Introduction
Langlitz and colleagues wrote about clinically used psychedelic drugs and the possibility
of a “moral psychopharmacology” earlier in this journal (1). They emphasized the context-dependency
of the effects of these substances (e.g., ayahuasca, psilocybin) and the importance
of understanding their impact on social and moral cognition, particularly now that
there is more research on their possible clinical applications (e.g., to facilitate
psychotherapy). In this opinion article, I want to, first, reflect on the context-dependency
from the perspective of recent research on placebo effects, and, second, clarify different
meanings of “moral psychopharmacology”. The latter will be placed in the context of
a broader conceived view on drug instrumentalization (2–4) and different values associated
with it (5–7). In the conclusion, I will briefly distinguish aspects primarily relevant
to theoretical, research-oriented, or applied perspectives, respectively, to inform
further theoretical, empirical, and ethical discussion of these topics.
Context-Dependency of Drugs
In line with the proverbial emphasis on the importance of “set and setting” for the
consumption of psychedelic substances, Langlitz and colleagues discuss research reporting
different experiences associated with the use of hallucinogens in different social
contexts (“settings”). In what has become a classic study, Wallace indeed found that
the experiential content of dreams and hallucinations depends on one's cultural and
social background (8). Similarly, in another classic study, Bourque and Back found
that lower-educed people with lower incomes in the USA described transcendental experiences
more in religious terms, whereas higher-educated people with higher incomes used more
aesthetic concepts (9). More recently, a comparison between schizophrenia patients
from more or less religious regions in East and West Germany showed that the former
reported less religious content in their hallucinations than the latter (10).
Langlitz and colleagues, like other researchers, subsequently distinguish pharmacological
and extra-pharmacological effects, particularly in research on psychedelics (1, 11–13).
But actually already a classic social-psychological study with adrenaline investigated
“cognitive, social, and physiological determinants of emotional state” (14). Subjects
either received correct, incorrect, or no information about the expected physiological
responses (such as increased heart rate, feeling warmth) of an adrenaline injection
presented to them as a vitamin shot.
1
In a following funny social interaction, they reported significantly more euphoria
and were more active without the correct information, that is, when they had no rational
explanation of the physiological symptoms. A similar outcome was reported for an anger-inducing
social interaction. Thus the outcome variable (here: emotional state and behavior)
depends on an interaction of the expectation, social setting, and pharmacologically
induced physiological state of the subjects.
2
Research on the placebo effect provides more examples. In recent years, scientists
focused on explaining its psychobiological mechanisms and clinical potential (15,
16). Some actually criticize the notion of placebo as an “inert substance” as inconsistent,
for something inert can conceptually not have any effects. They thus propose the alternative
notions of a “context effect” or “meaning response” (15, 17–20), where the former
expression emphasizes social context and the latter term individual beliefs or expectations,
thus precisely what Langlitz and colleagues refer to when speaking of extra-pharmacological
effects. Strikingly, research has shown that the context effect or meaning response
can consist in activating the same physiological pathways as drugs used to alleviate
a certain medical problem, such as the activation of endogenous opioids and dopamine
for pain treatment (15).
Langlitz and colleagues refer to different epistemic cultures in the natural sciences
on the one hand and the humanities and social sciences on the other, with common dichotomies
of nature/culture or matter/mind. Consistent with a theoretical framework presented
by Greenberg and Bailey, I take the stance which Turkheimer coined “weak biologism”
(21, 22). This means that there is no strict dichotomy between the biological and
non-biological, because in some sense all of our perceptions, thoughts, behaviors,
and the like are biological—if only in that they have a biological basis, because
we are embodied beings (23–27). These bodies were shaped through an evolutionary history
and particularly their nervous systems enable a wide range of psychological and cultural
possibilities, which in turn also influence biological structure and function (i.e.
neuroplasticity). For example, certain brain (and other physiological) structures
allow us to acquire language; without them, we could not understand and express it.
But it's the psychosocial context determining whether one's primary language will
be, say, Chinese, English, or Spanish.
Applied to Langlitz's and colleagues' thoughts, this means that there is also no strict
dichotomy between pharmacological and extra-pharmacological factors: Just as anthropological
research-at least implicitly—always includes (and presupposes) certain bodies and
brains, pharmacological research always includes (and presupposes) a psychosocial
context, often a clinical setting in which a substance is administered. In other words,
our cognitive and emotional processes are not only (physically) embodied, but also
(psychosocially) embedded (23–27). Kaptchuk already described how, from an anthropological
perspective, common procedures even in Western medical systems could be described
as a “ritual”, thus a particular psychosocial context affecting the treatment effects
(28).
3
We may tend to overlook this, because we take that context for granted. It then also
makes sense that research on psychedelic substances in particular prompts scholars
to focus more on context effects, as these substances (like ayahuasca, religiously
used in South America and consisting of N,N-Dimethyltryptamine [DMT] and a monoamine
oxidase inhibitor [MAOI]) often originate in different cultures with particular practices
of consumption or “rituals” (13, 29–31).
The upshot of my proposal is that there is no intrinsic contradiction between the
tasks of anthropologists, psychologists, or pharmacologists: The latter often won't
investigate the psychosocial context (as an independent variable, that is) simply
for the reason that the respective drug is supposed to be taken within a particular
context. Thus, in principle, if pharmacologists became interested in cultural differences
or specific effects of psychosocial contexts, they could simply include them in their
experimental designs. Whether the resulting discipline deserved a new denomination
like “transcultural pharmacology” (cf. transcultural psychiatry) or “pharmanthropology”
lies in the eye of the beholder. Strictly distinguishing pharmacological and extra-pharmacological
effects, by contrast, carries the risk of reintroducing unnecessary dichotomies; unnecessary,
because, as we have seen, the former never comes without the latter.
One could speculate, though, whether the biological (or pharmacological) domain has
some primacy in the sense that without the enabling physiological structures and functions
there would simply be no anthropological, social, or psychological domain. Experimentally
this could be exemplified—within ethical boundaries—by increasing pharmacological
doses and thus eventually overriding or at least minimizing effects of psychosocial
context (32). Anesthesia is a clear example, as it transiently disables certain psychosocially
necessary functions—and thus psychosocial processes. In the words of a patient participating
in one of Delgado's early brain stimulation experiments: “I guess, Doctor, that your
electricity is stronger than my will” (33, 34). But this does not make the case for
“strong biologism”, on Turkheimer's account, which would mean that psychosocial functions
could be explained completely or at least for a large part in biological terms. Similar
to how Steven Hyman, former director of the US National Institute of Mental Health,
recently characterized psychiatric disorders, one could say: The psychosocial processes
of pharmaceutical drugs are grounded, but not exhausted in biology (35).
Moral Psychopharmacology and Values
Langlitz and colleagues continue to discuss a possible “moral psychopharmacology”.
Indeed, after neuroscientists began to investigate moral decision-making (36–40),
pharmacologists also addressed that domain (41, 42) and ethicists speculated about
“moral enhancement”, the possibility of using drugs to improve people's moral capacities
(43, 44). The ecological validity of the moral dilemmas often used in such studies
and their (alleged) social implications have subsequently been discussed critically
(45–48). Here it helps, in my view, to distinguish different meanings of “moral” on
the one hand and individual or collective perspectives on morality on the other.
In a loose sense of “moral”, as pertaining to moral implications, the term is more
or less equivalent with “social”. Then, according to Langlitz's and colleagues' call
for more awareness for the potential effects of psychedelic drugs in the social domain,
one could say that this is valid for psychoactive drugs in general. Unless one imagines
the life of a hermit, it would be difficult to think of a psychopharmacological application
that could not, in principle, have any social implications (e.g., think of the possibility
of substances to interfere with people's capacity to control vehicles or machines,
which could in turn harm themselves or others and thus become socially relevant).
Here it would be important to distinguish transient and permanent effects (with the
latter possibly altering personality). It would still make sense, as Langlitz and
colleagues suggest, to investigate the potential effects of psychedelic (and other
psychoactive) drugs on social cognition. But this should be distinguished, in my view,
from a “moral psychopharmacology” in a narrower sense: that is, one that specifically
aims at improving subjects' moral cognition. To my knowledge, pharmacological experiments
so far used the previously mentioned moral decision-making paradigms, but did not
explicitly try to enhance moral competence as operationalized by, for example, Lind's
Moral Judgment Test (49, 50). The general problem remains that there is no single
accepted standard of what a “good” moral decision is; this would always presuppose
a particular moral theory or stance, of which there are many different ones competing
in moral philosophy.
Before improving moral capacities, there thus has to be a value judgment on what kind
of moral capacity is deemed desirable. Langlitz and colleagues, just like Evers before,
particularly address the possibility of increasing empathy on the neurobiological
level (1, 51–53). But this raises the question whether more empathy is always morally
good. It could make people prone to overrate the preferences of others at the cost
of their own wellbeing (54). What if egoists or “successful psychopaths” who hardly
care for the interests of others as a value in itself disagree to become “morally
enhanced” in this sense? This also raises the question who is to decide: Without informed
consent of the subjects themselves, some would consider “moral psychophamarcology”
as coercive or even totalitarian. And based on informed consent there would probably
always be individuals deciding against such an intervention, simply because people
have different values and think differently about the instrumental use of substances.
This dilemma was actually already debated by psychiatrist Klerman and bioethicist
Veatch half a century ago (5–7). Different value systems and different understandings
of human nature shape people's views on drug use differently. The category not addressed
by Langlitz and colleagues, but central in medical ethics, is that of autonomy. People
should primarily decide for themselves, but also considering their own psychosocial
context (55). From this perspective, “moral psychopharmacology” could be understood
as one example of instrumental substance use more broadly conceived (2–4). People
might use drugs to become more moral—if and only if they themselves desire so.
Conclusion
On the theoretical level, psychosocial effects (contexts and beliefs/expectations)
on how a drug (psychedelic or not) work are neither surprising nor an insurmountable
problem. When humans are conceived as a psychobiological unity in a social environment,
“weak biologism” or the idea of the psychosocial domain being grounded in biology
predicts precisely that: that psychological processes are embodied and thus also reflected
in a subject's physiological state (23–27). An example from research on the placebo
effect has shown that the physiological pathway can actually be the same as the one
activated by a drug to treat a particular medical problem (15). This also exemplifies
the clinical relevance of understanding the effects of context or beliefs/expectations
on drugs. Generally speaking, there is no reason why anthropologists should disregard
pharmacology or why pharmacologists should neglect psychosocial context—if that becomes
salient for their research questions. Whether this justifies the establishment of
a “transcultural pharmacology” or “pharmanthropology” as an independent sub-discipline
may be a matter of taste. But from the point of view presented here, the context-dependency
of psychedelic drug effects rather means opportunities, not problems for anthropology
and pharmacology.
Author Contributions
The author confirms being the sole contributor of this work and has approved it for
publication.
Funding
This publication has been supported by the History of Neuroethics grant by the Dutch
Research Foundation (NWO), Grant Number: 451-15-042.
Conflict of Interest
The author declares that the research was conducted in the absence of any commercial
or financial relationships that could be construed as a potential conflict of interest.
Publisher's Note
All claims expressed in this article are solely those of the authors and do not necessarily
represent those of their affiliated organizations, or those of the publisher, the
editors and the reviewers. Any product that may be evaluated in this article, or claim
that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.