We propose a formal framework for interpretable machine learning. Combining elements from statistical learning, causal interventionism, and decision theory, we design an idealised explanation game in which players collaborate to find the best explanation for a given algorithmic prediction. Through an iterative procedure of questions and answers, the players establish a three-dimensional Pareto frontier that describes the optimal trade-offs between explanatory accuracy, simplicity, and relevance. Multiple rounds are played at different levels of abstraction, allowing the players to explore overlapping causal (...) patterns of variable granularity and scope. We characterise the conditions under which such a game is almost surely guaranteed to converge on a optimal explanation surface in polynomial time, and highlight obstacles that will tend to prevent the players from advancing beyond certain explanatory thresholds. The game serves a descriptive and a normative function, establishing a conceptual space in which to analyse and compare existing proposals, as well as design new and improved solutions. (shrink)

We propose a formal framework for interpretable machine learning. Combining elements from statistical learning, causal interventionism, and decision theory, we design an idealisedexplanation gamein which players collaborate to find the best explanation(s) for a given algorithmic prediction. Through an iterative procedure of questions and answers, the players establish a three-dimensional Pareto frontier that describes the optimal trade-offs between explanatory accuracy, simplicity, and relevance. Multiple rounds are played at different levels of abstraction, allowing the players to explore overlapping causal patterns of (...) variable granularity and scope. We characterise the conditions under which such a game is almost surely guaranteed to converge on a (conditionally) optimal explanation surface in polynomial time, and highlight obstacles that will tend to prevent the players from advancing beyond certain explanatory thresholds. The game serves a descriptive and a normative function, establishing a conceptual space in which to analyse and compare existing proposals, as well as design new and improved solutions. (shrink)

Recent years have seen a surge in online collaboration between experts and amateurs on scientific research. In this article, we analyse the epistemological implications of these crowdsourced projects, with a focus on Zooniverse, the world’s largest citizen science web portal. We use quantitative methods to evaluate the platform’s success in producing large volumes of observation statements and high impact scientific discoveries relative to more conventional means of data processing. Through empirical evidence, Bayesian reasoning, and conceptual analysis, we show how information (...) and communication technologies enhance the reliability, scalability, and connectivity of crowdsourced e-research, giving online citizen science projects powerful epistemic advantages over more traditional modes of scientific investigation. These results highlight the essential role played by technologically mediated social interaction in contemporary knowledge production. We conclude by calling for an explicitly sociotechnical turn in the philosophy of science that combines insights from statistics and logic to analyse the latest developments in scientific research. (shrink)

The Gettier Problem (henceforth GP) is a milestone in contemporary epistemology. Half a century after the philosophical earthquake caused by the well-known Gettier article (1963), solutions continue to appear and the discussion resurfaces. Because we, like many, choose to adopt an optimistic stance towards the possibility of finding a plausible answer to the GP, in this paper we present an unconventional solution to solve it. Crucially, we argue that the only plausible and non-falsifiable definition of the propositional knowledge is the (...) one given by an “open” formula that contains a placeholder or a “variable” – on the third necessary condition – which takes a value according to which, depending on the circumstances, is necessary and sufficient for epistemic success. (shrink)

This paper addresses the problem of upgrading functional information to knowledge. Functional information is defined as syntactically well-formed, meaningful and collectively opaque data. Its use in the formal epistemology of information theories is crucial to solve the debate on the veridical nature of information, and it represents the companion notion to standard strongly semantic information, defined as well-formed, meaningful and true data. The formal framework, on which the definitions are based, uses a contextual version of the verificationist principle of truth (...) in order to connect functional to semantic information, avoiding Gettierization and decoupling from true informational contents. The upgrade operation from functional information uses the machinery of epistemic modalities in order to add data localization and accessibility as its main properties. We show in this way the conceptual worthiness of this notion for issues in contemporary epistemology debates, such as the explanation of knowledge process acquisition from information retrieval systems, and open data repositories. (shrink)

The philosophy of information (PI) is a new area of research with its own field of investigation and methodology. This article, based on the Herbert A. Simon Lecture of Computing and Philosophy I gave at Carnegie Mellon University in 2001, analyses the eighteen principal open problems in PI. Section 1 introduces the analysis by outlining Herbert Simon's approach to PI. Section 2 discusses some methodological considerations about what counts as a good philosophical problem. The discussion centers on Hilbert's famous analysis (...) of the central problems in mathematics. The rest of the article is devoted to the eighteen problems. These are organized into five sections: problems in the analysis of the concept of information, in semantics, in the study of intelligence, in the relation between information and nature, and in the investigation of values. (shrink)

Epistemic diversity is the ability or possibility of producing diverse and rich epistemic apparati to make sense of the world around us. In this paper we discuss whether, and to what extent, different conceptions of knowledge—notably as ‘justified true belief’ and as ‘distributed and embodied cognition’—hinder or foster epistemic diversity. We then link this discussion to the widespread move in science and philosophy towards monolingual disciplinary environments. We argue that English, despite all appearance, is no Lingua Franca, and we give (...) reasons why epistemic diversity is also deeply hindered is monolingual contexts. Finally, we sketch a proposal for multilingual academia where epistemic diversity is thereby fostered. (shrink)

The article contains the replies to the collection of contributions discussing my research on the philosophy of information.

In this article, I outline a logic of design of a system as a specific kind of conceptual logic of the design of the model of a system, that is, the blueprint that provides information about the system to be created. In section two, I introduce the method of levels of abstraction as a modelling tool borrowed from computer science. In section three, I use this method to clarify two main conceptual logics of information inherited from modernity: Kant’s transcendental logic (...) of conditions of possibility of a system, and Hegel’s dialectical logic of conditions of in/stability of a system. Both conceptual logics of information analyse structural properties of given systems. Strictly speaking, neither is a conceptual logic of information about the conditions of feasibility of a system, that is, neither is a logic of information as a logic of design. So, in section four, I outline this third conceptual logic of information and then interpret the conceptual logic of design as a logic of requirements, by introducing the relation of “sufficientisation”. In the conclusion, I argue that the logic of requirements is exactly what we need in order to make sense of, and buttress, a constructionist approach to knowledge. (shrink)

The use of “levels of abstraction” in philosophical analysis (levelism) has recently come under attack. In this paper, I argue that a refined version of epistemological levelism should be retained as a fundamental method, called the method of levels of abstraction. After a brief introduction, in section “Some Definitions and Preliminary Examples” the nature and applicability of the epistemological method of levels of abstraction is clarified. In section “A Classic Application of the Method ofion”, the philosophical fruitfulness of the new (...) method is shown by using Kant’s classic discussion of the “antinomies of pure reason” as an example. In section “The Philosophy of the Method of Abstraction”, the method is further specified and supported by distinguishing it from three other forms of “levelism”: (i) levels of organisation; (ii) levels of explanation and (iii) conceptual schemes. In that context, the problems of relativism and antirealism are also briefly addressed. The conclusion discusses some of the work that lies ahead, two potential limitations of the method and some results that have already been obtained by applying the method to some long-standing philosophical problems. (shrink)

In this article, I define and then defend the principle of information closure (pic) against a sceptical objection similar to the one discussed by Dretske in relation to the principle of epistemic closure. If I am successful, given that pic is equivalent to the axiom of distribution and that the latter is one of the conditions that discriminate between normal and non-normal modal logics, a main result of such a defence is that one potentially good reason to look for a (...) formalization of the logic of “ $S$ is informed that $p$ ” among the non-normal modal logics, which reject the axiom, is also removed. This is not to argue that the logic of “ $S$ is informed that $p$ ” should be a normal modal logic, but that it could still be insofar as the objection that it could not be, based on the sceptical objection against pic, has been removed. In other word, I shall argue that the sceptical objection against pic fails, so such an objection provides no ground to abandon the normal modal logic B (also known as KTB) as a formalization of “ $S$ is informed that $p$ ”, which remains plausible insofar as this specific obstacle is concerned. (shrink)

This article offers an account and defence of constructionism, both as a metaphilosophical approach and as a philosophical methodology, with references to the so-called maker's knowledge tradition. Its main thesis is that Plato's “user's knowledge” tradition should be complemented, if not replaced, by a constructionist approach to philosophical problems in general and to knowledge in particular. Epistemic agents know something when they are able to build (reproduce, simulate, model, construct, etc.) that something and plug the obtained information into the correct (...) network of relations that account for it. Their epistemic expertise increases with the scope and depth of the questions that they are able to ask and answer. Thus, constructionism deprioritises mimetic, passive, and declarative knowledge that something is the case, in favour of poietic, interactive, and practical knowledge of something being the case. Metaphilosophically, constructionism suggests adding conceptual engineering to conceptual analysis as a fundamental method. (shrink)

This is the revised version of an invited keynote lecture delivered at the "1st Australian Computing and Philosophy Conference". The paper is divided into two parts. The first part defends an informational approach to structural realism. It does so in three steps. First, it is shown that, within the debate about structural realism, epistemic and ontic structural realism are reconcilable. It follows that a version of OSR is defensible from a structuralist-friendly position. Second, it is argued that a version of (...) OSR is also plausible, because not all relata are logically prior to relations. Third, it is shown that a version of OSR is also applicable to both sub-observable and observable entities, by developing its ontology of structural objects in terms of informational objects. The outcome is informational structural realism, a version of OSR supporting the ontological commitment to a view of the world as the totality of informational objects dynamically interacting with each other. The paper has been discussed by several colleagues and, in the second half, ten objections that have been moved to the proposal are answered in order to clarify it further. (shrink)

In this paper, I lend credence to the move toward non-reductive religious epistemology by highlighting the systematic failings of Alvin Plantinga’s seminal, religious epistemology when it comes to surmounting the Gettier Problem. Taking Plantinga’s account as archetypal, I argue that we have systematic reasons to believe that no reductive theory of knowledge can viably surmount the Gettier Problem, that the future of religious epistemology lies in non-reductive models of knowledge.

The Gettier dilemma facing reductive analyses of knowledge has not been properly appreciated by virtue epistemologist or even virtue epistemology’s most vocal critics. In §1, we start by considering how recent critics of virtue epistemology understand the Gettier Problem facing virtue-theoretic accounts of knowledge. I highlight how the dilemma facing virtue- theoretic analyses of knowledge is more general than these critics seem to suggest. In §2, I elucidate the worry that the threat facing virtue epistemology is really a dilemma between (...) Gettier counterexamples and radical skepticism. In §3, we will consider how some recent virtue epistemologists have tried to viably defuse the Gettier Problem. We will see (i) just how the critiques elucidated in §1 have (mis)shaped the dialectic between virtue epistemology and what is required in solving Gettier counterexamples and (ii) how this has led to virtue epistemologists underestimating the widespread insidiousness of Gettier counterexamples. (shrink)

If the history of the Gettier Problem has taught us anything, it is to be skeptical regarding purported solutions. Nevertheless, in “Manifest Failure: The Gettier Problem Solved” (2011), that is precisely what John Turri offers us. For nearly fifty years, epistemologists have been chasing a solution for the Gettier Problem but with little to no success. If Turri is right, if he has actually solved the Gettier Problem, then he has done something that is absolutely groundbreaking and really quite remarkable. (...) Regrettably, however, while Turri’s account is both intuitive and elegant—improving upon many seminal projects within contemporary epistemology—I argue in this paper that any success against Gettier counterexamples it affords is merely fleeting. Straightforwardly, this is done in two sections. In §1, I briefly sketch Turri’s proposed solution to the Gettier Problem. Then, in §2, I level a counterexample against it. Unfortunately for Turri and his solution, in this paper we will see history repeat itself. (shrink)

That in Gettier's alleged counterexamples to the traditional analysis of knowledge as justified true belief the belief condition is satisfied has rarely been questioned. Yet there is reason to doubt that a rational person would come to believe what Gettier's protagonists are said to believe in the way they are said to have come to believe it. If they would not, the examples are not counter-examples to the traditional analysis. I go on to discuss a number of examples inspired by (...) Gettier's and argue that they, too, fail to be counter-examples either for reasons similar to those I have urged or because it is not clear that their subject does not know. (shrink)

The definition of knowledge as justified true belief is the best we presently have. However, the canonical tripartite analysis of knowledge does not do justice to it due to a Platonic conception of a priori truth that puts the cart before the horse. Within a pragmatic approach, I argue that by doing away with a priori truth, namely by submitting truth to justification, and by accordingly altering the canonical analysis of knowledge, this is a fruitful definition. So fruitful indeed that (...) it renders the Gettier counterexamples vacuous, allowing positive work in epistemology and related disciplines. (shrink)

The logic of ‘being informed’ gives a formal analysis of a cognitive state that does not coincide with either belief, or knowledge. To Floridi, who first proposed the formal analysis, the latter is supported by the fact that unlike knowledge or belief, being informed is a factive, but not a reflective state. This paper takes a closer look at the formal analysis itself, provides a pure and an applied semantics for the logic of being informed, and tries to find out (...) to what extent the formal analysis can contribute to an information-based epistemology. (shrink)

Cognitive states as well as cognitive commodities play central though distinct roles in our epistemological theories. By being attentive to how a difference in their roles affects our way of referring to them, we can undoubtedly accrue our understanding of the structure and functioning of our main epistemological theories. In this paper we propose an analysis of the dichotomy between states and commodities in terms of the method of abstraction, and more specifically by means of infomorphisms between different ways to (...) classify states of information, information-bases, and evidential situations. (shrink)

n various publications over the past years, Floridi has developed a theory of semantic information as well-formed, meaningful, and truthful data. This theory is more or less orthogonal to the standard entropy-based notions of information known from physics, information theory, and computer science that all define the amount of information in a certain system as a scalar value without any direct semantic implication. In this context the question rises what the exact relation between these various conceptions of information is and (...) whether there is a real need to enrich these mathematically more or less rigid definitions with a less formal notion of semantic information. I investigate various philosophical aspects of the more formal definitions of information in the light of Floridi’s theory. The position I defend is that the formal treatment of the notion of information as a general theory of entropy is one of the fundamental achievements of modern science that in itself is a rich source for new philosophical reflection. This makes information theory a competitor of classical epistemology rather than a servant. In this light Floridi’s philosophy of information is more a reprise of classical epistemology that only pays lip service to information theory but fails to address the important central questions of philosophy of information. Specifically, I will defend the view that notions that are associated with truth, knowledge, and meaning all can adequately be reconstructed in the context of modern information theory and that consequently there is no need to introduce a concept of semantic information. (shrink)

In various publications over the past years, Floridi has developed a theory of semantic information as well-formed, meaningful, and truthful data. This theory is more or less orthogonal to the standard entropy-based notions of information known from physics, information theory, and computer science that all define the amount of information in a certain system as a scalar value without any direct semantic implication. In this context the question rises what the exact relation between these various conceptions of information is and (...) whether there is a real need to enrich these mathematically more or less rigid definitions with a less formal notion of semantic information. I investigate various philosophical aspects of the more formal definitions of information in the light of Floridi’s theory. The position I defend is that the formal treatment of the notion of information as a general theory of entropy is one of the fundamental achievements of modern science that in itself is a rich source for new philosophical reflection. This makes information theory a competitor of classical epistemology rather than a servant. In this light Floridi’s philosophy of information is more a reprise of classical epistemology that only pays lip service to information theory but fails to address the important central questions of philosophy of information. Specifically, I will defend the view that notions that are associated with truth, knowledge, and meaning all can adequately be reconstructed in the context of modern information theory and that consequently there is no need to introduce a concept of semantic information. (shrink)

Abstract: Luciano Floridi has impressively applied the concept of information to problems in semantics and epistemology, among other areas. In this essay, I briefly review two areas where I think one may usefully raise questions about some of Floridi's conclusions. One area is in the project to naturalize semantics and Floridi's use of the derived versus nonderived notion of semantic content. The other area is in the logic of information and knowledge and whether knowledge based on information necessarily supports closure, (...) in every instance. I suggest that it does not and, thereby, raise a challenge to Floridi's logic of being informed. (shrink)

In this paper we introduce three methods to approach philosophical problems informationally: Minimalism, the Method of Abstraction and Constructionism. Minimalism considers the specifications of the starting problems and systems that are tractable for a philosophical analysis. The Method of Abstraction describes the process of making explicit the level of abstraction at which a system is observed and investigated. Constructionism provides a series of principles that the investigation of the problem must fulfil once it has been fully characterised by the previous (...) two methods. For each method, we also provide an application: the problem of visual perception, functionalism, and the Turing Test, respectively. (shrink)

This thesis addresses the problem of the analysis of knowledge. The persistent failure of analyses of knowledge in terms of necessary and sufficient conditions is used to motivate exploring alternative approaches to the analytical problem. In parallel to a similar development in the theory of truth, in which the persistent failure to provide a satisfactory answer to the question as to what the nature of truth is has led to the exploration of deflationary and minimalist approaches to the theory of (...) truth, the prospects for deflationary and minimalist approaches to the theory of knowledge are investigated. While it is argued that deflationary approaches are ultimately unsatisfactory, a minimalist approach to epistemology, which characterises the concept of knowledge by a set of platitudes about knowledge, is defended. The first version of a minimalist framework for the theory of knowledge is developed. Two more substantive developments of the minimalist framework are discussed. In the first development a safety condition on knowledge is derived from the minimalist framework. Problems for this development are discussed and solved. In the second development, an ability condition is derived from the minimalist framework. Reason is provided to believe that, arguably, the ability condition can avoid the problems that beset traditional analyses of knowledge. It is also shown that even if this argument fails, minimalist approaches to epistemology may serve to provide a functional definition of knowledge. Reason is thus provided to believe that minimalist approaches to epistemology can make progress towards addressing the problem of the analysis of knowledge. (shrink)

The use of "levels of abstraction" in philosophical analysis (levellism) has recently come under attack. In this paper, we argue that a refined version of epistemological levellism should be retained as a fundamental method, which we call the method of abstraction. After a brief introduction, in section two we make clear the nature and applicability of the (epistemological) method of levels of abstraction. In section three, we show the fruitfulness of the new method by applying it to five case studies: (...) the concept of agenthood, the Turing test, the definition of emergence, quantum observation and decidable observation. In section four, we further characterise and support the method by distinguishing it from three other forms of "levellism": (i) levels of organisation; (ii) levels of explanation and (iii) conceptual schemes. In this context, we also briefly address the problems of relativism and antirealism. In the conclusion, we indicate some of the work that lies ahead, two potential limitations of the method and some results that have already been obtained by applying the method to some long-standing philosophical problems. (shrink)

This chapter addresses two questions. First, if knowledge is accounted information, how are we supposed (to apply this analysis in order) to understand perceptual knowledge and knowledge by testimony? In the first part of the chapter, I articulate an answer in terms of a re-interpretation of perception and testimony as data providers rather than full-blown cases of knowledge. Second, if perception and testimony are correctly understood as data providers, how are we supposed (to apply this analysis in order) to understand (...) the semantic value of the data provided by such processes? In the second part of the chapter, I argue in favour of a constructionist hypothesis about how data may become meaningful for human cognitive agents through a process of repurposing of natural data/signals. The conclusion of the chapter is that human agents are natural-born data hackers. (shrink)

One of the open problems in the philosophy of information is whether there is an information logic (IL), different from epistemic (EL) and doxastic logic (DL), which formalises the relation “a is informed that p” (Iap) satisfactorily. In this paper, the problem is solved by arguing that the axiom schemata of the normal modal logic (NML) KTB (also known as B or Br or Brouwer’s system) are well suited to formalise the relation of “being informed”. After having shown that IL (...) can be constructed as an informational reading of KTB, four consequences of a KTB-based IL are explored: information overload; the veridicality thesis (Iap → p); the relation between IL and EL; and the Kp → Bp principle or entailment property, according to which knowledge implies belief. Although these issues are discussed later in the article, they are the motivations behind the development of IL. (shrink)

Philosophers of information, according to Luciano Floridi (The philosophy of information. Oxford University Press, Oxford, 2010, p 32), study how information should be “adequately created, processed, managed, and used.” A small number of epistemologists have employed the concept of information as a cornerstone of their theoretical framework. How this concept can be used to make sense of seemingly intractable epistemological problems, however, has not been widely explored. This paper examines Fred Dretske’s information-based epistemology, in particular his response to radical epistemological (...) skepticism. We discuss the relationship between information, evidence and knowledge in relation to the problem of skepticism and the options available to an information-based epistemology for dealing with it. (shrink)

The concept of distributed moral responsibility (DMR) has a long history. When it is understood as being entirely reducible to the sum of (some) human, individual and already morally loaded actions, then the allocation of DMR, and hence of praise and reward or blame and punishment, may be pragmatically difficult, but not conceptually problematic. However, in distributed environments, it is increasingly possible that a network of agents, some human, some artificial (e.g. a program) and some hybrid (e.g. a group of (...) people working as a team thanks to a software platform), may cause distributed moral actions (DMAs). These are morally good or evil (i.e. morally loaded) actions caused by local interactions that are in themselves neither good nor evil (morally neutral). In this article, I analyse DMRs that are due to DMAs, and argue in favour of the allocation, by default and overridably, of full moral responsibility (faultless responsibility) to all the nodes/agents in the network causally relevant for bringing about the DMA in question, independently of intentionality. The mechanism proposed is inspired by, and adapts, three concepts: back propagation from network theory, strict liability from jurisprudence and common knowledge from epistemic logic. (shrink)

El término “conocimiento” y la disciplina filosófica que lo estudia —la teoría del conocimiento— han experimentado notables cambios hasta el presente. La teoría clásica concibe el conocimiento en íntima unión con la verdad, como una captación intelectual de realidades necesarias e inmutables. Con la llegada de la modernidad, la difusión de un clima escéptico puso en duda esta pretensión, cuestionando la aptitud misma del conocimiento para la verdad. Esta duda ha presidido toda la modernidad hasta el presente. Para responder al (...) desafío escéptico, las principales corrientes de la llamada epistemología analítica contemporánea han intentado, sin éxito, explorar el carácter justificante del conocimiento. En este intento, han destacado teorías encaminadas a mostrar que el conocimiento recibe su justificación desde un fundamento, como el fundacionismo y el coherentismo; teorías encaminadas a mostrar que el conocimiento recibe su justificación desde sus fuentes, como el externismo y el internismo; y teorías encaminadas a mostrar que la recibe por las facultades cognitivas mismas como instrumentos de conocimiento fiables, como la epistemología de la virtud. Todas ellas generan, a su vez, nuevas paradojas e imponen la necesidad de volver a los fundamentos de esta disciplina, presentes en la teoría clásica, para descubrir la conexión entre conocimiento y verdad, e intentar detener el desafío escéptico. (shrink)

El artículo muestra la forma en que laepistemología de las virtudes puederesolver algunos de los problemasepistemológicos clásicos, y la manera enque puede lograrse una concepciónintegrada de las variantes confiabilistay responsabilista de la epistemología delas virtudesThis paper shows the way in which Virtue Epistemology can solve some ofthe classical epistemological problems,and the way in which an integrated approach of the Reliabilist and Responsibilist variants of Virtue Epistemology can be achieved.

The Philosophy of Information is a new area of research at the intersection of philosophy and computer science [4]. It concerns (a) the critical investigation of the conceptual nature and basic principles of information, including its dynamics (especially computation), utilization (especially computer ethics) and sciences; and (b) the elaboration and application of computational and information-theoretic methodologies to philosophical problems. Past work by members of our group has concentrated on (a), and in this paper we explore (b). In a nutshell, we (...) ask what computer science can do for philosophy, rather than what the latter can do for the former. -/- . (shrink)

This dissertation is on infinite regress arguments in philosophy. Its main goals are to explain what such arguments from many distinct philosophical debates have in common, and to provide guidelines for using and evaluating them. Two theories are reviewed: the Paradox Theory and the Failure Theory. According to the Paradox Theory, infinite regress arguments can be used to refute an existentially or universally quantified statement (e.g. to refute the statement that at least one discussion is settled, or the statement that (...) discussions are settled only if there is an agreed-upon criterion to settle them). According to the Failure Theory, infinite regress arguments can be used to demonstrate that a certain solution fails to solve an existentially or universally quantified problem (e.g. to demonstrate that a certain solution fails to settle all discussions, or that it fails to settle even one discussion). In the literature, the Paradox Theory is fairly well-developed, and this dissertation provides the Failure Theory with the same tools. (shrink)

This paper contributes to the current debate on the nature of semantic information by offering a semantic argument in favour of the veridical thesis according to which p counts as information only if p is true. In the course of the analysis, the paper reviews some basic principles and requirements for any theory of semantic information.