Academic
Publications
Joint attention, openness, and self-other (in)differentiation
Journal of Consciousness Studies (2025)
Joint attention is characterised by openness: when two agents jointly attend to an object, they are immediately and fully aware of each other’s attentional states. Existing accounts of openness involve a mental picture in which two agents are attending to the same object and where openness is then ‘added’. This is problematic. The experience that attentional states are open comes first. When we look at young infants, we can see that they operate under a tacit assumption of openness: they behave as if attentional states were open even when they are not. The ability to engage in joint attention doesn’t arise when infants begin to experience openness, but rather when they can limit these experiences to open interactions. This development depends on cognitive processes that enable infants to detect non-open interactions so that such interactions can be excluded from those in which openness is experienced. Some of these processes develop early and do not require the representation of mental states. Other processes develop later and require the infant to differentiate between herself and the other as subjects of attentional states. The result is an account that is pegged at the right level of cognitive sophistication and in line with empirical findings. The account explains the experiential nature of openness and how joint attention can provide a rational basis for joint action. It does so, moreover, without relying on relationalism about perception.
Phenomenal transparency and the boundary of cognition (with Hadeel Naeem)
Phenomenology and the Cognitive Sciences (2024)
Phenomenal transparency was once widely believed to be necessary for cognitive extension. Recently, this claim has come under attack, with a new consensus coalescing around the idea that transparency is neither necessary for internal nor extended cognitive processes. We take these recent critiques as an opportunity to refine the concept of transparency relevant for cognitive extension. In particular, we highlight that transparency concerns an agent’s employment of a resource – and that such employment is compatible with an agent consciously apprehending (or attending to) a resource. This means it is possible for an object to be transparent and opaque to an agent, even at a single moment in time. Once we understand transparency in this way, the detractors’ claims lose their bite, and existing arguments for transparency’s necessity for cognitive extension return to apply with full force.
Should we discourage AI extension? (with Hadeel Naeem)
Philosophy & Technology (2024)
We might worry that our seamless reliance on AI systems makes us prone to adopting the strange errors that these systems commit. One proposed solution is to design AI systems so that they are not phenomenally transparent to their users. This stops cognitive extension and the automatic uptake of errors. Although we acknowledge that some aspects of AI extension are concerning, we can address these concerns without discouraging transparent employment altogether. First, we believe that the potential danger should be put into perspective – many unreliable technologies are unlikely to be used transparently precisely because they are unreliable. Second, an agent who transparently employs a resource may also reflect (opaquely) on its reliability. Finally, agents can rely on a process transparently and be yanked out of their transparent use when it turns unreliable. When an agent is responsive to the reliability of their process in this way, they have epistemically integrated it, and the beliefs they form with it are formed responsibly. This prevents the agent from automatically incorporating problematic beliefs. Responsible (and transparent) use of AI resources – and consequently responsible AI extension – is hence possible. We end the paper with several design and policy recommendations that encourage epistemic integration of AI-involving belief-forming processes.
Sharing is caring vs. stealing is wrong: A moral argument for limiting copyright protection.
International Journal of Technology Policy and Law (2017)
Copyright is at the centre of both popular and academic debate. That emotions are running high is hardly surprising - copyright influences who contributes what to culture, how culture is used, and even the kind of persons we are and come to be. Consequentialist, Lockean, and personality interest accounts are generally advanced in the literature to morally justify copyright law. I argue that these approaches fail to ground extensive authorial rights in intellectual creations and that only a small subset of the rights accorded by copyright law is justified. The pared-down version of copyright that I defend consists of the right to attribution, the right to have one’s non-endorsement of modifications or uses of one’s work explicitly noted, and the right to a share of the profit resulting from the commercial uses of one’s work. I also cursorily explore whether contribution to another person’s work gives rise to moral interests.
Under review
Thinking of oneself as someone: the structure of minimal self-representation
One question we can ask when investigating the nature of self-representation concerns the types of property that must figure in its content. Here, authors have claimed that self-representations need to be about spatial, temporal, bodily, or mental properties. However, we can also ask a second question: how do we need to represent a property to self-represent it? It is this latter question that I address. I argue that a distinction between egocentric and non-egocentric forms of representation – known primarily from the literature on spatial cognition – also applies to representations of other kinds of property. I use examples drawn from animal cognition and developmental psychology to show how creatures non-egocentrically represent their temporal, bodily, and cognitive properties. These representations are, I submit, minimal self-representations: they involve representing one’s properties so that an explicit differentiation is made between the system’s and other objects’ properties (or between the system’s actual and merely possible properties), they are directly linked to behaviour and sensation, and they are immune to error through misidentification. The upshot is a view on which different creatures may self-represent (in this minimal sense) more or fewer kinds of property. More substantive forms of self-representation (for instance, as exemplified by neurotypical adult human beings) then require integrated minimal self-representations of the right kinds of property.
Dispositions and objects’ changing properties
Analyses of dispositions share the following formal structure: O has disposition D if O fulfils modal conditions C. This simple structure hides a difficult question: what is the relation between O in the analysandum and O in the analysans? Clearly, they must be numerically identical – and just as clearly, this is insufficient. Whether a thirty-year-old is disposed to wake up early is unaffected by their night owl teenage years. What, then, is an appropriate additional constraint? This paper argues that no suitable constraint has so far been advanced and that finding one presents important difficulties. We might think that the objects need to share all intrinsic properties – but that renders dispositions largely useless in explanation and prediction. As most objects change over time, we cannot, for instance, use our knowledge of someone being an early riser to infer that they will get up early tomorrow. We might, in contrast, think that the objects need to share only some of their intrinsic properties. This approach is more promising but requires explaining which properties need to be shared. I develop a constraint according to which the objects need to share the causal basis (inspired by Lewis’s reformed conditional analysis), but ultimately find it wanting. Finally, I argue that the puzzle of the relation between O in the analysandum and O in the analysans can help motivate some re-evaluation of how dispositions are affected by objects’ dynamic natures.
Work in progress
AI am I
A paper arguing that the rise of AI assist might bring with it a radical change in self-conception. Personal assistants will be tightly enmeshed with their users, so that these will likely attribute abilities, beliefs, and so forth to themselves even if they are realised by the human-AI system jointly. However, AI assistants are also conversational partners with differing states (for instance, they know things we do not), and hence posited as an other. This tension between seeing them as part of the self and as an other might lead to a self-conception that is far less unified that what we might think necessary.
A paper on immunity to error through misidentification
Egocentric representations’ success conditions involve the agent’s actual state. Hence, the self figures as an unarticulated constituent in such representations. I argue that when this unarticulated constituent is articulated in an allocentric representation, the resulting self-attribution is IEM.
Non-academic
- ELSA Synergy Magazine: The evolution of the concept of privacy: From the American revolution, to big data and the internet of things (March 2015, republished as EDRi-gram ENDitorial)
- EDRi-gram: Bad analogies and the threat to “cybersecurity” (March 2015)
- EDRi-gram: France implements Internet censorship without judicial oversight (March 2015)
- EDRi-gram: UN report on copyright – is the EU really a beacon of human rights? (February 2015)
- Netzpiloten.de: EU-Urheberrechtsreform: Antwort auf den Berichtsentwurf (February 2015, translation into German of the EDRi-gram article on copyright reform)
- EDRi-gram: Copyright Reform: Responses to Draft Report and what comes next (February 2015)
- EDRi-gram: Data retention in Kosovo and Switzerland – legalising illegal laws (January 2015)
Videos
- What’s wrong and meta-wrong with copyright: philosophy and the copyright debate at Copycamp in Warsaw, Poland (November 2015)
- Urheberrecht: Sharing is caring vs. stealing is wrong (in German) at 40. Netzpolitischer Abend in Berlin, Germany (September 2015)