Research

Cosim Sayid works in law and philosophy. A unifying theme of his thought is that analytic rigor -- especially from the philosophy of language -- can better elucidate legal, moral, political, and social concepts and how they relate to our history and traditions. I make extensive use of philosophy and philosophical methods – especially at the intersection of law and other disciplines, like cognitive science, economics, epistemology, or philosophy of language and linguistics. I use this to elucidate and resolve issues that arise in private law (especially torts and contracts) as well as the law of evidence. To my mind, this is a two-way street between law and philosophy, which functions as a proving ground for current philosophical accounts, and generates demands about what sorts of theories the law requires for future progress. This analytic discipline also addresses an agenda in legal theory, jurisprudence, and philosophy of law more generally, especially as respects topics in law and language – like the meaning of legal terms and texts. I am also interested in more foundational philosophical work, often with an eye towards potential moral, political, or social applications.

Several projects are outlined below. Rather than just have a giant philosophy of law/political and social theory category, projects are grouped under the conceptually nearest more fine-grained research area from amongst Torts, Evidence, Contracts, and Jurisprudence/Philosophy of Law. These are only brief statements, and Cosim very glad to discuss any of them at greater length; please write him. 

***** PAGE UNDER CONSTRUCTION *****

Private Law

Across private law subjects, I grapple with the idea that private law is simultaneously shaped by instrumentalist thinking but not dealing in thick concepts that are irreducible to an instrumentalist paraphrase.

Torts

I’m interested in specific areas of tortious liability (like defamation and false light depiction) as well as the justification of a particular mechanism via which tortious liability operates in a specific area, e.g., wrongfulness-tracking strict liability, market-share liability. I am also trying to advance a contractualist justification of strict liability torts as part of a potentially larger effort to study contractualist justification in law, which may offer more general lessons concerning contractualism.

Defamatory Harm [completed; under review]

Defamatory harm can cause grievous, oft-irreparable, injury. That injury runs via intrinsically reducing one’s agential well-being, by perpetrating an epistemic injustice, which is then fleshed out using game theory. Defamatory harms usher in a game-theoretic cheap talk environment, where it is rational to wholly discount otherwise credible testimony about someone’s reputation. I show that there is a Kantian imperative to recover the good of generic credibility by imposing a mechanism to stop cheap talk. I also show that it is important to address defamatory harm that arises indirectly (as in false light portrayal), even though it is relatively neglected compared to defamation. The threat of cheap talk cannot be addressed unless both what’s strictly false about someone is handled by a mechanism as well as what’s implied about someone’s reputation; letting false uptake alone while dealing exclusively with false assertion is not viable. 

I then consider a variety of mechanisms for stopping cheap talk, which are bifurcated into two trios: non-tort mechanisms and tort law mechanisms. Non-tort mechanisms include an informal self-help regime, criminal law, and property law. Based either on their inefficacy in a globalized world or on their remedial heavy-handedness in the latter two cases, I rule out these options. The three tort alternatives then surveyed are the actual malice regime expounded by the United States Supreme Court, a negligence regime, and a strict liability scheme propounded by Arthur Ripstein for defamation. Against the first two of these, I contend that it denies victims of defamatory harms their dignity in the sense that they are unable to make a demand against those harmed them; they’re made to totally bear losses that they did not cause. The actual malice regime also legally permits defamatory misfeasance; negligence is explicitly not legally actionable. Negligence fares somewhat better, since it covers both misfeasance and malfeasance, but it fails to hold legally accountable those who engage in defamatory feasance. Permitting those whose conduct is even a smidge better than negligent are not made to pay for the harm they cause. One should have a right against his reputation being falsely diminished, and if we enforce this right through private law, in tort, it is up to the victim to decide whether to litigate. The negligence regime makes it as if the victim has waived her right or consented to bad treatment; this aspect of the negligence regime is shared with actual malice.

As against actual malice and negligence, I urge that we consider contractual alternatives, whereby an attention-starved agent desiring more coverage can grant a license to be covered under either a negligence regime or per actual malice. In either case, this is an autonomous choosing – a choice to forego holding someone responsible in all cases. But it is wrong to make this choice a fait accompli. In the case of negligence, insurance is also an option that has not been explored.

As against Ripstein’s form of strict liability for negligence, I question how it fails to address false light depiction, how it turns on having a good name, and how it ushes in a strict but inflexible regime.

The notion of a wrongfulness-tracking strict liability rule is introduced. Liability is without regard to fault, but damages assessed are via a multiplier that tracks wrongfulness. For egregious conduct, punitive damages are assessed, amounting to a multiplier greater than one. For defamatory harm that happened even after due diligence, a multiplier of less than one may be appropriate. A victim’s dignity is addressed in that she’s able to claim against the party that caused her harm, but free expression is not excessively chilled, since well-nigh unavoidable damage is discounted. 

Market-Share Liability as Corrective Justice

(with Yuval Abrams (Michigan State, Law))

We defend market-share liability in appropriate circumstances as a response within tort law that does not do violence to the traditional causation requirement or adjust the evidentiary burden for a platintiff in an ad hoc way. To do this, we rely on the technique of causal grouping, where defendants are pooled together based on their acknoweldged causal behavior.

We then justify this procedure using classical contractarianism and contractualism. 

Omissive Causation and Mereology

How absences (or omissions) cause in law, morals, or elsewhere is mysterious. How does an absence – what is non-existent and paradoxical to denote – cause events in the world? Criminal law and tort law are rife with this sort of talk – I focus here on the latter – but it seems thoroughly wrongheaded from the naturalistic and scientifically respectable worldview. 

By appealing to mereology – the study of parts of wholes – and reformulating the ontology of causation, I show that we can dissolve these alleged difficulties. There is, thus, nothing metaphysically questionable about liability for omissions. Such liability is not derivative or second-class. There is no need to resort to questionable notions like ‘quasi-causation’ when a common-sense understanding of how our concepts of liability, omissions, and ontology work is feasible.

Targeted by this exercise are accounts where liability via absence-causation is (ceteris paribus) less bad than productive causation. The felt intuition in some that there should be no duty to rescue, even in cases where rescue is easy, nearly costless, or trivial, depends on the idea that omissions – failures to do – are in some deep metaphysical sense different from doings. That difference is sometimes seen as a fundamental difference in how one’s liberty is restricted by responsibility for acts versus responsibility for omissions. It’s further argued that tortious acts leave victims worse-off, whereas tortious omissions failed to improve someone’s lot. This project will tackle all three sorts of justifications and explore the so-called easy rescue or ‘bad samaritan’ cases as a paradigm. The easy rescue cases are a matter of political morality rather than metaphysics. 

Contractualism and Strict Liability in Tort

This project seeks to present a unified account of strict liability that addressed both strict liability for harms that were very much avoidable by the tortfeasor (like intentional torts) and strict liability for harms that were not (like blasting and other abnormally dangerous activity). Ex ante contractualism is used to show that strict liability for neither form of harm can be rejected. Ex ante contractualism is argued to be the correct decision-procedure, on grounds that subjunctive contractors should be in ignorance of their position or how they particularly would be affected by some scheme, at least when risk of harm is not overwhelming. Contractors are understood to permit imposition of relatively low-probability risks in a live and let live regime, but a rule that imposes huge burdens on some should not be allowed for the sake of relatively small benefits to others.

To be sure, contractualism concerns the moral rights (and correlative duties) that one should possess. And tort law is part of a legal regime. Still, we can meaningfully ask the normative questions: when, and why, is strict liability in tort justified? And, consider a further upshot of contractualist justification: if no-one can reasonably reject a strict liability regime in tort for such-and-such sorts of cases, then we obtain the result that morality itself accords us the rights (and concomitantly assigns correlative duties) pursuant to that scheme. It may still be the case that such rights and duties do not give rise to a legal cause of action, of course.

Methodologically, this is similar to, but different from, approaches that have been used extensively in tort theory like some version of rule-consequentialism or idealized deliberation behind a veil of ignorance. Contractualist contractors are trying to formulate norms at a level of generality and with a particular sort of ignorance, namely: the contractors lack extant deontic principles. Contractualism is meant to be an aufbau rising up from a stage prior to deontic norms to the level where we have deontic norms that are agreeable to via unanimous consent under substantive reasonableness; contractors who would reject some candidate rule are required to produce a personal (to them, non-aggregative) reason for their nay. Moreover, some version of the ‘greater burdens’ principle is in play: a contractor acts unreasonably in rejecting a candidate principle because it imposes some burden on him when every alternative thereto would imposed much greater burdens on others. To be sure, there are complexities to be considered with risks rather than definite outcomes, but recent work in contractualism is of much use on this front, and can pave the way for further advances in tort theory that, like prior advances, import in relevant respects the state of the art in moral philosophy. In yesteryear’s tort theory, that was, say Kantianism or rule-consequentialism; today it’s contractualism. Finally, this can be a two-way street, improving contractualist accounts. 

Evidence

In, and around, the law of evidence I bring to bear tools from epistemology and other subjects to throw light on debates and discussions. Some of this work seeks to get clearer about what our standards of proof are and their justification. Other projects try to tackle what legal evidence needs to do. Some of the work is more theoretically oriented, while some of it seeks to explicate exactly what a particular rule covers and what it does not.

Law, Language, and Aboutness: Diaz v. United States as Case-Study [accepted]

The Supreme Court’s recent decision in Diaz v. United States is a simple interpretive dispute concerning when expert testimony is ‘about’ a defendant’s possession of mens rea conducted by justices who are avowedly textualist and yet reached diametrically opposite conclusions. While the court reached the right result in permitting testimony concerning behavior of members of a class including the criminal defendant, the rationale it offered is at best incomplete and at worst incorrect.

This essay introduces into legal analysis the rigorous study of aboutness in analytic philosophy, which is used to augment, rather than attack, the textualist analyses offered by the justices. The essay develops the tools of aboutness in a straightforward way that should serve as a primer for those interested in the technique and its value – a group which should definitely include those interested in textualist interpretation. 

In addition to this, the paper explores the problematic results of rationales for the result proffered by the justices. This exploration clarifies the difference between testimony about subjective versus objective probability. It also shows how unintended consequences arise from some versions of a blanket rule, such as the dissent in Diaz proposed.

Taking aboutness seriously elegantly justifies the correct result in Diaz. The essay concludes on the speculative note that aboutness can be used to resolve other intra-textualist disputes, such as the thorny one in Bostock v. Clayton County. Thus, attending to aboutness allows us to cleanly explain the meaning of Federal Rule of Evidence 704, and it may lead to further cogenr explanations where thorny disputes have pervaded the law even against the common ground of shared textualism.

Cosim will present some of this work to the Evidence Summer Workshop at Vanderbilt Law School in May 2025, where his focus will be when expert testimony is about an accused's mental state. 

Naked Stats and Putting Knowledge First

There are well-known puzzle cases that arise in connection with naked statistical evidence, where, standardly, there is a probability on bare statistics well in excess of whatever applicable threshold value is thought required. Think of the gatecrasher case, where exactly one ticket was sold for the show, the gate was crashed, and there are 10,000 people in the stands as a result of the gate having been crashed. It’s not known to whom the ticket was sold, nor is it known who crashed to gate. But for any person in the stands, the probability that it’s not a gatecrasher is only 10 – 4. (Obviously, these numbers can be adjusted to add more zeroes.)

Suppose that folks in the stands are sued for trespass, and the applicable standard is that of a preponderance of the evidence, which is usually taken to be satisfied only when liability is more likely than not. I endeavor to show that puzzles like the gatecrasher case and variations on it cannot be successfully handled by various proposed accounts of legal proof, including (inter alia) Martin Smith’s normalcy condition, the sensitivity condition proposed by David Enoch and colleagues, and the promising individualized evidence criterion that Judith Jarvis Thomson enunciated. The reason why counterexamples to these (and other) approaches succeed seems to be that there is an epistemic anti-luck requirement, as with knowledge, but the foregoing proposed conditions on legal evidence all permit some sort of luck to go through unchecked. 

This strongly suggests that an account based on knowledge of liability would be strong enough to stop counterexamples. And while that’s certainly true, knowledge is far too onerous of a requirement, at least in most civil cases, and likely even in criminal cases, since while knowledge plausibly entails the lack of reasonable doubt, lacking reasonable doubt doubtless does not entail knowledge. I propose to examine alternatives that are weaker than requiring knowledge, including (inter alia) considering the probability that one knows, evaluating a knowledge-first internalist position advanced by Sven Rosenkranz where there is justification for some proposition just in case one’s in no position to know one is in no position to know that proposition, and appraising a proposal that looks at whether evidence adduced at trial together with other kinds of plausible related evidence would make liability knowable. 

Only the Best Explanation Wins: Legal Proof, Explanation, and Probability

(with Yuval Abrams (Michigan State, Law))

To legally prove a claim p, it must, qua explanation, and qualitatively, be the best explanation for the evidence. If a competing claim q equally or better explains the evidence qua explanation than does p, then p is clearly not proven. In other words: a version of Inference to the Best Explanation (IBE) is a necessary condition for legal proof. This holds for all legal proof (civil, criminal, administrative) and on all standards of proof (from preponderance to beyond reasonable doubt). We explore what sorts of explanations can be combined in which sorts of ways. Further, this insight re: IBE sheds special light on naked statistical evidence (NSE), since NSE always fails to satisfy the necessary IBE constraint. It’s shown how this requirement relates to, and differs from, alternative possibilities in the NSE literature.

***** PAGE UNDER CONSTRUCTION *****

Cosim wrote a piece about knowledge-norms and the mere likelihood evidentiary standard that's applicable in most common-law civil matters; it's now published. It's part of Cosim's interest in issues at the intersection of language and norms. Cosim is writing up an account of lying that takes seriously the semantics of ascribing lying, and which uncovers a species of lying that has not received as much attention in its moral dimensions.