In 1976, Edwards and Gross provided a provisional description of an Alcohol Dependence Syndrome which differed markedly from previous conceptions of 'alcoholism as an entity.' It is instructive to summarise the three species of argument which led to the postulation of this syndrome, as outlined in Edwards (1977). Firstly, Edwards was impressed by alcohol consumption statistics which demonstrated that within populations the distribution of consumption was uni-modal. The prevailing assumption till that time had been that consumption was bi-modal with 'alcoholics' forming their own second mode around some higher mean consumption level. This in turn suggested that 'alcoholism' was therefore experienced by a special group of people whose consumption was discontinuous from that of normal drinkers. According to Edwards, however, contemporary arguments concerning bi-modality versus uni-modality (see for example Davies 1982) overwhelmingly favoured the uni-modal view, and implied that 'alcoholics' were 'not a species standing on their own, but a segment of the population defined only by a cutting point on a continuum. The concept of alcoholism as an entity seems therefore to take a knock.'
Secondly, Edwards took heed of evidence from a number of surveys which investigated population drinking practices. These showed that people moved in and out of periods of troubled drinking behaviour, often without any outside intervention; consequently, people who were 'alcoholics' at one time frequently did not behave like 'alcoholics' some time later. In addition, Edwards noted that what constituted a drinking problem varied according to the social setting; what was seen as problem drinking by one person might be seen as normal in a different social or class setting. To some extent, the definition thus appeared to be arbitrary. Edwards concluded that any satisfactory definition would have to take such differences into account; and that troubled drinking did not reside in the individual but resulted from the interaction of the individual with his/her environment.
Thirdly, Edwards examined D.L. Davies' claims (1962, 1969a, 1969b) to have identified 'alcoholics' whose drinking had returned to normal levels. Such an outcome would be impossible within existing conceptions of 'alcoholism', since the 'disease' would inevitably reinstate itself should the 'alcoholic' return to any form of drinking. However, the idea 'once an alcoholic, always an alcoholic,' did not appear to hold according to D.L. Davies' data. Furthermore, Edwards referred in general terms to studies in which controlled drinking rather than abstinence had been the treatment goal, and to other studies showing the extent to which drinking was influenced by external factors such as environmental cues. It should be stated that the issue of controlled drinking amongst former alcoholics remains a fairly contentious issue in some circles, but Edwards' 1977 paper shows how his thinking was nonetheless influenced by this phenomenon.
After reviewing this evidence, Edwards wrote 'In the face of such a mass of evidence coming from at least three directions, to retain the notion of a specific syndrome of alcohol dependence might seem obdurate.'
Edwards then described an alternative syndrome, under the heading of the Alcohol Dependence Syndrome; a provisional description of which was first offered in Edwards and Gross (1976 op cit). The revised syndrome is an attempt to reconcile the conflicting evidence, and at the same time come up with a conception of alcohol dependence in which pharmacological, physiological, environmental, social, cognitive and phenomenological (experiential) variables are all taken into account. The syndrome is thus more broadly based than traditional concepts based on 'alcoholism' and 'the alcoholic', and takes into account types of influence not considered by those non-interactive and mechanistic conceptions of alcoholism which the Alcohol Dependence Syndrome seeks to replace.
In the Alcohol Dependence Syndrome an attempt is made to bring together the various different ways of conceptualising alcohol dependence under a common banner; one of its specific aims being to produce a synthesis which will facilitate communication between those with differing perspectives. In the end, there is some disagreement about how successful this attempted integration has been. It can be argued that one problem with the Edwards and Gross syndrome is precisely that it can mean all things to all people, particularly since the empirical status of the variables said to comprise it differs rather widely, all the way from physiological indices of withdrawal, to verbal statements about subjective awareness of compulsion to drink.
Two of the most severe critics have been Heather and Robertson (1981, 1985) who argue that certain aspects of controlled drinking amongst 'alcoholics' (i.e. the ability of 'alcoholics' to successfully achieve the goal of returning to normal patterns of alcohol consumption) appear to contradict not merely earlier conceptions of 'alcoholism', but also conflict with certain central features of the Alcohol Dependence Syndrome itself. They suggest that the new syndrome is little more than the old 'disease' model in sheep's clothing.
Nonetheless, Edwards and Gross's proposition is important because it represents an attempt to address and resolve a number of problems arising from simplistic notions of alcoholism as a single entity. It remains a matter for concern therefore that simplistic notions about 'alcoholism', 'alcoholics' and 'disease' still prevail in some treatment agencies; and that the cautious and far-from-revolutionary propositions of Edwards and Gross still set a pace which is too hot for many people to follow.
The point now arises that the comments made by Edwards in the context of alcohol problems might logically apply, directly or in modified form, to addictions of other kinds. The mere fact of arguing that 'addiction' to alcohol is more differentiated and less monolithic than hitherto conceptualised, highlights a need to examine 'addiction' to other drugs from a similar standpoint. Unfortunately, due to the illegal status of illicit drug use and the clandestine nature of the activity, comparable data are somewhat scarcer, and their quality is sometimes even more variable. In particular, data are still lacking on various aspects of normative drug use in the community, aspects of drug use which hospital- and clinic-based studies frequently have difficulty in addressing. Whilst hospital and clinic studies shed light on the medical aspects of drug use after something has gone wrong, they are generally less successful in providing a picture of what we may describe as normal drug use. In a similar way, a clinic-based study of 'alcoholics' would provide only oblique and ambiguous insights into the nature of normal alcohol use in the community, and a view of the drinker based on the 'alcoholic' would be highly misleading. However, before proceeding to the next stage of the argument, it should also be conceded that there are problems with self-reports of both alcohol and drug consumption and such self-reports form the basis for most of the population studies which have been carried out. Basically, the validity of self-reports remains problematic with respect to the absolute consumption levels that people report; a problem which is discussed in more detail in Chapter 8. However, the ordinal characteristics of such data are generally fairly robust; consequently, such reports may be expected to provide some idea of the distribution of consumption in a population, even if the validity of the absolute levels reported remains less certain.
Notwithstanding the above reservations, each of the three arguments advanced by Edwards can now with some qualifications be taken to apply to addictions involving other drugs. For example, although distribution statistics for illicit drug use are somewhat thin on the ground, studies of the self-reports of drugusers provide no evidence for the postulation of bi-modality. On the contrary, the evidence from studies of young people (Plant et al 1985; Coggans et al 1990 and 1991; Davies and Coggans, 1991) suggests that whilst experience with illicit drugs is fairly common (about one in five youngsters report having tried an illicit drug on at least one occasion) reports of use grow progressively less numerous as extent of use increases. In other words, the data from young people are certainly not compatible with a bi-modal theory, the normative evidence suggesting a uni-modal distribution with an extended upper tail; exactly what is described for alcohol consumption. Furthermore, despite the public perception of drug use as the prerogative of a specific 'junkie' group, evidence from surveys of old and young alike shows that drug use is not confined to a particular group, but is a regular occurrence in 'non-junkie' sections of the community. This is true not merely for Class B drugs like cannabis, but also for cocaine. Ongoing research by Ditton and the Scottish Cocaine Research Group (1990 unpublished) in the central belt of Scotland suggests that 61% of a 'snowballed' sample had college or university qualifications, and of these one third (21% of the total sample) had higher degrees or professional qualifications; and comparable research by Cohen (1989,1990) in Amsterdam highlights the prevalence of cocaine use in non-deviant subcultures. Finally, if one includes the widespread prescribing of mood-altering pharmaceuticals within the community, then it is clear that considerable numbers of people, far removed from the stereotyped junkie, are dependent on their drug of preference. Whilst none of these studies amounts to a detailed analysis of the precise nature of the drug-use distribution at the population level, they nonetheless attest to the fact that no instantly identifiable subset of drug users appears to exist, even though a stereotype of the junkie clearly exists. Certainly there is sufficient evidence to suggest, like Edwards, that 'drug-addicts' are not a species standing on their own,' and that 'the concept of addiction as an entity seems therefore to take a knock.'
With respect to Edward's second argument, the evidence is stronger. The studies of Stimson and Oppenheimer (1982) have shown how individuals move into and out of heroin and other addictions without outside intervention. The difference between public stereotypes of drug use, and the realities of the situation have been commented on widely (see for example Finnigan 1988,1996); and the idea that drug use does not reside in the individual but results from the interaction of the individual with his/her environment seems so uncontroversial as to merit little further comment, given the wealth of evidence showing associations between drug use, attitudes and values, and social and economic circumstances (see for example Dorn and South 1987; also the critique of Dorn and South by Mugford and O'Malley 1990).
The third argument, to be valid in the present context, would involve the return by heavy drug users to normal levels of drug use, and is thus rather more problematic. This is primarily because current conceptions of addiction do not permit, or enable the definition of, normal drug use. All use of illicit drugs is illegal and therefore 'abnormal', and public perceptions cannot at the present time entertain concepts such as 'normal heroin use' or 'normal cocaine use'. Consequently, the idea of returning to normal levels of drug use from levels that are abnormal cannot be demonstrated, primarily on account of the linguistic and moral contexts surrounding the words 'drugs' and 'normal'. However, progress may be made on this issue if we are prepared to abandon the use of the word 'normal' in favour of a somewhat looser conceptualisation. For example, research from a number of Scottish studies of heroin-and-polydrugs use in the community (O'Doherty and Davies 1988; Hammersley et al 1990) suggests that heroin use typically follows a cyclical pattern. The O'Doherty and Davies study showed a repeated, cyclical pattern of drug use typically of about three-to-four months duration, during which use escalated to a high level. However, use was not maintained at this high level; a period of non-use followed, after which the user was able to get 'cheap highs' once more on fairly modest amounts. The problem for these users was therefore not one of stopping, nor of returning to low levels of use, which many of them were able to do repeatedly; but it must also be said that such levels were characteristically not maintained. Nonetheless, despite this serious flaw in the argument, the idea of compulsive every-day use was not supported, and in the place of a 'compulsion to use' model, a type of cost-benefit analysis seemed to fit the bill better. Users often appeared to control their use spontaneously when the habit reached proportions which lead to too much 'hassle' in terms of economics, time spent finding supplies, risk of detection, value for money, and other practicalities. In an analogous way, the study by Hammersley et al revealed that 55% of opioid users used on less than 90% of days; and that roughly 25% of those who had taken opioids in the past had not used during the past year. Once again, however, the available data do not provide unambiguous support for the argument that drug users regularly return to lower and more controlled levels of consumption; and whilst the data from O'Doherty and Davies, and Hammersley et al, clearly have something to do with this issue, it is not clear that they address the problem squarely.
Fortunately, there are signs that more direct evidence on this difficult topic may be forthcoming. In the studies by Cohen (op cit) and by Ditton (op cit), cocaine users were asked to examine a series of graphical representations which depicted their cocaine-use career. These graphical representations, first used by Cohen, are illustrated below. The graphs represent in pictorial form fluctuations in use over time, and drug users are asked to choose the graph which best describes their use career.
|I immediately started using large amounts after I first tried cocaine, but gradually decreased since then.|| |
|My cocaine use has gradually increased over the years.|| |
|I started using cocaine at the same level that I still use, and the amount and frequency haven't changed.|| |
|My use increased gradually until it reached a peak, then it decreased.|| |
|I have started and stopped using cocaine many times.|| |
|My use pattern has been very varied over the years.|| |
|Fig. 3 Patterns of cocaine use (from Cohen P, 1989)|| |
In Cohen's Amsterdam study, only 3. 1 % of respondents chose the classic 'increasing dependency' pattern (graph 2); and the modal choice (39.4%) was graph 4, showing increased use followed by a return to a lower level. According to Cohen, these and other data suggest that typically the cocaine user experiences only 'a relatively short duration of the top period of consumption.' Preliminary data from the Ditton study (op cit), for which data are not publicly available at the time of writing, show that the same pattern is the modal selection amongst Scottish cocaine users also. Accordingly, whilst one automatically bridles at the concept of a return to normal cocaine use, these data show that users certainly describe returning to more controlled and limited patterns of consumption with some regularity, and of maintaining that reduced level of use over considerable periods of time.
Overall, therefore, there is justification for suggesting that Edwards' line of reasoning should at least be examined to see whether it has applicability where other drugs are concerned. If the case is more difficult to make out, we must acknowledge that there is a shortage of comparable data due to the difficulties of investigating an illegal activity; but in principle there is no reason to argue a priori that such a case is necessarily untenable.
Unfortunately, the line of reasoning that led Edwards to suggest that alcoholism was not an 'it' has made less headway in the area of drug 'addiction' than in the specific area of alcohol problems. Whilst specialists with differing areas of expertise will have different detailed perceptions as to the relative importance of different aspects of drug misuse e.g. social, economic, pharmacological, psychological etc.) it is nonetheless possible to discern two strands of common meaning which run through the literature, and which appear to be of a categorical nature, whenever the word 'addiction' is used. That is, the use of the word signals that a given group of researchers or treatment specialists, regardless of area of specialisation, have agreed to make certain common judgments about the behaviour in question, which derive from common social perceptions rather than from any particular specialisation or expertise. These judgments are as follows:
i) the word 'addiction' is taken to signify a state. The state is different from the state of being normal though, as in the case of hypnotism, the components of the state remain a mystery. Individuals from different disciplines will, however, have quite different ideas about what the underlying basis of this state is likely to be.
ii) the most salient feature of the supposed state is that it interferes with, or in the extreme case removes, the capacity for voluntary behaviour with respect to a substance or drug. The process which is thought to be responsible for this can range from the 19th century idea of a 'disease of the will' through to more modern conceptions based on biochemistry or pharmacology. Whatever the case, a metaphysical or physical mechanism is proposed which in the extreme case, so it is believed, makes a person unable not to take their drug of preference. This inability to make certain types of choice differentiates them from other 'normal' people.
It may be argued that the above characterisation is false, for two reasons. Firstly, some would probably claim that use of the word 'addiction' does not have to imply a state, but may be used to specify a group of people showing a particular behaviour to an extreme degree. In other words, it is merely quantitative. However, whilst this may be true in other instances, it is argued here that psychologically the word is categorical in function, and that sooner or later its categorical nature imposes itself on our thinking. 'Addicted' is the opposite of 'not-addicted' rather than 'less addicted', and with the sureness of inevitability the categorical nature of the word leads to the search for differences between those who are 'addicted' and those who are 'not addicted'; and subsequently to cures or treatments for those who have 'got it' as opposed to those who 'haven't got it'.
Although it is possible to argue that words like 'addicted' and 'dependent' refer to continuous variables, the postulation of concepts like 'slightly addicted' or 'somewhat dependent' removes from the central concepts most of the denotative or diagnostic value they might otherwise have; and like the oxymoron 'fairly unique', they confuse rather than clarify the issue. The concept of addiction as an 'it' cannot be salvaged by towing it to safety with linguistic qualifiers, like some broken-down vehicle.
The second argument suggests that ultimately there is no difference between 'scientific' explanations (e.g. explanations in terms of the pharmacological impact on neural transmission) and explanations in terms of will-power, decision-making, intentions, and so forth. The latter, it can be argued, are terms of convenience pitched at the level of phenomenology, simply because they refer to events with which no current pharmacology or physiology can deal, but which in principle are explicable ultimately at that level. Consequently, there are not really several competing types of explanation, but only one.
There are a number of rebuttals to this position, both pragmatic and logical. Pragmatically, one can merely say that one will be prepared to accept the usefulness of the pharmacological (or whatever) basis for motives and intentions and so forth, when the day comes that they are understood at that level. However, there is a more fundamental issue to be discussed here, namely that understanding drug action is not the same as understanding the causes of voluntary behaviour. For example, physiological states giving rise to feelings of pain or discomfort are not in themselves sufficient to specify behaviour. People given heroin in hospital seldom become dependent; people will undergo fatigue, stress and even torture rather than divulge a secret; and two 'addicts' can make different decisions about whether to continue use or not. In other words, such decisions derive not just from interoceptive cues (perception of bodily state) but on the situation in which an individual finds him/herself, as perceived and interpreted by that individual.
Addiction as Disease
The most problematic concept in the addiction area is 'disease'; and despite repeated and consistent assertions in the recent literature that the disease notion has either (a) a highly qualified and constrained application in this area (cf. the Alcohol Dependence Syndrome), or (b) has no applicability whatsoever (cf. Heather and Roberston's account of controlled drinking), it remains a notable fact of life that the idea of addiction-as-disease is alive and well amongst many drug and alcohol misusers and their families, and in many treatment agencies. It will not go away for one simple reason. Namely, it is highly functional.
This functionalism will become the main theme of the present book in later chapters, but for the time being we must content ourselves with observing that as a general rule, where a disease interferes with behaviour it replaces something purposive and coordinated with something chaotic (for example, Parkinsons disease, Huntingdon's chorea, peripheral neuritis). Commonsense suggests that the disease definition should indeed normally make reference to something which disrupts or is inimical to integrated and purposive behaviour patterns. It does not make sense as a category description for the replacement of one behaviour with a new, equally integrated, coordinated and purposive pattern. If we postulate a disease which has the direct capacity to force people to steal, to lift up glasses, or to stick needles in their arms when they are actually trying not to, and furthermore to execute long strings of appetitive goal-directed behaviour as precursors to these actions, we have to accept the possibility that any integrated chain of goal directed behaviour in any realm might be nothing more than a disease symptom.
The fact that the short and longer-term disruptions of behaviour which sometimes result from taking drugs can become the occasion for postulating drug taking as a disease manifestation shows a familiar confusion; namely the confusion of intentions with outcomes. For example, during the late 19th and early 20th centuries, missionaries went to Africa where many of them caught malaria and died. The disease was malaria; not the decision to go to Africa. Furthermore, whilst a doctor can in principle treat the malaria, he can only advise people not to go to Africa, on the basis of his own beliefs and opinions about Africa. In a similar way, damage to health caused by drugs does not imply that the decision to take them is pathological, any more than deciding to go to Africa is pathological.
There is little evidence that any disease has the capacity to impose specific integrated behavioural strings in the place of previous ones. However, it may be the case that, in response to the conditions created by a disease, people will voluntarily adopt various strategies to cope with it; and that in a probabilistic sense, some strategies are more useful, and hence more likely, than others. However, these are purposeful and voluntary adaptations to conditions created by the disease, and their inclusion alongside other symptoms creates important dilemmas. Is it reasonable to conceive of the practise of injecting insulin as a symptom of diabetes? If so, where was this symptom before the discovery of insulin therapy? However we resolve this problem, it is clear that, even if we classify such things as symptoms, there is a clear difference in the way the word is being used when (a) we describe something such as high temperature, shortage of breath, or a skin rash, as a symptom, and (b) we talk about going into a pub and buying a pint of beer as a symptom. Whatever we mean by the word 'voluntary' (i.e. whether we take it to imply a metaphysical exercise of the power of 'free will', or a particular mode of cerebellar functioning) it is clear that going into a pub is voluntary in a sense that having a high temperature is not. For these reasons, the inclusion of acts of drug- or alcohol-directed appetitive behaviour as parts of the disease symptomatology, alongside involuntary bodily changes, lumps together two sets of phenomena which require different levels of explanation.
Craving; 'Having to Have', or 'Just Wanting'?
Within the literature on 'addiction', it is often assumed that voluntary acts are brought into the mainstream of symptomatology by the interaction of two mechanisms. These mechanisms are (a) craving and (b) withdrawal symptoms. These two factors are in some way responsible for the re-classifying of behaviours which are voluntary in normal people as non-voluntary or symptomatic in 'addicts.'
Discussions of the concept of craving are available in Gossop (1990) and in West and Kranzler (1990). In both these pieces of work, the authors accept self-ratings of craving as an indicant of a specific inner (subjective) state, central to which is some sort of compulsive desire which can be accessed through verbal report. it is clear from their discussions, however, that both sets of authors are aware of difficulties with this approach. It appears that there are problems with the concept of craving, especially where this is postulated as an independent entity. In ordinary usage, craving is a response to some basic biological need, giving it an implied compulsive quality, and semantically distinguishing it from a simple want. Thus, lack of food may produce craving for food, and various associated physiological states will give the craving its particular quality. The implication of craving is that the person in question does not simply want, but in some sense has to have, something. The relationship between an emotional state and a bodily state is a thorny one, and goes back at least as far as William James (1884, 1890) who posed some perplexing questions on this issue. Basically James suggested that people do not, for example, run away from a tiger because they feel afraid. Rather, they feel afraid because they perceive the urgency with which they run away from the tiger. Fear, he reasoned, is merely the perception of a bodily state; and in the absence of such a state, the notion of fear per se becomes meaningless, a mere idea robbed of its essential ingredients. By analogy, James' idea suggests that people do not seek drugs because they crave, but crave because they seek drugs.
There are a number of criticisms of James' theory, not the least of which is that it ranks alongside other 'mind-as-epiphenomenon' explanations discussed in the introduction. In addition, insufficient types/combinations of discriminable bodily states exist to enable perception of the whole range of felt emotions through this means; for example fear and excitement cannot be readily differentiated in these terms. Nonetheless, it is too easy to couple together the semantic argument that says people use drugs because they feel craving. What is craving for drugs in the absence of perception of bodily state? On what grounds do we postulate an irresistible drive rather than a desire to have? And if craving only makes sense when underlaid by withdrawals, what is the need to postulate it as an independent entity?
In fact, the evidence for the existence of craving is basically that people say they feel it, when asked the appropriate question. Craving cannot be inferred from merely observing behaviour. Thus, in animal experiments, the fact that unfortunate creatures regularly overdosed on some substance suffer from withdrawals and show 'drug-seeking behaviour' is an established fact; but whether the animal craves drugs remains problematic, in the absence of verbal reports from animals. Drug seeking is not in itself evidence of craving if we wish to postulate craving as an independent entity. To have any meaning, the poor creature has to be able to sit and crave quietly to itself, without necessarily going in search of its drug; and even then we cannot assume that it is craving, because there is no independent measure. If we ask a person to say how they feel in such circumstances, however, his/her verbal report will be taken as evidence for an independent craving process. The postulation of craving thus illustrates one more way in which we can be deceived by the preeminence we give to our own verbal behaviour. It illustrates the problem of reification, the process whereby a convenient semantic symbol becomes transmuted into an entity which is assumed to have actual existence. In its starkest form, the process involves the psychologist (or whoever) asking someone to provide ratings in terms of some semantic label, on the customary five- or seven-point scale, the labels coming in the first instance from the psychologist. The fact that the subject complies is carelessly taken as proof that the word must refer to something real, an entity; otherwise how could he/she produce systematic ratings? In fact however, the subject's acquiescence, and subsequent performance, can more parsimoniously be described in terms of the demand characteristics of the interview, and any systematic variation in terms of strategy and response bias, or 'making sense of the task'.
Put simply, craving is an alternative word which we can use to describe an experience of discomfort, and an accompanying desire to curtail or avoid it. If we examine craving from a Kelley-type standpoint, it looks as though a craving explanation is offered in circumstances where i) people consistently choose to reduce their discomfort, and ii) there is a consensus belief about the biological determinants of the discomfort. Thus, whilst drug users 'crave' (have to have) drugs and hungry people 'crave' food, people merely 'want' colour T.V. sets or holidays in Venice. The use of the word 'craving' is an interesting exercise in attribution, and its primary purpose is to convey how we are intended to perceive the addiction process. It refers to the fact that sometimes people feel a strong desire to use, or use more of, their preferred drug, but it gives the impression of an autonomous force whose power cannot be resisted; hence its attraction. In fact, whether people resist the experience depends on whether they have good reasons, or no good reasons, for doing so. People in the dentist's chair have a craving to get up and leave; but by and large they stay put.
Perhaps the last word on the craving issue should be given to a pharmacologist, since it may be assumed in some quarters that the pharmacology of drug action provides a simple and sufficient account of craving as an irresistible force. It is clear however that modern pharmacology takes a more differentiated view. Wise (1990), in an account of 'Reward Pathways and Drug 'Craving' ' (pp 43-45) says the following:
'If there is no depression of reward pathways when chronic use of cocaine or heroin is terminated, what other explanation is there for feelings of drug craving and instances of drug relapse? One possibility is simply that the subject remembers the last reinforcement….'
'Rats clearly remember their last few rewarding brain stimulations, adjusting their speed of running for access to more brain stimulation in proportion to the amount of stimulation received on the last trial. The running speed is the same whether the last trial was five minutes or five days earlier; the rat simply remembers what it has been receiving for lever pressing (Gallistel et al 1974). If the rat remembers strong reinforcement, it runs quickly; this has nothing, apparently, to do with the state of its dopamine receptors, since five days should make a difference in any changes in receptor supersensitivity which were caused by the last reinforcement experiment.'
'Craving for cocaine or heroin may, like craving for nicotine in a smoker who has been nicotine free for many years, simply be triggered by memories of past experience. Like a cat that has tasted fish, a human that (sic) has tasted cocaine may be unwilling to give up the hope of repeating the experience. If this view is correct, then it may be more appropriate to look for the biological correlate of craving in the neurobiology of memory, and not in the neurobiology of positive reinforcement.'
Wise (op cit) thus implies strongly that craving derives not from some drugrelated change to the reinforcement mechanisms of the brain which turns a 'want to' into a 'have to'; but more mundanely from people's simple recollections that some experience was pleasurable the last time it occurred, coupled to the hope that it might happen again. In other words, a 'want' deriving from the normal everyday pharmacology of memory; not a 'have to' underlaid by some alien drug-induced pharmacology.
It is necessary to note from the outset that withdrawal symptoms are real, in the sense that changes to homeostasis brought about by regular drug use lead to discomfort and temporary illness when the drug is absent or withdrawn; and that extreme withdrawal symptoms have been induced in laboratory animals on a number of occasions (Stewart et al, 1984). The most convincing theoretical argument about possible mechanisms for withdrawals implicates endogenous opiates, namely the enkephalins, beta-endorphins, and dynorphins (Bloom 1983). These are opiate-like substances which occur naturally in the body, and they have the function of inhibiting neurotransmitters and so producing analgesia. Taking opiate drugs over a prolonged period makes the role of the endogenous opiates redundant in some sense, and their production is therefore reduced as part of a homeostatic feedback mechanism. When the opiate drug is withdrawn the body is left without its natural analgesic defences against over-activity in the excitatory systems (Johnston 1990).
However, whilst withdrawals are real enough and convincing pharmacological models exist to explain their existence, it is also the case that the precise nature and extent of the symptoms, particularly the behavioural manifestations, are not uniquely specified by the drug and its pharmacological effects. The form taken by withdrawals, their severity, and the significance attached to them by the sufferer, depend on a variety of situational and cognitive factors in addition to straightforward pharmacological effects. This has been demonstrated with animals as well as observed with people. For example, rats orally addicted to morphine show physiological and behavioural signs of withdrawal when in their customary cage (i.e. the cage where the addiction process took place); but show more exploratory behaviour and less external signs of withdrawal when in a new and different cage. More importantly, the animals show greater avidity for morphine solution during the 're-addiction' phase when in the cage where the original addiction has been acquired (Thompson and Ostlund, 1965) than in the changed environment.
An even more striking demonstration of the way in which withdrawals depend on contextual factors, and thus cannot be conceptualised as some sort of powerhouse or driving force for future drug use, was provided by MacRae and Seigal (unpublished; cited in Balfour 1990, p. 81) in an experiment with 'yoked' rats. Rats worked in trios, each member being in a different cage. One rat could self-administer morphine through a cannula, by means of a lever press. The second animal passively received the same dose at the same times as the first animal by means of a yoked or 'slave' syringe pump, but could not actively self-administer. A third control animal passively received Ringer solution, also in a yoked manner. The experiment ran for six days, but on the seventh day no drugs were available to any animal. On the seventh day there was a clear and substantial difference between the animals in terms of withdrawal distress, with the rat who actively self-administered showing by far the most severe symptoms. The authors conclude:
'It would appear that interoceptive signals of a drug, incidental to voluntary self-administration, can importantly influence the magnitude of withdrawal symptorns.' (The reader's attention is specifically drawn to the use of the word 'voluntary', which is essential to making phenomenological sense of this study, despite its underlying determinist philosophy.)
At the human level it can be observed that the florid withdrawal symptoms portrayed in films like 'The French Connection' are by no means typical, and that prisoners on remand who endure forced withdrawals from heroin often seem to suffer from something more like an attack of influenza, rather than screaming and clawing at the walls. Finally, just to confuse the issue, many of us will have seen spoiled children who can writhe about on the floor and make themselves sick if mummy refuses to give them another Easter Egg.
This variability creates the final dilemma. If, as seems to be the case, there is variability in severity of withdrawals as a function of time, place, expectation or whatever, then it becomes increasingly difficult to conceptualise withdrawals as the basic powerhouse or engine-room for 'addictive' behaviour. Furthermore, since we have argued that 'craving' lacks its essential property of compulsion, so 'addiction' becomes less monolithic and more amenable to explanation in human terms.
Cures for Taking Drugs
Returning to our main theme, it is suggested that the concept of 'addiction' might conceivably have some value if it gave emphasis to the normal and nonpathological decisions people make about drugs; but in fact it is usually employed to encapsulate certain assumptions about what drugs do to people, thereby implying a process from which the powers, wishes and intentions of the drug user are specifically excluded. The idea that the pharmacology of drugs makes people into addicts against their 'will' has to be contrasted with the idea that people make addicts of themselves because they choose to do so. The latter is a challenging suggestion which deserves serious consideration, and it certainly makes sense of the fact that treatment for addictions frequently seems to have more in common with procedures for attitude change, than with medical intervention.
From such a standpoint, the term 'addiction' appears to refer not so much to some medical condition as to certain disapproved-of ways of thinking and deciding, certain acts of choice which are not qualitatively or quantitatively different from thinking, deciding and choosing in any other area of human life, but which happen to involve drugs. Because of the pharmacological action of drugs, the consequences can be disastrous to health, family and all aspects of living, but such consequences do not in themselves warrant the postulation of a special state which compels their use. Furthermore, there is no cure for drug taking because there is fundamentally nothing to be cured; no more in fact than there is a cure for rock climbing, football, or playing the violin. But if there are reasons for supposing that in a given case the rock climbing, football, or violin-playing are resulting in social and economic problems for the individual and for others, one might try to persuade him/her to reappraise the basis on which they make their decisions.
Because the basic decision to take drugs, notwithstanding the consequences of excessive use, is fundamentally non-pathological, much of the research into 'addictions' which seeks a better understanding of why people drink too much, smoke too much, or take drugs, is in reality a search for why people decide to do, commit themselves to, or dedicate themselves to, anything. Insofar as the search is for an understanding of the whole of human 'choice' behaviour, one can expect that no sudden breakthrough will occur; the search can go on more or less indefinitely. Furthermore, from some perspectives the problem appears to confound alternative metaphysical (i.e. free will) and materialist (i.e. determinist) propositions which can never be explained under a common rubric. In other words, the underlying philosophy is flawed. Within such a framework, theoretical progress becomes more a matter of good luck than of good management; and practical progress hinges on reactive and short-term expediencies, rather than presenting any clear or realistic view of how drug use fits into society in the longer term.
Certain central features of the received wisdom about drug addicts and addiction have to be challenged. The idea that addiction is a state in which the driving force for autonomous action becomes lost to the individual, and is taken over by craving, an irresistible psychological force fuelled by inevitable and excruciating withdrawal symptoms, is untenable since these concepts do not in fact possess the monolithic properties that they would require in order to assume the roles assigned to them. In their place, we require a conception of drug use which restores the user to centre stage, and within which his/her motives and intentions within particular contexts become the focus for attention and future theoretical development.