(EN)
(ES)
(IT)
Contemporary science is often presented in public debate, as well as in the offices of legislators and courts, as a monolithic institution providing absolute certainties, a methodological distortion that the philosophy of science has tried to dismantle over the last century [1, 2] . In their seminal work How Science Works , Michael Weisberg and Anastasia Thanukos present a realistic and detailed view of actual scientific practice, contrasting it with the sterile and linear stereotype of the “scientific method” found in school textbooks [1, 3] .
Science does not operate through a rigid five-step recipe, but is defined in a dual way: it is both an accumulating body of knowledge and the dynamic, complex, iterative, and social process designed to build and refine that knowledge based on empirical evidence obtained through observation, experimentation, and simulation [1] . This knowledge-building process is accurately described by the “flow diagram of science,” a model of multiple, non-linearly interacting paths where hypotheses are iteratively tested and errors are progressively corrected through the interaction of a diverse community [1] .
To understand the usefulness of science in bioethical deliberation, it is imperative to establish its principles of scientific demarcation, which are summarized below.
Investigation of the natural world
Experimental science is strictly limited to investigating the physical and natural world (which encompasses everything from matter and physical forces to human societies and their products) using exclusively natural explanations [1, 2] . Hypotheses or explanations that appeal to supernatural or metaphysical causes are methodologically outside its scope because they are not susceptible to empirical testing [1] .
Generation of testable hypotheses and predictions
For a proposal to be scientific, it must generate logical predictions, that is, expected observations if the hypothesis is correct or incorrect. Ideas that shield themselves from empirical testing or that are compatible with any possible observation are not part of science [1] .
Dependence on and response to evidence
Empirical evidence is the supreme arbiter of scientific validity. The scientific community has an obligation to assimilate the available evidence, modifying or abandoning hypotheses when the data contradict them [1] .
Fallibilism
One of Weisberg and Thanukos’s central theses is the principle of fallibilism: science never irrevocably “proves” a hypothesis [1, 4] . However robust the empirical support for a theory may be, it always remains open to refinement, revision, or rejection if new evidence or more coherent interpretations emerge, as demonstrated by the historical replacement of classical mechanics by the theory of relativity or the late acceptance of the prion hypothesis [1, 5] .
The influence of context
Furthermore, science is an inherently human and communal endeavor. Researchers are not perfectly objective automatons, but human beings conditioned by their own contexts, biases, career ambitions, and funding incentives [1] . Scientific objectivity does not reside in the individual mind of an isolated researcher, but in the cross-scrutiny and collective scrutiny of a diverse scientific community that subjects work to rigorous controls [1] .
Control mechanisms
Among the control mechanisms proposed to carry out this scrutiny, peer review stands out as an essential filter to guarantee a minimum standard of quality and methodological validity in the official register of science [1] .
However, Weisberg and Thanukos caution that peer review is not infallible, as it is not designed to detect sophisticated scientific fraud or systematic data manipulation, as demonstrated by the late retraction of key Alzheimer’s research originally published in the journal Nature [1, 6, 7] .
On the other hand, the prestige of the journal (impact factor), the eminence of the author or isolated statistical significance (p-values less than 0.05) are misleading heuristics that do not automatically equate to methodological correctness or scientific consensus [1] .
The rise of preprints (pre-publications without peer review) also introduces agility, but lacks standardized community filters, forcing evaluators to apply an extraordinarily cautious methodological examination before validating their conclusions [1] .
Finally, scientific consensus is not an instantaneous event resulting from a single critical experiment, but an emergent property that arises slowly over the years as multiple independent and separate lines of evidence converge towards the same theoretical explanation [1, 8] .
The definitive delimitation of this epistemological framework is condensed at the boundary of moral judgments: experimental science is a descriptive and predictive instrument capable of revealing how the real world works, but it lacks the methodological and conceptual competence to make prescriptive or moral judgments. It can measure, for example, the psychological effects, social acceptance, or physiology of a practice, but it is epistemologically incompetent to determine whether that practice is ethically permissible or illegitimate [1, 5] .
2. Significance in the assessment of the biological status of the human embryo
The bioethical debate on the moral status of the human embryo directly benefits from the application of the demarcation principle between fact and value proposed by Weisberg and Thanukos. Clinical embryology and molecular genetics provide highly precise and testable descriptions of the physical reality of the early embryo: the conformation of its unique genome and the developmental program it contains and initiates from fertilization, the sequential patterns of its cell division, and the totipotency of the blastomeres are all empirically verifiable facts of the natural world.
However, the step from biological description to the attribution of “human nature” in a normative sense (i.e., the consideration of whether the early embryo is a person with intrinsic dignity and an inalienable right to life) represents a moral and metaphysical judgment that exceeds the competence of the experimental sciences [1, 5] .
Therefore, as we have previously published, the basis for the absolute ethical protection of an embryo—or, on the contrary, its total lack of protection—must be based, in addition to its biological nature, on the consideration of its ontological quality, which will shape the anthropological-philosophical interpretation in which the dignity or status of person is inserted, on which the treatment and recognition that must be given to it and the rights that will be recognized will depend.
Furthermore, the application of the principle of fallibilism is of paramount importance in evaluating the biological milestones traditionally used by bioethicists to draw lines of moral demarcation. Biological concepts that once seemed absolute, such as cellular totipotency, the impossibility of organized development without prior fertilization, or the fourteen-day limit associated with the primitive streak, are undergoing profound scientific revision [1] . Advances in developmental biology and the creation of synthetic embryo models from the reprogramming of stem cells demonstrate that scientific categories are not static, but rather provisional explanations subject to constant self-correction. Bioethics, therefore, cannot be anchored in immutable biological dogmas from textbooks, but must instead develop dynamic conceptual frameworks capable of engaging with a science in constant empirical refinement.
3. Significance in the ethical debate on euthanasia
Weisberg and Thanukos explicitly address the case of euthanasia to illustrate the insurmountable limits of scientific demarcation: experimental science can provide empirical data of the highest significance for this debate—such as rigorous surveys on the public acceptance of assisted suicide in different cultures [9] , qualitative studies of the psychology of the terminally ill patient in the face of death [10], or detailed quantitative descriptions of the psychosocial and emotional impact on surviving relatives [11] —but it is radically incapable of decreeing whether euthanasia is a morally right or wrong act [1] .
The significance of this epistemological limit in the field of bioethics is twofold. On the one hand, it prevents the technocratization of morality: a bioethics committee or a legislature that seeks to justify the legalization or prohibition of euthanasia based exclusively on public opinion polls or clinical effectiveness statistics would be falsifying the scientific process, using it as a shield of apparent technical objectivity to conceal underlying ethical value choices.
Secondly, a rigorous philosophical and anthropological foundation is required: The evaluation of euthanasia must necessarily be resolved within the framework of moral philosophy and law, weighing rational principles such as the patient’s personal autonomy, the sanctity and intrinsic value of human life, and the duties of beneficence, non-maleficence, and therapeutic compassion within the ethics of care. Data provided by clinical medicine and behavioral sciences serve as an essential empirical context for understanding the factual reality of the terminally ill patient, but never as a direct source of normative justification.
4. Transcendence in Gender Theories (felt sex versus biological sex)
The debate surrounding the discrepancy between perceived gender identity or “felt sex” and biological chromosomal and anatomical sex represents one of the most complex challenges in current clinical bioethics. Weisberg and Thanukos explicitly argue that the social and behavioral sciences are legitimate disciplines capable of scientifically investigating complex human dynamics through rigorous quantitative, qualitative, or mixed-methods methodologies [1, 12] . Consequently, felt sex and the processes of psychological self-perception and social integration of gender are entirely scientific objects of study.
However, contemporary epistemology demands that these behavioral sciences be subjected to the same level of critical scrutiny, bias control, and methodological rigor as is usually required of the natural sciences [1] . When analyzing the postulates of gender theories that advocate for medical transition interventions, both in adults and at an early age in minors (transition hormone therapies, puberty blockers, or reassignment surgeries), bioethics must subject the alleged medical “consensus” to a rigorous epistemological evaluation in light of the current pathologies of science described in the text [1] , such as:
a) The replication crisis and methodological fragility
Behavioral sciences and psychology are at the epicenter of the so-called “replication crisis,” where an alarming proportion of published studies fail to replicate their results when subjected to independent testing [1, 13] . Many clinical guidelines recommending early medical transitions rely on observational studies with small sample sizes, high dropout rates during follow-up, and a lack of adequate control groups, severely limiting their statistical power and external validity.
b) Publication and funding biases
Science, as a human social practice, is strongly affected by systemic biases that favor the publication of positive and innovative results over those that do not report significant effects or that suggest long-term harm [1, 14]. In the field of gender health, institutional and ideological pressures, as well as funding sources (with strong market interests in the chronicity of hormonal treatments), can introduce profound confirmation biases, actively discouraging the publication of critical studies or research on withdrawal and detransition processes [1, 15, 16].
c) Manipulation of public perception of consensus
Weisberg and Thanukos warn that certain sectors and interest groups design communication strategies to simulate the existence of a consolidated scientific consensus where, in reality, intense internal scientific debate persists [1] . In gender bioethics, the label of “expert consensus” is often politically instrumentalized, ignoring systematic and independent reviews from countries with a strong bioethical tradition (such as the United Kingdom, Sweden, or Finland), as we have previously published, which have restricted hormonal interventions in minors due to the lack of robust convergence of evidence and significant rates of clinical error.
Guided by the principle of scientific prudence, clinical bioethics must warn that implementing experimental and irreversible medical protocols on the pediatric population based on “bad science” or on theories without a solidly grounded consensus in multiple lines of converging clinical evidence of long-term follow-up constitutes a transgression of the principles of non-maleficence and therapeutic beneficence.
5. Conclusions: Epistemological vigilance in the face of biotechnological disruption
Applying the philosophy of science proposed by Weisberg and Thanukos to contemporary bioethics allows us to articulate a defense of clinical rationality based on what we can call active epistemological vigilance. Given the exponential acceleration of advances in biomedicine and the social and market pressure for their immediate application, scientists, legislators, and bioethics committees must operate as critical interpreters guided by the following fundamental methodological principles:
Weighing uncertainty and the precautionary principle
Given that scientific knowledge is inherently provisional and fallible [1, 4] , and that rigorous replication of medical innovations takes years, bioethics committees must make regulatory decisions under conditions of high uncertainty. In these cases, prudence dictates that, in the absence of a robust long-term convergence of empirical evidence, the precautionary principle be applied, halting or restricting the widespread clinical use of invasive or irreversible technologies until their methodological error margin and therapeutic safety are firmly defined by the scientific community [1] . This is especially necessary in the field of gene editing and diagnostics, where uncertainty about their effects must weigh the advisability of undertaking certain experiments.
Demand for financial and methodological independence
In addition to evaluating the validity of the results presented by the promoters of a biotechnology, the sources of sponsorship and funding must be actively investigated, recognizing that there is a statistically proven bias that correlates the corporate origin of the money with methodological results that are consistently favorable to the sponsors [1, 16] . True independence of ethical judgment requires purging clinical evidence of underlying conflicts of interest.
Critical filters for preprints and provisional literature
Given the rapid pace of current biomedical research, bioethical judgments that condition scientific procedures or clinical or regulatory recommendations should not be based on preliminary literature ( preprints ) or studies hastily published in journals with low methodological standards [1, 17] . They must require the maturity of the scientific community process and empirical confirmation through independent systematic reviews and methodologically sound meta-analyses.
Julio Tudela. Bioethics Observatory. Institute of Life Sciences. Catholic University of Valencia
***
Bibliographic References
1. Weisberg M, Thanukos A. How Science Works. In: Reference Manual on Scientific Evidence. Washington (DC): Federal Judicial Center / National Academies Press; 2025.
2. Pigliucci M, Boudry M, editors. Philosophy of Pseudoscience: Reconsidering the Demarcation Problem. Chicago: University of Chicago Press; 2013.
3. McComas WF. The Principal Elements of the Nature of Science: Dispelling the Myths. In: McComas WF, editor. The Nature of Science in Science Education: Rationales and Strategies. Dordrecht: Kluwer Academic Publishers; 1998. p. 53-70.
4. Haack S. Inquiry and Advocacy, Fallibilism and Finality: Culture and Inference in Science and the Law. Law, Probability and Risk. 2003;2(3):205-14.
5. Godfrey-Smith P. Theory and Reality: An Introduction to the Philosophy of Science. Chicago: University of Chicago Press; 2003.
6. Lesné S, Koh MT, Kotilinek L, Kayed R, Glabe CG, Yang A, et al. A Specific Amyloid-β Protein Assembly in the Brain Impairs Memory. Nature. 2006;440(7082):352-7. Retraction in: Nature. 2024 Jun 24;630:1024.
7. Piller C. Blots on a Field? A neuroscience image sleuth finds signs of fabrication in scores of Alzheimer’s papers, threatening a dominant theory of the disease. Science. 2022;377(6604):358-63.
8. Nye MJ. Molecular Reality: A Perspective on the Scientific Work of Jean Perrin. London: Macdonald; 1972.
9. Cohen J, Marcoux I, Bilsen J, Deboosere P, van der Wal G, Deliens L. European Public Acceptance of Euthanasia: Socio-Demographic and Cultural Factors Associated with the Acceptance of Euthanasia in 33 European Countries. Social Science & Medicine. 2006;63(3):743-56.
10. Leichtentritt RD, Rettig KD. Meanings and Attitudes Toward End-of-life Preferences in Israel. Death Studies. 1999;23(4):323-58.
11. Swarte NB, van der Lee ML, Conyn-van Spaendonck MA, de Treslong J, Heintz AP. Effects of Euthanasia on the Bereaved Family and Friends: A Cross Sectional Study. BMJ. 2003;327(7408):189.
12. Hedges LV. How Hard Is Hard Science, How Soft Is Soft Science? The Empirical Cumulativeness of Research. American Psychologist. 1987;42(5):443-55.
13. National Academies of Sciences, Engineering, and Medicine. Reproducibility and Replicability in Science. Washington (DC): National Academies Press; 2019.
14. Smaldino PE, McElreath R. The Natural Selection of Bad Science. Royal Society Open Science. 2016;3(9):160384.
15. Krimsky S. Do Financial Conflicts of Interest Bias Research? An Inquiry into the «Funding Effect» Hypothesis. Science, Technology, & Human Values. 2012;38(4):566-87.
16. Sismondo S. Pharmaceutical Company Funding and Its Consequences: A Qualitative Systematic Review. Contemporary Clinical Trials. 2008;29(2):109-13.
17. Rieger OY. Preprints in the Spotlight: Establishing Best Practices, Building Trust. Ithaca (NY): Ithaka S+R; 2020.
