How much authority should parents have?

Suppose a seven-year-old child, Alex, has a rare metabolic disorder for which few pediatric treatments exist. Alex's parents are faced with the decision of whether to enroll him in a high-risk clinical trial. This example highlights the 'therapeutic orphan' problem: researchers must include children in studies to generate knowledge relevant to pediatric care, but there is also a moral obligation to protect children from being used merely as a means (McCormick, 1974). Because children differ from adults in metabolism and enzyme function, data from adult drug trials are not sufficient for pediatric care. However, children lack both the legal and practical capacity to provide informed consent for research participation (McCormick, 1974).

Parental permission is required in pediatric clinical research because parents act as fiduciaries for their children. Their role is to protect the child, who is not autonomous, and to ensure that the child's welfare is prioritized over the aims of research.

I will develop this argument in four sections. First, I will explain why morally transformative consent depends on volitional stability, which is different from cognitive understanding. This distinction shows that children, including mature minors, are vulnerable to outside influence and authority. Second, I will argue that Institutional Review Boards (IRBs) set a general risk ceiling for groups, but do not have the specific knowledge to judge whether a study's risks are appropriate for a particular child's physical or psychological state. Third, I will defend a modified version of McCormick's view on non-therapeutic research. On this view, parents may authorize participation in minimal-risk situations to fulfill a child's presumed civic duty, but their fiduciary duty requires them to veto any risk above this level. Finally, I will address the objection that some parents may be negligent or motivated by self-interest. I will propose a double-lock system: the IRB sets the maximum risk for the group, and the parent ensures the study is suitable for the individual child. Without this approach, children could be exposed to excessive harm or used as mere means for research, which would undermine both their welfare and the integrity of science.

Section I: Volition and Mature Minors 

In this section, I will argue that children and mature minors lack the volitional stability required for morally transformative consent. I will begin by outlining the 'mature minor' objection, which holds that older adolescents, particularly those aged sixteen or seventeen, possess adult-level cognitive capacity (The Belmont Report, 1979). Proponents of this view argue that if a teenager can clearly understand a research protocol's risks and benefits, then requiring parental permission is an unnecessary and paternalistic barrier to the adolescent's developing autonomy (The Belmont Report, 1979). On this view, the 'therapeutic orphan' problem is addressed by recognizing the adolescent's capacity to weigh the risks of a trial against the need for knowledge about pediatric treatments (McCormick, 1974). Critics suggest that if the subject can process information, their assent should be treated as informed consent, making the parental role unnecessary for those close to the age of majority (The Belmont Report, 1979).

The mature minor objection fails because it conflates cognitive capacity with volitional stability. While a sixteen-year-old may possess the intellectual capacity to understand the mechanics of a clinical trial, morally transformative consent requires more than a mere grasp of facts; it requires the agent to be “substantially noncontrolled” (Bromwich & Millum, 2013). Bromwich and Millum (2013) distinguish between the disclosure of information and the quality of the will behind the consent token. For consent to be morally transformative, meaning it successfully alters the moral obligations between the researcher and the subject, it must issue from a stable set of values that the agent can maintain against external pressures (Bromwich & Millum, 2013). 

Adolescents, despite their cognitive scores, often lack this volitional stability. They are biologically prone to impulsivity and uniquely vulnerable to “relational inducement,” the desire to comply to please authority figures such as doctors, researchers, or even their own parents (The Belmont Report, 1979; Bromwich & Millum, 2013). A child might provide a token of assent not because they have performed a rational risk-benefit calculus, but because they feel a social pressure to be a “good patient” (The Belmont Report, 1979). This relational inducement constitutes a form of illegitimate control that undermines the voluntariness of the decision (Bromwich & Millum, 2013). 

Because the child lacks the stability to resist these pressures, the parent must act as a “temporal guardian (McCormick, 1974). In this fiduciary role, the parent is not exercising a right of ownership over the child rather the parent is protecting the “future adult” from potentially regrettable or compliant choices of the “present child” (McCormick, 1974; Werthemer, 1999).  The parent provides the rational buffer that the child’s developing brain cannot yet maintain, ensuring that the subject is not used merely as a means to a scientific end (Ramsey, 1970). 

To respect the developing personhood of the minor, the research ethics system utilizes "assent," the child’s affirmative agreement to participate (The Belmont Report, 1979). However, because the child lacks volitional stability, assent is an ethical courtesy rather than a legal authorization (The Belmont Report, 1979). The morally vital prerequisite is parental permission, which functions as the asymmetric fiduciary veto. The interaction between two mechanisms depends on whether the research is therapeutic or non-therapeutic. In a therapeutic context, where the research offers a direct, life-saving medical benefit unavailable elsewhere, the parental fiduciary duty to preserve the child’s life takes precedence over the child’s dissent (McCormick, 1974). In such cases, the principle of beneficence is paramount, as it prioritizes preserving life, a fundamental value that any rational person should uphold (Faden, 1986). This hierarchy of principles ensures that the child’s right to life takes precedence over limited autonomy in high-stakes scenarios. As McCormick (1974) argues, a construction of the child’s presumed wishes is valid when it aligns with what any rational person ought to choose for their own flourishing. Therefore, the parent overrides the child’s “no” to fulfill the primary fiduciary obligation of protection.

In a non-therapeutic context, where the research offers no direct benefit to the subject, the child’s dissent acts as an absolute veto that cannot be overridden by parental permission. While McCormick suggests that children have a presumptive “civic duty” to bear minimal burdens for the common good, I argue that this “civic duty” floor is insufficient to justify forcing a resisting child to participate. Paul Ramsey (1970) correctly identifies that performing non-therapeutic procedures on a non-consenting subject is an “offensive touching: that treats the child as an object. In these cases, the fiduciary duty compels the parent to honor the child’s dissent to prevent the child from being sacrificed for generalizable knowledge (Ramse,y 1970; The Belmont Report, 1979). 

Section II: The Core Conflict in Non-Therapeutic Research

In this section, I will examine the main ethical conflict in non-therapeutic research, which does not offer direct medical benefit to the child. I will explain the positions of Ramsey and McCormick, and then argue for a modified version of McCormick's view. On this view, parental permission sets a civic duty floor for minimal risk and also acts as a fiduciary ceiling that blocks any risk above that level.

A fundamental disagreement over the nature of personhood and social obligation defines the ethical inclusion of children in non-therapeutic research. Ramsey (1970) argues for a strict protectionist stance rooted in the Kantian principle that persons must never be treated merely as a means to an end (Ramsey, 1970). For Ramsey, any non-therapeutic intervention on a child who cannot consent constitutes an “offensive touching.” Because a child is not a moral agent, they cannot volunteer for charitable work or become a “joint adventurer” in the medical process (Ramsey, 1970; Mcormic, 1974). Ramsey contends that a proxy’s duty is purely fiduciary, to protect the child’s welfare and survival, and thus, authorizing a child’s participation in research that offers no benefit to that child is a breach of this duty (Ross, 1998; Ramsey, 1970). On this view, the “Therapeutic Orphan” dilemma cannot be solved by treating children as objects for the benefit of future generations (Ramsey, 1970). 

McCormick (1974) offers a competing “civic duty” model grounded in the natural law tradition. He argues that humans are inherently social beings whose flourishing is tied to the well-being of the community (McCormick, 1974). McCormick rejects the view of the person as an “atomistic” individual, suggesting instead that there are some fundamental values any rational person ought to support. Because sharing in the “common burden” of medical progress is a good that resides in others, McCormick (1974) concludes that a child would want to participate in such efforts if they were capable of choice, because they ought to do so. Consequently, parents may authorize a child’s participation in research to “socialize” them into these civic obligations, provided the risks are not significant (McCormic, 1974; Ross, 1998). 

I will now evaluate these positions. Ramsey's view is overly restrictive. By treating every non-therapeutic procedure as 'offensive touching,' he would prohibit even minimal data collection, such as routine weighing or buccal smears, which are necessary to address gaps in pediatric knowledge. If we adopt Ramsey's absolutism, we risk leaving children in a permanent state of 'therapeutic orphanhood' (McCormick, 1974). On the other hand, McCormick's view may be too permissive. By basing permission on what a child ought to want, he risks allowing the interests of the group to override the interests of the individual child (McCormick, 1974). Without clear limits, this approach could be used to justify exposing children to hazardous studies for the sake of the 'greater good' (Ramsey, 1976).

To address this conflict, I propose a 'floor-ceiling' model. In this model, McCormick's civic duty view establishes the minimum standard for participation (the floor), while Ramsey's protectionist view sets the maximum allowable risk (the ceiling).

I will now elaborate on the civic duty floor. Parents may authorize participation in non-therapeutic research only when the intervention meets the threshold of “minimal risk.” Following the scholastic principle parum pro nihilo reputatur, “very little counts for nothing,” minimal risk refers to harms no greater than those encountered in daily life or routine examinations (McCormick, 1974; The Belmont Report, 1979). In these scenarios, the child’s civic duty to bear negligible burdens for the community is a reasonable presumption (McCormick, 1974). At this floor, the parent is not violating the child’s personhood but, in turn, guiding their development toward a socially responsible adult (Ross, 1998).

However, the parental fiduciary duty must function as an absolute ceiling. Parents must veto any research participation where the risk exceeds the "minimal" threshold (McCormick, 1974). Some might argue that allowing participation could yield greater aggregate benefits, advancing societal progress. However, while utilitarian considerations might lead adults to choose to accept greater risks for the collective good, parents cannot impose self-sacrifice on behalf of their child when there is no direct benefit to the child (McCormick, 1974). In non-therapeutic contexts, the fiduciary duty to protect the child from dangerous experiments trumps any perceived civic obligations. If a study carries a minor increase over minimal risk without the prospect of direct benefit, it fails the fiduciary test because it treats the child’s body as a tool for generalizable knowledge (The Belmont Report, 1979; Ramsey, 1970).

Parental permission is justified because it protects the child. It is valid when it serves as an asymmetric veto to prevent harm. Parents should not be allowed to permit participation in research with significant risks. This approach prevents children from being used as means, while still allowing minimal-risk research that is necessary for pediatric care. By maintaining this asymmetry, the parent acts as a rational buffer and ensures that scientific aims do not override the child's status as a person.

Section III: Risk, Inducement, and Epistemic Access

In this section, I argue that Institutional Review Boards (IRBs) set only a general risk ceiling, which leaves an epistemic gap that only parents can fill. First, I will explain the institutional context of risk assessment and the recruitment of vulnerable populations. Then, I will show that risk is subjective and that parents must act as buffers to protect the child's welfare over scientific aims.

The recruitment of human subjects often occurs within what Elliott and Abadie (2008) describe as a “shadow economy” where participation in clinical trials becomes a functional substitute for traditional employment for people with low incomes. In this environment, inducements, such as monetary compensation or free medical care, can exert coercive pressure on vulnerable populations (Elliott & Abadie, 2008; The Belmont Report, 1979). For the “research underclass,” a sum of money that a wealthy individual could easily resist may be a “seductive offer” that causes them to discount long-term costs or risks (Wertheimer, 1999; Elliot & Abadie, 2008). 

Institutional safeguards, primarily IRBs, are designed to protect these populations by ensuring that research designs are reasonable (The Belmont Report, 1979). The IRB evaluates risk through an objective statistical lens, focusing on the magnitude of the physical or psychological harm (The Belmont Report, 1979; Ross, 1998). Under the minimal risk standard, a procedure is deemed permissible if the risks do not exceed those encountered in the “daily life” of healthy children, or during “routine medical examinations (The Belmont Report, 1979). For example, a blood draw is classified as minimal risk because it has a statistically negligible chance of causing significant physical injury (McCormick, 1974; Ross, 1998). From the IRB perspective, if a study falls below this objective ceiling, it is deemed ethically acceptable for the target population (Ross, 1998). 

I argue that the IRB’s statistical framework is insufficient because it treats risk as a universal context, failing to account for its subjective nature. While a procedure like a needle stick may be “minimal” risk according to population data, it can constitute a “maximal” psychological trauma for a specific child. As Ross (1998) observes, a child with a history of medical trauma or a condition such as autism may experience intense psychological distress from a routine examination that a “usual” child would find innocuous. Because IRBs evaluate protocols based on population averages, they lack the epistemic access required to identify these individual vulnerabilities (Ross, 1998; The Belmont Report, 1979).

This creates an epistemic gap between institutional policy and individual welfare. The IRB sets the risk ceiling for the study population, but cannot determine whether that ceiling is appropriate for a particular child. Only the parent, acting as a fiduciary, has the specific knowledge of the child's history, temperament, and psychological state. For this reason, parental permission is a necessary moral requirement, since the parent is best positioned to judge whether participation is consistent with the child's welfare.

Furthermore, research participation often involves power imbalances where the researcher’s scientific objectives may overshadow the subject’s interests (The Belmont Report, 1979). In the “shadow economy” of Phase 1 trials, where subjects often view participation as a job, there is a risk that financial inducements will lead to the exploitation of those who participate precisely because they need money (Elliott & Abadie, 2008). For a child or adolescent with limited volitional stability, a small incentive, such as a $50 gift card, can constitute an “undue inducement” that overwhelms their ability to appreciate the real burden of the research (Wertheimer, 1999; Bromwich & Millum, 2013).

The parental fiduciary duty serves as a necessary counterweight (the veto) to these scientific and economic pressures. In this case, parents also act as a rational buffer, looking past the immediate lure of inducements to evaluate the long-term cost to the child (Wertheimer, 1999). In this capacity, parental permission is not an act of volunteering a child for harm, but a prospective mechanism that blocks any participation where the subjective or physical risk exceeds the child’s capacity to flourish (McCormick, 1974). By exercising this asymmetric veto, the parent ensures that the child is treated as an “end in himself” rather than a means to generalizable knowledge (Ramsey, 1970; Ross, 1998). While the IRB ensures the study is “good enough” for a population, the parent ensures the study is “good enough” for their child, filling in the epistemic gap left by institutional oversight (Ross, 1998; The Belmont Report, 1979).

Section IV: The Inverted Objection

I will now address the “Inverted Objection,” which challenges the moral necessity of parental permission by highlighting the possibility of parental failure. I will begin by explaining the objection that negligent, abusive, or greedy parents render the permission mechanism an unreliable safeguard. I will then evaluate this objection by defending what I call a double-lock structural framework, arguing that the IRB and the parental fiduciary serve as distinct, non-redundant safeguards that ensure the child is protected even when one mechanism fails.

The moral justification for the asymmetric fiduciary veto rests on the premise that parents are the primary protectors of their child’s individual welfare. However, a significant objection arises: what happens when parents are greedy, abusive, negligent, or simply unfit? If the moral necessity of parental permission is derived from its protective function, some argue that this necessity vanishes when a parent is willing to trade a child’s safety for a “seductive offer” (Wertheimer, 1999). 

In the shadow economy of clinical research, where participation often becomes a survival strategy for the disadvantaged, inducements, such as financial incentives for medical professionals, can amount to coercive measures. An objector might posit that a parent motivated by financial desperation could authorize a child’s participation in a high-risk research that a rational fiduciary would veto. If a parent fails to act as a rational buffer and instead becomes an agent of inducement, the objection suggests that parental permission is not a morally vital prerequisite but rather a point of failure. On this view, the IRB serves as the only real protection for the child, and the requirement for parental permission should be seen as a fallible legal formality that can be bypassed if the state determines the research is in the child’s best interest.

This objection relies on a false choice between institutional oversight and parental permission. Parental permission is necessary, but not sufficient, for ethical research. Protecting the non-autonomous child requires a double-lock system, in which two independent agents must both authorize participation before a subject is enrolled.

The first safeguard is “Lock 1,” the IRB. As Emanuel, Wendler, and Grady (2000) establish, independent review is a fundamental principle of ethical research that is conceptually and operationally distinct from the permission process. The primary duty of the IRB is to set a risk ceiling for the population (The Belmont Report, 1979). The IRB ensures that the research design is scientifically valid and that the risks are “reasonable in relation to anticipated benefits” for the target population as a whole (The Belmont Report, 1979). 

Lock 1 prevents a negligent parent from enrolling a child in bad science. No parent, regardless of their level of greed or negligence, is ever presented with the choice to enroll a child in a study that carries an unethical risk-to-knowledge ratio (The Bemont Report, 1979). By the time a parent is asked for permission, the IRB has already determined that the study meets a minimum baseline of social value and safety (Emanuel et al., 2000). Thus, Lock 1 removes the most egregious forms of exploitation from the table before the parent’s fiduciary duty is even engaged. 

While Lock 1 protects children as a class, “Lock 2” parental permission protects the individual child from a “bad fit.” This second lock is morally vital because IRBs lack the epistemic access required to assess the subjective and unique vulnerabilities of a specifc child (Ross, 1998) The IRB evaluates risk statistically, but a procedure classified as minimal risk by a board can constitute a maximal psychological trauma for a partucular child, such as one with history of medical trauma or autism (Ross, 1998). Only the parent, acting in their fiduciary capacity, can judge whether a specific child has the temperament and physical state to withstand the study (Ross, 1998). The fiduciary does not have the power to lower the risk cceilingset by the IRB, but they retain the absolute right to veto participation if the bad science lock is insufficient to cover the child’s bad fit.

The core inverted objection is the fear that if a negligent parent compromises the two, the child is left defenseless. However, the double lock system would provide a fail-safe. If a parent is negligent and authorizes a “bad fit,” the subjects are still protected by Lock One, which has guaranteed that the study is not “hazardous” in the extreme (The Belmont Report, 1979). Furthermore, the investigator has the professional duty to know and assess the subject (Bromwich & Millum, 2013). Suppose an investigator has a professional duty to know and assess the subject (Bromwich & Millum, 2013). If an investigator realizes a parent is acting in a way that is “disrespectful of the child’s developing personhood” or is ignoring the child's obvious distress for the sake of money, the investigator’s own fiduciary duty to the subject compels them to refuse the enrollment (Ross, 1998; Bromwich & Millum, 2013).

The failure of a minority of parents to act as fiduciaries is an argument for greater IRB oversight and training in investigatory procedures, not for removing parental veto. For the vast majority of children, the parent remains the most motivated agent to prioritize the child’s status as an “end in himself (Ramsey, 1970). The moral necessity of parental permission remains a structural requirement because it provides an individualized protection that an institutional body is epistemically and legally incapable of providing (Ross, 1998). Parental permission is the essential second lock that ensures research participation is a rational choice for the child, even when background conditions are less than ideal.

Section V: Closing Remarks

Parental permission functions as an asymmetric fiduciary veto. This requirement is essential because it recognizes that the child is not autonomous and lacks the volitional stability needed to manage the power imbalances present in clinical research (Bromwich & Millum, 2013). Institutional Review Boards are necessary to set an objective risk ceiling and to protect the research protocol, but only the parent has the specific knowledge required to protect the individual child (The Belmont Report, 1979; Ross, 1998). The double-lock system demonstrates that both institutional and parental safeguards are necessary. The IRB prevents exploitation at the group level by ensuring scientific validity and social value, but cannot assess the subjective or psychological impact on a particular child (Emanuel et al., 2000; Ross, 1998). In non-therapeutic research, the parental veto prevents the child's presumed civic duty from being used to justify treating the child as a means to an end (McCormick, 1974; Ramsey, 1970). By maintaining the risk threshold at minimal risk, the parent acts as a rational buffer against the pressures of the research system, which often targets vulnerable individuals. Without this check, the child could be exposed to the utilitarian aims of medicine and research. The asymmetric fiduciary veto ensures that the child's welfare remains the primary concern. For this reason, parental permission is not merely a legal formality, but a necessary moral requirement.

 Sources:

Bromwich, D., & Millum, J. (2013). Disclosure and Consent to Medical Research Participation. Journal of Moral Philosophy.

Elliott, C., & Abadie, R. (2008). Exploiting a Research Underclass in Phase 1 Clinical Trials. The New England Journal of Medicine, 358(22), 2316–2317.

Emanuel, E. J., Wendler, D., & Grady, C. (2000). What Makes Clinical Research Ethical? JAMA, 283(20), 2701–2711.

Emanuel, E. J., Wendler, D., Killen, J., & Grady, C. (2004). What Makes Clinical Research in Developing Countries Ethical? The Benchmarks of Ethical Research. The Journal of Infectious Diseases, 189(5), 930–937.

McCormick, R. A. (1974). Proxy Consent in the Experimentation Situation. Perspectives in Biology and Medicine, 18(1), 2–20.

National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research. (1979). The Belmont Report: Ethical Principles and Guidelines for the Protection of Human Subjects of Research.

Pryor, J. (2012). Guidelines on Writing a Philosophy Paper.

Ramsey, P. (1970). The Patient as Person: Explorations in Medical Ethics. Yale University Press.

Ramsey, P. (1976). The Enforcement of Morals: Nontherapeutic Research on Children. The Hastings Center Report, 6(4), 21–30.

Ross, L. F. (1998). Children, Families, and Health Care Decision Making. Oxford University Press.

Ross, L. F. (2019). Better than Best (Interest Standard) in Pediatric Decision Making. The Journal of Clinical Ethics, 30(3), 183–195.

U.S. Department of Health and Human Services. (2018). Code of Federal Regulations, Title 45: Public Welfare, Part 46—Protection of Human Subjects (The Common Rule).

Wendler, D. (1996). Deception in Medical and Behavioral Research: Is It Ever Acceptable? The Milbank Quarterly, 74(1), 87–114.

Wertheimer, A. (1999). Exploitation. Princeton University Press.

Wertheimer, A. (2011). Rethinking the Ethics of Clinical Research. Oxford University Press