Skip to Content
PhilosophiesPhilosophies
  • Article
  • Open Access

11 January 2025

14 Pages

Social Media, Informed Consent, and the Harm Principle

Faculty of Law, University of Oxford, Oxfordshire OX1 3UL, UK

Abstract

This article examines whether social media users can validly consent to their own use of social media. It argues that, whether or not social media use is analogous to public health interventions, there is an obligation to provide users with information about risks and benefits, and absent that provision, there is no valid consent. Many or most users, in any event, do not have the capacity to consent, according to the criteria for capacity articulated in the ‘four abilities’ model: the ability to express a choice, the ability to understand the facts pertinent to the decision in question, the ability of a subject to believe that the information applies to them, and the ability to reason—in the sense of being able to consider and weigh (with reference to the patient’s own concerns, circumstances, and values) the main possible outcomes of the decision to opt for the intervention and the decision to opt not to undergo it. Even if an individual social media user is capacitous according to these criteria, many will fail to be judged capacitous if (as it is argued should be the case), a further criterion, identified by Jennifer Hawkins must be satisfied, namely that the individual can look after their own interests at least as well as most other people can. It follows from this consideration that not only can regulation of social media (in the form of a ban) be justified under Mill’s harm principle, but that non-regulation cannot be justified.

1. Introduction

Almost all Western teenagers use some sort of social media. This article focuses on use by teenagers. The literature on use by adults is much more sparse, but there is no reason to suppose that adults do not face at least some of the dangers social media poses to teenagers, or that (since the issue canvassed here is the effect of social media on the ability to consent to social media) the ethical or regulatory consequences are materially different. Nearly 97% of US teenagers have at least one electronic item in their bedroom [1] and 60% of time after school is spent on some form of screen media activity [2].
There is a vast body of literature dealing with the effects of this activity, and this article cannot adequately summarise it. Generalisations are difficult and dangerous, but one can confidently be made: social media use, viewed on a population level, does more harm than good [3,4,5]. In some individuals, a sense of community and connectedness can be beneficial, but detrimental effects are sufficiently common and sufficiently serious to outweigh any benefit to the cohort of users as a whole.
Generally, the deleterious effects of social media activity (SMA) are greater in females than in males, in younger than in older adolescents, and are more associated with passive SMA than forms that involve active engagement. They have been systematically reviewed by Paulus et al. [4], who emphasised the complexity of questions relating to the effects of SMA and the need to see SMA as one element in a multifaceted ecosystem within which teenagers live, but nonetheless noted a significant association between SMA and mental health issues (and particularly depression and anxiety) and between SMA and both short- and long-term effects on sleep duration and quality.
Twenge et al. noted that amongst girls in particular there is a consistent and substantial association between SMA and mental health issues, which were statistically stronger than links between mental health and injectable drug use generally, obesity, exercise, having four or more sexual partners, and having sex before the age of 13, and comparable in size to the effects of binge drinking, sexual assault, and the use of heroin [6,7,8].
Van den Eijnden et al. devised a ‘Social Media Disorder Scale’ based on the architecture of the DSM5’s definition of Internet Gaming Disorder, and using the criteria of compulsive internet use, self-declared social media addiction, self-esteem, depression, attention deficit, impulsivity, loneliness, and the frequency of daily social media use [9]. In three samples of disordered social media users (all teenagers), one sample showed more disordered boys than girls, but this gender difference was not replicated in the other two samples. It may be, then, that a holistic assessment of social media disorder, rather than an examination (as in the Twenge et al. analysis [6]) of less specific mental health disorders, might eliminate the apparent over-representation of girls as psychiatric victims of social media, and hence boost the case for intervention across the board.
Comparisons with the use of addictive drugs invite the question of whether addiction to social media use is a distinct pathology. There is mounting evidence that it is. Discussion is bedevilled by difficulties of definition and by (vain) dispute about whether various degrees of dependence on social media amount to a recognisable psychiatric illness [10]. Whatever the small print of the definition of social media addiction, the prevalence, though varying widely between nations, is alarmingly high. Cheng et al., in a meta-analysis, reported a pooled prevalence estimate in ‘collectivist’ nations of 31%, which is two-fold higher than in ‘individualist’ nations (14%) [11].
Dependence seems, unsurprisingly, to be linked to particular online activities rather than to the internet itself [12], and social media use and gaming seem to be particularly addictive. The American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders 5 (DSM-5) 2013 [13] recognises Internet Gaming Disorder, as does the latest DSM iteration, DSM-5-TR (2022) [14].
Identification of the contribution of social media to psychiatric dysfunction is complicated by the fact that many individuals with ‘Internet Addiction’ have other psychiatric diagnoses too [15] (88.3% had an average of two other diagnoses in one study) [16]. This fact has led some to see internet-related disorders as mere manifestations of other conditions, but it is increasingly recognised that internet disorders should be regarded as distinct psychiatric entities, with clear parallels with conventionally recognised disorders, such as impulse-control disorders or addictive behaviours [17,18,19,20].
Twenge et al. addressed the question of whether the effects they identified justified some sort of public health intervention in relation to social media use [6]. There were, they noted, two relevant questions: one empirical and one theoretical. The empirical question relates to the size of the relationship between SMA and the damaging effects on mental health. (I note that they assume that any positive effects of SMA should be discounted, either because they are too small in themselves, or because they are disproportionately small compared to the detriments. I have already indicated my agreement with the second of those propositions.)
In addressing the empirical question, they noted that the relationship between SMA and mental health issues was greater than that for screen time per se and that there is an emerging consensus that it was generally considered to be more than r = 0.10 in absolute value.
The theoretical question is whether an effect of that size warrants any action by parents or policymakers. Twenge et al. acknowledge that the much-followed rule of thumb articulated by Cohen et al. [21] suggested that correlations between 0.1 and 0.3 should be regarded as ‘small’, those of 0.3–0.5 as ‘moderate’, and those of 0.5 and above as ‘large’, but observe the evolving criticism of this rule as a basis for intervention [6]. Effects which are ‘small’ in Cohen’s terms can have an immense cumulative impact when applied in a large population over a long period of time. Twenge et al. cite the example of the association between childhood exposure to lead and adult IQ (where r = 0.09—‘small’ in Cohen’s terms). The removal of lead has nonetheless had a significant effect on the part of the population most at risk. They calculate the overall risk to adolescent girls of exposure to social media as 0.20 and assert that the case for intervening to restrict SMA is at least as strong as that for lead reduction [6].
There is evidence that, for some users, some types of social media may in some respects be beneficial. Foerster et al. found that youths who played video games could allocate their temporal and spatial attention better than those who did not [22]. Sauce noted a positive (though small) relationship between gaming and indices of generalised intelligence. Children who played more video games at the ages of 9–10 showed greater gains in intelligence 2 years later [23]. Huang et al. noted that the ‘pervasive connectivity’ brought about by smartphone use could be advantageous, and might in some respects be ‘conducive to their mental health’, but commented that online interactions might supplant face-to-face interactions, thus offsetting the potential advantages of smartphone use for mental health [24].
This article is not primarily concerned with the question of whether social media (a term that covers a wide variety of different activities, to each of which different considerations may apply when considering the risk/benefit balance) does more harm than good. I have noted the elements of the empirical debate about risk and benefit simply to set the following considerations in context. I certainly do not purport to resolve that debate.
I consider here two opposing possibilities, each painted with a very broad brush. I then identify and discuss the ethical considerations relating to each.
The first is that, on balance, and considering the cohort of all users, social media do more good than harm. By way of shorthand, I will simply call this situation ‘Social media are good’. The second is that, on the same basis, they do more harm than good. I will call this ‘Social media are bad’.
I begin that consideration by identifying situations analogous to social media. This will help by locating the discussion of the relevant ethics in more familiar and less febrile contexts.

2. Should Social Media Be Considered in the Framework of Public Health Ethics?

If social media are good, the goods identified in the literature are, as described above, increased connectivity and, possibly, the enhancement of some cognitive functions. There is also an arguable ‘political good’ associated with permitting social media use on the basis that prohibition of any activity is in principle to be deprecated, but that ‘good’ only falls into account when considering the (logically later) question of whether regulation can be justified.
If social media are good, their use is analogous to the addition of fluoride to the public water supply, which is widely (but not always uncontroversially) accepted for the reduction of dental caries. There is an option—though an expensive and onerous option—not to drink water from the public water supply. Social media are as ubiquitous as is water from the public system.
This analogy rests on the premise that conferring a benefit (increased connectivity and enhanced cognition in the case of social media) is ethically comparable to the avoidance of a detriment (dental caries in the case of fluoridation). This premise seems to me to be reasonable, not least because the detriment avoided by fluoridation can equally well be described in terms of a benefit (healthy teeth), and the benefit conferred by social media can equally well be described in terms of avoiding a detriment (reduced connectivity and relatively impaired cognition).
If social media are bad, they are usefully analogous to the contamination of the public water supply with a toxin such as lead. This is a better analogy than the one suggested by Twenge et al. [6], of air contaminated with lead, because while not drinking from the public water supply is an option, not breathing the air is not. The only option in relation to contaminated air is to move to a distant place, which is analogous, in the case of social media, to the situation of successfully banning all access to the internet. Such a ban is easier to state than to enforce: it would mean restricting not just access to devices at home and at school, but also restricting access to friends who had such devices.
If these analogies are sound, the debate about the ethics of social media use can be framed in terms of public health ethics. And if this is so, the notion of informed consent is central to the debate. The notion has been described as the ‘bedrock of clinical bioethics’ [25] (p. 5). Its place outside the clinic is more controversial, as we will see.

6. Social Media and the Four Abilities

On the basis of our presumption about the consensus relating to social media risks and benefits, many social media users will not have the capacity to consentvalidly to the use. They may well fall at the second and/or third and/or fourth hurdle.
So far as the second ability is concerned, the facts (as we have seen in the brief review above of the literature about risks and benefits) are complex and contentious, and the field is fast evolving. Few specialists in the field would purport to know all the nuances. Some of those nuances may be relevant to the decision-making of an individual.
The third ability may be relevant. Extrapolation from data pertaining to a cohort may be particularly difficult for the adolescents most likely to be affected by social media. Since engagement with social media is perceived to be (and to some degree is) voluntary, the user is likely to believe that whatever the problems experienced by others, they themselves have control over their usage rather than being controlled by it [35,36].
The fourth ability will be engaged in those social media users who show some signs of SMA. Addiction, by definition, involves an activity or a substance overriding normal psychological methods of restraint. The ability to reason—in the sense of conducting some sort of audit of risks and benefits and some kind of dispassionate use of the results of that audit to one’s own situation—is impaired in addiction. In a study of Turkish students who perceived themselves as social media addicts—and therefore acknowledged at some level that social media presented a problem for them—23 out of the 25 students in the study stated that they would definitely not give up using social media. Their comments included the following:
‘I do not think about giving social media up because I think that it contributes to my personal development, it is not harmful to me, I am having fun, I am spending time, why should I give up?’
‘I do not think about giving social media up, it complements my daily life; giving it up means to cut off all communication with my friends. I would give up if I only want to stop communicating with my friends, there needs to be a valid reason.’
‘I have social media friends that may not be my real friends but I share the same feelings and thoughts; this makes me social and even they [sic] are more concerned than my real life friends. I definitely not [sic] consider giving it up.’
‘Give social media up!! Why would I give up, it does not cost anything, I can follow the current events without watching TV, I communicate with my friends, I meet people whom I share [sic] the same thoughts, I would use it more if they make the internet connection fees a little bit more reasonable’ [37] (p. 863). These comments in themselves might not connote SMA, any truncation of autonomy, or indeed any problem (it seems likely that similar observations would be made by social media users who do not regard themselves as suffering from SMA), but taken in the context of the users’ acknowledgement that their social media use was problematic, are perhaps best seen as illustrations of the kind of justification of problematic use that might mask real and corrosive interference with autonomy.
Levy, writing about addiction in general, is curtly dismissive of the suggestion that addiction might undermine autonomy. ‘At worst, addicts are subject to impairments of autonomy… [when] they encounter cues that trigger their cravings… [but] for the great majority of time… addicts are as autonomous as you and me’ and hence, he says, have a capacity for informed consent [38]. Norlock contends that the very nature of social media means that those cues continually bombard the user, meaning that users who have entered the social media bubble freely are often not free to leave. She writes, ‘[O]ur capacities are literally undermined by engineered ludic loops so that we do not always retain competence to consent’ [39].
If my observations about the applicability to social media use of the four abilities are correct, then at least many users cannot, using those criteria, be regarded as giving informed consent to its use.
It may be that some social media users whose use is to some degree problematic may nonetheless manage to satisfy all the criteria. They may, perhaps despite an acknowledgement of a problem (showing insight, which contributes positively to the third and fourth ability components), continue to want to use social media. Is this then not a classic illustration of the legitimate use of one’s autonomy to do something personally harmful in an act of self-sovereignty? (I leave aside in this article the argument that the autonomous participation in social media by an individual that is itself legitimate may, because social media are social and thus depend on interaction, become ethically unacceptable because of the actual or potential risk to more vulnerable constituencies of social media per se, or even of that individual’s particular contribution to social media.)
Jennifer Hawkins has contended, in a clinical context, that even where the four abilities are satisfied, it may nonetheless be appropriate to regard an individual as lacking capacity [40]. (I have criticised her analysis insofar as it relates to the structure of legal decision-making, but endorse her bioethical conclusions.)

7. Incapacity Beyond the Four Abilities

Hawkins centres her discussion around two fictional—but realistic—scenarios.
The first concerns a young man (‘Terence’) who has anorexia nervosa. If he is not medically fed he will, as he knows, die. He knows and understands his diagnosis. But he says that he would rather die than put on weight. The MacCAT-T assessment concludes that he has capacity. So far as the fourth ability is concerned, he can ‘reason’ sufficiently to pass the test. Given his preference to die rather than gain weight his choice, says Hawkins, is ‘broadly speaking, instrumentally rational—it makes sense given the values he currently has and the priority he gives them.’
The second concerns a woman in her mid-50s (‘Donna’) who has type 1 diabetes which, a few years earlier, caused her to need an above-knee amputation. She did well after the operation, and after an episode of depression, adapted fully to the new situation. She works as a visual artist, which she finds interesting, and is socially active, with a number of close friends. She is told that she will need a second above-knee amputation. Initially, she deals well with the news and has the surgery. The clinical outlook is good. But post-operatively, she tells the clinicians that she wants only treatment going to her comfort, and that she does not want the standard post-operative protocols of antibiotics and anticoagulants. If she gets an infection, she says, she wants it left untreated, even if that means her death.
A psychiatric assessment reveals that while she was in hospital, Donna’s husband told her that he was leaving her for someone else, and would be moving out of their home.
In lay terms, comments Hawkins, we would say that her decision about treatment was made when she was ‘in shock’. Her emotions are in turmoil, and so she is ‘not currently attending to, or giving weight to, her future as it would most likely be’. Hawkins rightly identifies the question as it would be framed in terms of the four abilities: does she have the capacity right now to make the decision about treatment? The psychiatrist concludes that she does. Shortly after the assessment, Donna develops an infection. It is untreated, and she dies.
Hawkins notes three ethical considerations underpinning the four abilities model. The first is ‘value neutrality’, which seeks to ensure that patients’ values are respected, even if they differ radically from those of the clinician, and even if they are objectively highly eccentric and life-threatening. The second is ‘Diagnostic neutrality’, which means that capacity should not be determined simply on the basis of a diagnosis. One should not conclude that a patient lacks capacity only because they suffer from a condition that commonly truncates capacity, such as dementia or depression. As implied above, I would want to insert a caveat to diagnostic neutrality in clear cases of addiction. There, at least in relation to the activity to which the addiction relates (for instance taking opiate drugs or alcohol or giving up one’s life and mental health to social media), the diagnosis surely connotes incapacity (at least in relation to the fourth ability). It is of course true, though, that the incapacity of an addict in relation to the specific activity that is the subject of the addiction does not—or does not necessarily—connote incapacity in relation to any other domain.
The third ethical consideration is what Hawkins calls ‘inclusivity’, by which she means that in assessing capacity one must be careful not to set the bar too high. This is easily accomplished if one relies on idealised models of decision-making. We should start with an assumption of capacity, and conclude incapacity only if this assumption is convincingly rebutted.
It is self-evident how these considerations operated in Terence’s and Donna’s cases.
But did the four abilities and their underlying principles achieve an ethical result in either or both of these cases? Many would intuitively say that they did not; that both Terence and Donna have been victims of an unethical formalism, and that they should and could have been protected from themselves, or the self that, in the narrow window afforded by the clinical circumstances, was making the relevant decision. Each case is crucially fact-specific, but certainly in Donna’s case, and possibly in Terence’s, it seems that they have identifiable (and in Donna’s case, self-identified) goods which can be preserved by a clinical intervention, and there are good reasons to suppose will be actively enjoyed by each patient should they survive to enjoy them. These sorts of considerations appear to fall foul of value neutrality, and possibly ‘inclusivity’ too. Is it offensively paternalistic to seek to accommodate these considerations? Can and should the four abilities approach be revised?
Hawkins exhorts us to ask why we value decisions that satisfy certain formal criteria. It is a question so foundational that it is almost never asked. She notes, following Buchanan and Brock [41], that we need to assess capacity because of a tension between well-being and free choice. Any of us might, on occasion, make a dangerously foolish choice—a choice deeply inimical to our well-being. The only way to protect against this possibility is to limit free choice drastically. Unrestricted free choice would endanger some vulnerable individuals with a suboptimal ability to make objectively sensible decisions. How should we resolve the tension? How should we strike the balance? However we do it we must, says Hawkins, remember the presumption of capacity. If we do, she proposes the following resolution: the individuals we need to protect are those ‘who are less able to look after their own interests than most ordinary people are’, i.e., individuals who are more likely than others to make ‘serious prudential mistakes’. This involves expressing the object of medical decision-making in terms of welfare. A patient with relevant decision-making capacity is therefore, she says, one who can look after their own interests ‘at least as well as most other people can’.
Hawkins goes on to consider (and to my eye deal convincingly with) the objection that the patient’s own welfare is not always the goal of medical decision-making. A patient might choose less good care (for instance) because to do so would save money. These considerations are unlikely to apply directly to the social media context, and I do not consider them here.
Hawkins proposes that the traditional four abilities should be regarded as necessary for capacity, but that an individual who has all four capacities should be regarded as capacitous if, but only if, two further requirements are both satisfied. These are as follows:
First: there must be good evidence for thinking that the individual is making a serious, prudential mistake here and now in making the decision she has made.
Second: the individual must be known to have a condition that is known to make those who have it more likely than usual to make prudential mistakes—in Donna’s case, ‘being in shock’; in Terence’s case, anorexia nervosa.
Applying Hawkins’ criteria to the social media context, we can see that social media users deemed capacitous according to the four abilities will often fail to clear the other two hurdles to deemed capacity erected by Hawkins. On the basis of our presumption about the badness of social media, a decision to continue to use it will be a serious, prudential mistake. Even a decision to use it made on the basis of equivocal evidence about badness may be such a mistake, because the risks, if real, are risks of serious conditions. This may be so even if there is equivocal evidence of the overall benefit of social media. Here, the identifiable potential benefits will be small, but the identifiable potential detriments very significant.
Hawkins’ second criterion, relating to a condition known to predispose to prudential mistakes, should uncontroversially apply to cases of SMA. Since social media generally are known to be addictive in a sense falling short of a recognisable psychiatric illness, the criterion may apply more broadly too.

8. Discussion

Informed consent, as O’Neill demonstrated, is ubiquitous. Although we tend to think of it in clinical contexts, it infuses most transactions, formal and informal, in everyday life. Medical examples—both in clinical and public health contexts—are helpful simply because much of the academic deliberation on informed consent has been in those contexts. There is no reason to suppose that the principles of informed consent articulated in medical contexts should not apply outside those contexts and every reason to suppose that they should.
I have detailed above the various ways in which informed consent has been justified. It may help to paint with a rather broader brush. There is, as McNish has illustrated [42], a debate between Beauchamp and Childress, who see the doctrine as enshrining and respecting respect for autonomy [43], and Monson and O’Neill, who see it as primarily concerned with the limitation of harm [44].
I need not enter the lists on either side. Both sets of protagonists support the contention in this article. Beauchamp and Childress should be concerned because social media use appears to make users less autonomous; I would expect Monson and O’Neill to join me in regretting the harms that social media can cause, and in suggesting that the notion of informed consent, properly deployed, might reduce the incidence or mitigate some of the severity of those harms.
Social media usage may in some respects be analogous to interventions in the public health contexts. If social media are on balance beneficial, they may be analogous to, for instance, the fluoridation of water. If they are on balance detrimental, they may be analogous to the presence of toxins in the water.
There are difficulties in applying the whole doctrine of informed consent to policymaking in public health (because individuals often cannot opt out of public health policies). But there are, as demonstrated in this article, public health policies based on some of the effects of SMA from which an individual may opt out. One can opt out of social media, and thus to the extent to which the public health analogy applies to social media, social media are an example of ‘interventions’ to which the undiluted doctrine of informed should apply. There is, in any event, widespread agreement that even when an informed decision is not possible (e.g., where there is no possibility of opting out), there is still an ethical obligation to provide information relevant and pertinent to the risks posed by the relevant intervention.
Applying the criteria of informed consent as deployed in public health contexts, then, there is an ethical duty to inform about the risks associated with social media use.
Because, per O’Neill [6], informed consent is a principle of general application in transactions of all kinds, a duty to provide pertinent information about risks and benefits (in substance identical to a ‘medical’ duty) applies to social media risks even if the public health analogy is defective.
This information is not routinely supplied. This amounts to a breach of an ethical duty.
Many social media users may not, in any event, have the capacity to consent to their use, where capacity is judged according to the conventional ‘four abilities’ and, even more so, if it is judged according to those capacities with the addition of the two further criteria identified by Hawkins.
However one looks at the issue, then, and whatever the empirical evidence about the risks and benefits of social media use, social media users do not give informed consent to the use (because they are not supplied with the relevant information about risks and benefits), and/or (in many cases) could not in any event be regarded as having the capacity to give consent.
These conclusions have obvious corollaries for regulation.
We should be slow to regulate.
The most strenuous test for a would-be regulator is the test in Mill’s harm principle, which states that ‘the sole end for which mankind are warranted, individually or collectively, in interfering with the liberty of any of their number is self-protection’ [45]. Interference will be unjust unless it is necessary to prevent harm to others.
Whether the threshold for regulation is passed will depend not only on the probability of an identified risk eventuating but also on the magnitude of the harms/wrongs that would occur (both in terms of the consequences to individuals and the number of individuals affected if the risk eventuates). Moreover, passing the threshold for legitimate regulation will also depend on the same criteria being applied to potential but non-identified risks, and the same criteria being applied conversely to the benefits of the intervention but including the political harms/wrongs of intervention and the harm to autonomy interests.
It is not only legitimate but necessary to incorporate the precautionary principle into the operation of the harm principle in order to give practical effect to the harm principle [46]. The harm principle, properly construed, cannot require evidence beyond reasonable doubt of serious and irreversible harm.
When applying these criteria in the case of social media in relation to the causation of mental ill-health, the question of whether or not a ban is justified or mandatory involves the determination of empirical questions that are strictly outside the scope of this article. It follows, though, from what I have already stated (and without taking the informed consent arguments into account: I will come to their relevance shortly) that a strong empirical case in this respect is building against social media. I suggest that it is likely that the harm principle simpliciter (without the support of the precautionary principle), on the basis of current empirically based understandings of the risks and benefits, justifies regulation.
If, contrary to my suggestion above, the current state of the evidence does not justify a ban based on the harm principle alone, but the harm principle and the precautionary principle do indeed work in synergy, there can, on the view of the evidence most favourable to social media, be no principled objection to regulation.
And fairly draconian regulation at that. It is hard to see how, practically, any regulation falling short of a complete ban could vindicate the harm principle if it applies.
Even if a social media user could demonstrate that she was entirely immune to all the identified dangers, it would still seem likely that regulation in the form of a complete ban would be justified by the harm principle since social media are by definition social, and the activities of a wholly immune user would inevitably expose potentially vulnerable users to harm.
If the arguments articulated above about informed consent are correct (or, if the precautionary principle applies may be correct), then two further forms of harm (and/or, possibly wrongs) fall into account when considering regulation. First, the harm to individual social media users of undertaking an activity to which they have not consented and/or, as the case may be, cannot validly consent. This is a harm to the individual’s perceived welfare and/or autonomy interests, and may well constitute a wrong too. And, second, to permit an activity to which a large sector of participants have not consented and/or cannot validly consent has a wider societal fallout. As we discussed when considering utilitarian justifications for informed consent, society as a whole will benefit if informed consent and related autonomy interests are vindicated, and correspondingly, worse off (justifying regulation) if informed consent and autonomy interests are not protected.
If the case for regulation gets past John Stuart Mill, it necessarily follows that all the less stringent tests for regulation would also be satisfied.
It is possible to imagine situations where regulation is justified, but also non-regulation can be justified. (Many situations falling into the ‘margin of appreciation’ in the jurisprudence of the European Convention on Human Rights will be examples.) But given the factual background of the debate about social media harms, it is hard coherently to suggest that a compelling case (such as I have argued there is) for regulation does not necessarily imply a compelling case against non-regulation.
O’Neill, as discussed above, saw informed consent as part of everyday life: a principle grafted into every transaction, formal and informal, clinical, commercial, and otherwise. I agree. And if O’Neill is right, the corollaries of the conclusions in this article go far beyond the social media context. Problems with social media are just a convenient way to highlight the way in which informed consent, as articulated by Hawkins, is routinely violated. To follow the strands of that thought into political philosophy and jurisprudence is well beyond the scope of this article.

9. Conclusions

It seems likely, from what we know (and from what we do not know) of the risks and benefits of social media use, that users do not, using the conventional canons of construction of informed consent, validly consent to it. Many will not have the capacity to do so.
Even applying the most stringent test that potential regulation of social media (realistically, a total ban) can face (Mill’s harm principle), not only is regulation justified, but non-regulation is not justified.

Funding

This research received no external funding.

Institutional Review Board Statement

Not applicable.

Data Availability Statement

No new data were created or analyzed in this study. Data sharing is not applicable to this article.

Conflicts of Interest

The author declares no conflicts of interest.

References

  1. Hale, L.; Guan, S. Screen Time and Sleep among School-Aged Children and Adolescents: A Systematic Literature Review. Sleep Med. Rev. 2015, 21, 50–58. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  2. Arundell, L.; Fletcher, E.; Salmon, J.; Veitch, J.; Hinkley, T. A Systematic Review of the Prevalence of Sedentary Behavior during the After-School Period among Children Aged 5–18 Years. Int. J. Behav. Nutr. Phys. Act. 2016, 13, 93. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  3. Montag, C.; Demetrovics, Z.; Elhai, J.D.; Grant, D.; Koning, I.; Rumpf, H.-J.; Spada, M.M.; Throuvala, M.; Van Den Eijnden, R. Problematic Social Media Use in Childhood and Adolescence. Addict. Behav. 2024, 153, 107980. [Google Scholar] [CrossRef] [Scilit]
  4. Paulus, M.P.; Zhao, Y.; Potenza, M.N.; Aupperle, R.L.; Bagot, K.S.; Tapert, S.F. Screen Media Activity in Youth: A Critical Review of Mental Health and Neuroscience Findings. J. Mood Anxiety Disord. 2023, 3, 100018. [Google Scholar] [CrossRef] [Scilit]
  5. Fumagalli, E.; Shrum, L.; Lowrey, T.M. The Effects of Social Media Consumption on Adolescent Psychological Well-Being. J. Assoc. Consum. Res. 2024, 9, 119–130. [Google Scholar] [CrossRef] [Scilit]
  6. Twenge, J.M.; Haidt, J.; Lozano, J.; Cummins, K.M. Specification Curve Analysis Shows That Social Media Use Is Linked to Poor Mental Health, Especially among Girls. Acta Psychol. 2022, 224, 103512. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  7. Twenge, J.M.; Haidt, J.; Joiner, T.E.; Campbell, W.K. Underestimating Digital Media Harm. Nat. Hum. Behav. 2020, 4, 346–348. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  8. Heffer, T.; Good, M.; Daly, O.; MacDonell, E.; Willoughby, T. The Longitudinal Association between Social-Media Use and Depressive Symptoms among Adolescents and Young Adults: An Empirical Reply to Twenge et al. (2018). Clin. Psychol. Sci. 2019, 7, 462–470. [Google Scholar] [CrossRef] [Scilit]
  9. Van Den Eijnden, R.J.J.M.; Lemmens, J.S.; Valkenburg, P.M. The Social Media Disorder Scale. Comput. Hum. Behav. 2016, 61, 478–487. [Google Scholar] [CrossRef] [Scilit]
  10. Pies, R. Should DSM-V Designate “Internet Addiction” a Mental Disorder? Psychiatry 2009, 6, 31–37. [Google Scholar] [PubMed]
  11. Cheng, C.; Lau, Y.; Chan, L.; Luk, J.W. Prevalence of Social Media Addiction across 32 Nations: Meta-Analysis with Subgroup Analysis of Classification Schemes and Cultural Values. Addict. Behav. 2021, 117, 106845. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  12. Griffiths, M.D.; Szabo, A. Is Excessive Online Usage a Function of Medium or Activity? An Empirical Pilot Study. J. Behav. Addict. 2014, 3, 74–77. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  13. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 5th ed.; American Psychiatric Association: Richmond, VA, USA, 2013. [Google Scholar] [CrossRef] [Scilit]
  14. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders: DSM-5-TR; American Psychiatric Association Publishing: Richmond, VA, USA, 2022. [Google Scholar] [CrossRef] [Scilit]
  15. Dell’Osso, B.; Altamura, A.C.; Allen, A.; Marazziti, D.; Hollander, E. Epidemiologic and Clinical Updates on Impulse Control Disorders: A Critical Review. Eur. Arch. Psychiatry Clin. Neurosci. 2006, 256, 464–475. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  16. Potenza, M.N. Non-Substance Addictive Behaviors in the Context of DSM-5. Addict. Behav. 2014, 39, 1–2. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  17. Potenza, M.N. Should Addictive Disorders Include Non-substance-related Conditions? Addiction 2006, 101 (Suppl. S1), 142–151. [Google Scholar] [CrossRef] [Scilit]
  18. Montag, C.; Reuter, M. Internet Addiction; Springer: New York, NY, USA, 2017. [Google Scholar]
  19. Yau, Y.H.C.; Crowley, M.J.; Mayes, L.C.; Potenza, M.N. Are Internet Use and Video-Game-Playing Addictive Behaviors? Biological, Clinical and Public Health Implications for Youths and Adults. Minerva Psichiatr. 2012, 53, 153–170. [Google Scholar] [PubMed]
  20. Widyanto, L.; Griffiths, M. ‘Internet Addiction’: A Critical Review. Int. J. Ment. Health Addict. 2006, 4, 31–51. [Google Scholar] [CrossRef] [Scilit]
  21. Cohen, J. Statistical Power Analysis for the Behavioral Sciences, 2nd ed.; Routledge: Oxfordshire, UK, 2013. [Google Scholar] [CrossRef] [Scilit]
  22. Foerster, F.R.; Chidharom, M.; Bonnefond, A.; Giersch, A. Neurocognitive Analyses Reveal That Video Game Players Exhibit Enhanced Implicit Temporal Processing. Commun. Biol. 2022, 5, 1082. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  23. Sauce, B.; Liebherr, M.; Judd, N.; Klingberg, T. The Impact of Digital Media on Children’s Intelligence While Controlling for Genetic Differences in Cognition and Socioeconomic Background. Sci. Rep. 2022, 12, 7720. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  24. Huang, S.; Lai, X.; Li, Y.; Cui, Y.; Wang, Y. Beyond Screen Time: The Different Longitudinal Relations between Adolescents’ Smartphone Use Content and Their Mental Health. Children 2023, 10, 770. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  25. Salgo v Leland Stanford Jr. University Board of Trustees. 1957. Available online: https://caselaw.findlaw.com/court/ca-court-of-appeal/1759823.html (accessed on 1 December 2024).
  26. Schloendorff v Society of New York Hospital. 1914. Available online: https://www.quimbee.com/cases/schloendorff-v-society-of-new-york-hospital (accessed on 1 December 2024).
  27. Parmet, W.E. Informed Consent and Public Health: Are They Compatible When It Comes to Vaccines. J. Health Care Pol. 2005, 8, 71. [Google Scholar]
  28. Berg, J. All for One and One for All: Informed Consent and Public Health. Houst. Law Rev. 2012, 50, 1. [Google Scholar]
  29. O’Neill, O. Some Limits of Informed Consent. J. Med. Ethics 2003, 29, 4. [Google Scholar] [CrossRef] [Scilit]
  30. Grisso, T.; Appelbaum, P.S. Assessing Competence to Consent to Treatment: A Guide for Physicians and Other Health Professionals; Oxford University Press: New York, NY, USA, 1998. [Google Scholar]
  31. Kim, S.Y. Evaluation of Capacity to Consent to Treatment and Research; Oxford University Press: Oxford, UK, 2009. [Google Scholar]
  32. Appelbaum, P.S.; Grisso, T. The MacArthur Treatment Competence Study. I: Mental Illness and Competence to Consent to Treatment. Law Hum. Behav. 1995, 19, 105–126. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  33. Grisso, T.; Appelbaum, P.S.; Mulvey, E.P.; Fletcher, K. The MacArthur Treatment Competence Study. II: Measures of Abilities Related to Competence to Consent to Treatment. Law Hum. Behav. 1995, 19, 127–148. [Google Scholar] [CrossRef] [Scilit]
  34. Grisso, T.; Appelbaum, P.S. The MacArthur Treatment Competence Study. III: Abilities of Patients to Consent to Psychiatric and Medical Treatments. Law Hum. Behav. 1995, 19, 149–174. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  35. Maiese, M. Addiction, Autonomy, and Self-Insight. Philos. Psychiatry Psychol. 2024, 31, 351–363. [Google Scholar] [CrossRef] [Scilit]
  36. Aksoy, M.E. A Qualitative Study on the Reasons for Social Media Addiction. Eur. J. Educ. Res. 2018, 7, 861–865. [Google Scholar] [CrossRef] [Scilit]
  37. Hawkins, J. Affect, Values and Problems Assessing Decision-Making Capacity. Am. J. Bioeth. 2024, 24, 71–82. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  38. Levy, N. Addiction, autonomy and ego-depletion: A response to Bennett Foddy and Julian Savulescu. Bioethics 2006, 20, 16–20. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  39. Norlock, K. Free and Always Will Be? On Social Media Participation as it Undermines Individual Autonomy. Can. J. Pract. Philos. 2021, 5, 52–65. [Google Scholar]
  40. Foster, C. When Patients Are Not Themselves. Am. J. Bioeth. 2024, 24, 119–120. [Google Scholar] [CrossRef] [Scilit] [PubMed]
  41. Buchanan, A.; Brock, D.W. Deciding for Others: The Ethics of Surrogate Decision-Making; Cambridge University Press: Cambridge, UK, 1990. [Google Scholar]
  42. Macnish, K. Government Surveillance and Why Defining Privacy Matters in a Post-Snowden World. J. Appl. Philos. 2018, 35, 417–432. [Google Scholar] [CrossRef] [Scilit]
  43. Beauchamp, T.L.; Childress, J.F. Principles of Biomedical Ethics, 6th ed.; OUP: New York, NY, USA, 2009. [Google Scholar]
  44. Manson, N.; O’Neill, O. Rethinking Informed Consent in Bioethics; Cambridge University Press (CUP): Cambridge, UK, 2007. [Google Scholar]
  45. Mill, J.S. On Liberty; Macmillan Education UK: London, UK, 1859. [Google Scholar]
  46. Morris, J. (Ed.) Rethinking Risk and the Precautionary Principle; Butterworth-Heinemann: Oxford, UK; Boston, MA, USA, 2000. [Google Scholar]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Article Metrics

Citations

Article Access Statistics

Multiple requests from the same IP address are counted as one view.