Kantian Cognitivism

According to many of its advocates, one of the main attractions of Kantian moral philosophy is its metaethical innocence. The most interesting argument for such innocence appeals to Kantians' rationalism. Roughly, if moral action is simply rational action, then we do not need to appeal to anything beyond rationality to certify moral judgment. I assess this argument by reflecting on (dis)analogies between moral and logical forms of rationalism. I conclude that the Kantian claim to metaethical innocence is overstated. Kantians cannot avoid substantial metaethical commitments. Or if they can, it is not their rationalism that explains why this is so.     

Against Cognitivism About Supposition

A popular view maintains that supposition is a kind of cognitive mental state, very similar to belief in essential respects. Call this view “cognitivism about supposition”. There are at least three grades of cognitivism, construing supposition as (i) a belief, (ii) belief-like imagination or (iii) a species of belief-like imagination. I shall argue against all three grades of cognitivism and claim that supposition is a sui generis form of imagination essentially dissimilar to belief. Since for good reasons (i) is not supported in the literature, I shall dwell on (ii) and (iii). Without further explanation supposition has been very often merely postulated as being nothing but belief-like imagination—that is, (ii). I shall show that at least two considerations undermine (ii). First, supposition and belief-like imagination are governed by different norms, more precisely the former is freer than the latter and requires minimal or no mental effort. Second, contrary to belief-like imagination, supposition is “cold”, in that it is typically dissociated from emotional reactions. Proponents of (iii) face the pressure of explaining these differences between supposition and belief-like imagination too. I shall argue that they have not sufficiently motivated the claim that supposition is belief-like. In particular they fail to accommodate precisely the dimensions of supposition pertaining to its normativity and emotionality. I shall close with a sketch of a new account of supposition.     

Descombes' Critique of Cognitivism

Truth‐telling is a project that is both gripping and problematic for Rousseau, as he is both captured by an ideal of telling as complete, undistorted discernment, documentation and communication, and also haunted by the fear that telling can never be this innocent. For Rousseau, as for Kant, telling does not leave the told untouched; rather, telling gives us a type of contact with objects that is marked and mediated by the process of telling itself, and hence the possibility of immediately grasping objects through telling is forever lost to us. The drive to capture things in themselves, which originates, according to Kant, in a formal principle of reason, shows up in Rousseau's writings as a nostalgia that governs and animates inquiry. I will read Rousseau, and the traumas of truth‐telling he articulates, as important pretexts for Kant's critical epistemology. Rousseau discloses tensions that infect his truth‐telling practices. Kant seeks to neutralize these tensions, not by dissolving them, but rather by translating them into the terms of transcendental philosophy and thus showing how they can be defused and rendered harmless in their empirical form, so as to secure the possibility of proper truth‐telling.     

Moral Cognitivism and Character

It may seem to follow from Peter Winch's claim in ‘The Universalizability of Moral Judgements’ that a certain class of first‐person moral judgments are not universalizable that such judgments cannot be given a cognitivist interpretation. But Winch's argument does not involve the denial of moral cognitivism and in this paper I show how such judgements may be cognitively determined yet not universalizable. Drawing on an example from James Joyce's The Dead, I suggest that in the kind of situation Winch envisages where we properly return a different moral judgement to another agent it may be that we accept their judgement is right for them because we recognise that it is determined by values that, simply because of the particular people we are, we could never know or understand in just the same way.     

Against Cognitivism about Practical Rationality

Cognitivists about Practical Rationality argue that we can explain some of the (apparent) requirements of practical rationality by appealing to the requirements of theoretical rationality. First, they argue that intentions involve beliefs, and, second, they show how the theoretical requirements governing those involved beliefs can explain some of the practical requirements governing those intentions (or they show how these apparently practical requirements are actually theoretical requirements). This paper avoids the ongoing controversy about whether and how intentions involve beliefs and focuses instead on this second part of the Cognitivist approach, where I think Cognitivism faces significant difficulties. I proceed by considering two attempts by Cognitivists to explain (apparent) requirements of practical rationality and I argue that neither of them succeed.     

认知主义与联结主义之比较

认知主义的研究定向和联结主义的研究定向是广义的现代认知心理学的两种主要研究范式。这两种研究范式都各有自己的研究内容和方法论,也取得了各自不同的成就,同时也存在各自不同的问题。通过对两种研究范式的比较,我们可以清楚地认识到二者的相互沟通和融合才是认知心理学未来发展的必然趋势。     

Motivational Cognitivism and the Argument from Direction of Fit

An important argument for the belief-desire thesis is based on the idea that an agent can be motivated to act only if her mental states include one which aims at changing the world, that is, one with a “world-to-mind”, or “telic”, direction of fit. Some cognitivists accept this claim, but argue that some beliefs, notably moral ones, have not only a “mind-to-world”, or “thetic”, direction of fit, but also a telic one. The paper first argues that this cognitivist reply is deficient, for only the “dominant” direction of fit of an attitude is responsible for its character and function. Further, it seems that the dominant direction of fit of an attitude is determined by its psychological mode, and so all beliefs seem to have a dominant thetic direction of fit, and to be motivationally inert. The main part of this paper, however, is devoted to explaining how it is that attitudes, like moral attitudes, can truly have two directions of fit in a way which enables them to be both cognitive and motivational. Reflection on the nature of beliefs suggests that the claim that the dominant direction of fit of an attitude is determined by its psychological mode should be qualified. The reasons beliefs provide draw their authority for the agent – their demanding nature – from the objects represented by these beliefs, and so, it is the beliefs’ content which determine their dominant direction of fit, as far as their role in practical reasoning is concerned. Thus, in the sense relevant to practical reasoning a belief with a normative content does have a dominant telic direction of fit. At the same time, in the sense relevant to its satisfaction conditions a moral belief has a dominant thetic direction of fit, which underlies its classification as a cognitive attitude. Cognitivists, then, can have it both ways.     

Intentions,Intending, and Belief: Noninferential Weak Cognitivism

Cognitivists about intention hold that intending to do something entails believing you will do it. Noncognitivists hold that intentions are conative states with no cognitive component. I argue that both of these claims are true. Intending entails the presence of a belief, even though the intention is not even partly the belief. The result is a form of what Sarah Paul calls noninferential weak cognitivism, a view that, as she notes, has no prominent defenders.     

人工智能中符号奠基问题的几种解决策略

符号奠基问题(Symbol Grounding Problem:SGP)是人工智能中的一个重要理论问题。它由认知科学家哈纳德提出,意指在不借助外部中介的条件下,一个纯粹的符号系统应如何获取意义?目前为止,SGP的解决方案主要有三种:第一,以哈纳德的杂合系统为代表的认知主义方案;第二,以布鲁克斯的物理奠基策略为代表的生成主义方案;第三,以福格特等人为代表的指号学方案。但是,这些解决方案都受到了塔迪欧和弗洛里迪的批评。他们提出了"零语义承诺",即要求这些方案不能借助任何形式的内在主义(innatism)和外在主义(externalism)解决SGP问题。不过,"零语义承诺"本身也存在诸多问题,其合理性仍需进一步阐明。     

Against Strong Cognitivism: An Argument from the Particularity of Love

According to the view we may term “strong cognitivism”, all reasons for action are rooted in normative features that the motivated subject (explicitly or implicitly) takes objects to have independently of her attitudes towards these objects. The main concern of this paper is to argue against strong cognitivism, that is, to establish the view that conative attitudes do provide subjects with reasons for action. The central argument to this effect is a top‐down argument: it proceeds by an analysis of the complex phenomenon of love and derives a conclusion regarding the (motivational and normative) nature of more basic mental phenomena—particular desires. More specifically, its starting point is the crude intuition that the significance conferred by love upon its objects is of a distinctively personal kind—an intuition that is expressed by the apparent non‐substitutability of two similar subjects only one of whom is loved with respect to their importance for the lover. I argue that the initial notion of non‐substitutability can be refined and modified so as to form a real challenge to all versions of strong cognitivism and to establish the existence of attitude‐dependent reasons.     

Moving Beyond Computational Cognitivism: Understanding Intentionality,Intersubjectivity and Ecology of Mind

The emergence of the Cognitive Sciences, in the middle of the 20th Century, was initially based on an abstract model of the mind: the computer metaphor. The human mind was understood in analogy to the digital computer, as a rule-based, symbol processor. As a consequence, the human being was envisioned as logically-rationally guided, radically disembodied and isolated from culture. Over the last few decades, several disciplines, such as Biology, Mathematics, Philosophy, Psychology and Neuroscience, have begun to address the study of intentionality, intersubjectivity and natural cognition. Searching for a better understanding of these complex issues, a number of approaches have been developed with the promise of capturing the specific qualities of human cognition, radically omitted from a computationalist view of mind. Nevertheless, since these research programs are rather recent, concrete methodological designs and empirical approaches in the form of experimentally testable hypotheses are still scarce. This special issue brings together several perspectives in order to propose alternative research approaches in the topics of Intentionality, Intersubjectivity and Ecology of Mind. We believe it is necessary to discuss and advance towards explicit empirical frames in the form of actual experiments, specific predictions and formal models. The essays presented here constitute an attempt to move in this direction, with the specific aim of reconsidering the study of some forgotten properties of brain and mind.