studies have been performed which come quite close to this ideal. In a study by Alexander, Coambs and Hadaway (1978), rats living in isolation in standard laboratory cages were compared to rats living socially in a large open box. The rats were exposed to a series of three-day reinforcement cycles previously shown by Nichols et al to increase oral self-administration (drinking) of morphine solution. These cycles are referred to as periods of 'forced consumption'. Subsequently, when 'choice days' were introduced during which the rats could choose either morphine solution or water, the isolated/caged rats drank more morphine than the social rats. The authors write:
'A possible explanation is that…for the isolated rats the reinforcement value of morphine ingestion was enhanced by relief of the discomfort of spatial confinement, social isolation and stimulus deprivation. Similarly, in the social group, the reinforcement value of morphine may have been diminished by it's interference with the rat's natural activity patterns.'
'A housing-condition effect, if substantiated, needs to be accounted for by existing theories of addiction. Conditioning theories of opiate addiction generally state that self-administration of opiates is reinforced by relief of withdrawal symptoms. In this study, the effectiveness of training procedures specifically based on this principle.... appeared to be environment specific. Not only did the social group ingest less morphine than the isolated group during the choice days of the Nichols Cycle Period, but they decreased their consumption while the isolated rats increased theirs. A neurochemical change theory would also need to account for a housing conditions effect.'
In a further study, Alexander et al (1981) took things a step further by examining the effect of early, as opposed to later, colony housing on oral ingestion of morphine by rats. Rats housed socially at the time of testing drank less of a morphine-hydrochloride (MHCl) solution than did isolated rats, but drank no less of control solutions. Colony-dwelling rats who had been previously isolated drank more MHCl solution than did rats reared in a social setting since weaning, but only at the lower MHCl concentrations. Analysis of variance of the data showed only a significant main effect for conditions of housing at the time of testing; the effects of early environment were insignificant, though there was a marginally (0.06) significant interaction. The authors suggest that colony rats avoided opiates because opiate consumption interfered with the performance of complex species-specific social behaviours that had no relevance when the rat was artificially isolated.
Self-Reports from Humans
Despite the above arguments, it is regularly observed by clinicians that clients with drug problems report feelings of loss-of-control and lack of volition. These statements are frequently expressed in such earnest and self-deprecatory terms that to suggest clients are liars, or simply deluded, hardly seems to do justice to the horror of the situation. It is statements such as these that tend to become enshrined in the work on drug action; and in consequence the pharmacological quest becomes a search for 'scientific proof' to back up the verbal reports. However, there is a basic misconception about the nature of statements from drug users who so regularly assert that they 'cannot stop. The statements themselves are not 'scientific' in the first place, but functional. The earnestness with which the statements are made attest to their functional necessity, rather than their literal truth.
Bem (1968; 1972) suggested that people make inferences about others on the basis of observing what they do. Thus, our beliefs and knowledge of others are derived from observations, on the basis of which we make inferences. However, Bem went on to say that we make inferences about our own behaviour in exactly the same way. When a person reports not being able to control their drug use, we are in error if we assume they are revealing first-hand (proximate) knowledge about the fundamental principles underlying their behaviour. Rather, such reports are socially-functional inferences deriving from self observation.
If we observe that, with great regularity, we over-indulge in some activity to the detriment of health, family, friends and economic functioning, we require a linguistic formula that enables us to explain these circumstances in an acceptable fashion. The statement that 'I cannot stop' is not a statement of fact, but an inference based on the self-observation that I reliably fail to do so.
The statement 'I cannot stop' is thus primarily a metaphor; and no other linguistic device adequately captures the moral and behavioural dilemma in which the 'addict' finds him/herself.
With the permission of the author, the text of this sentence has been altered to reflect a more precise definition of crack cocaine. (return to text)