Can you moderate an unreadable message? 'Blind' content moderation via human computation

User-generated content (UGC) is fundamental to online social engagement, but eliciting and managing it come with many challenges. The special features of UGC moderation highlight many of the general challenges of human computation in general. They also emphasize how moderation and privacy interact: people have rights to both privacy and safety online, but it is difficult to provide one without violating the other: scanning a user's inbox for potentially malicious messages seems to imply access to all safe ones as well. Are privacy and safety opposed, or is it possible in some circumstance to guarantee the safety of anonymous content without access to that content. We demonstrate that such "blind content moderation" is possible in certain domains. Additionally, the methods we introduce offer safety guarantees, an expressive content space, and require no human moderation load: they are safe, expressive, and scalable Though it may seem preposterous to try moderating UGC without human- or machine-level access to it, human computation makes blind moderation possible. We establish this existence claim by defining two very different human computational methods, behavioral thresholding and reverse correlation. Each leverages the statistical and behavioral properties of so-called "inappropriate content" in different decision settings to moderate UGC without access to a message's meaning or intention. The first, behavioral thresholding, is shown to generalize the well-known ESP game.

Kant’s conception of self as subject and its embodiment

This paper examines Kant’s conception of the self as subject, to show that it points to an understanding of the self as embodied. By considering ways in which the manifold of representations can be unified, different notions of self are identified through the subjective perspectives they define. This involves an examination of Kant’s distinction between subjective and objective unities of consciousness, and the notion of empirical unity of apperception in the first Critique, as well as the discussion of judgements of perception and judgements of experience in the Prolegomena. If this identifies the self as subject through its embodied perspective, it leaves open the question of the self’s existence. The paper proposes an interpretation of Kant’s brief statements on this matter which provides grounds for an existence claim that extends to the embodied self. This suggests considering Kant’s view of the certainty of one’s existence as involving a feeling of self, albeit in a specifically Kantian form drawing on the transcendental unity of apperception. This feeling of self introduces a practical dimension that supports the claim that any determination of the self as subject has to be practical, namely as free agent subject to inclinations, insofar as he is embodied.

A non-Russellian treatment of the referential-attributive distinction

Kripke made a good case that “…the phi…” is not semantically ambiguous between referential and attributive meanings. Russell says that “…the phi…” is always to be analyzed attributively. Many semanticists, agreeing with Kripke that “…the phi…” is not ambiguous, have tried to give a Russellian analysis of the referential-attributive distinction: the gross deviations between what is communicated by “…the phi..”, on the one hand, and what Russell’s theory says it literally means, on the other, are chalked up to implicature. This paper shows that, when the phenomenon of implicature is scrutinized, there is overwhelming reason to doubt that a Russellian analysis can succeed. A positive, non-Russellian analysis is proposed: it is shown that, if definite descriptions are treated as referring expressions, it is easy to deal with the referential-attributive distinction. When “…the phi…” is functioning attributively, the definite description is seen as referring to some object described in an understood, antecedent existence claim.