In: First-Person Thought
Maik Niemeck
Search for other papers by Maik Niemeck in
Current site
Google Scholar
Free access

This is a book on first-person thought. These are thoughts which we typically express using the first-person pronoun singular (such as the English “I”) or one of its cognates (such as “me”, “mine” or “myself”).1 At least since Rene Descartes’ (1641) famous axiom cogito ergo sum and Immanuel Kant’s (1781) claim that I think must be able to accompany all my representations, first-person thought has taken center stage in many philosophical debates and has been viewed as the central phenomenon associated with self-consciousness. Nowadays, it is widely held that this is because first-person thought entails a peculiar kind of awareness about oneself (e.g. Smith 2017). In entertaining these types of thoughts, we do not simply happen to think about ourselves; rather, this fact itself is also something we become aware of in virtue of having this kind of thought. In this regard, first-person thought is directed at its thinker, but more importantly it also somehow reflects the identity relation holding between the thinker and the object at which this awareness is directed. Many have claimed that this double reflexive structure is appropriately captured with the characterization that first-person thinkers are aware of themselves as themselves.2

The overall aim of this book is to explore this and other peculiarities of first-person thinking and how they relate to a theory of self-consciousness more generally. There are two interrelated questions which will be the focus of what follows: 1.) Is there anything special about first-person thought? 2.) How is first-person thought related to other types of self-consciousness? I am quite confident that these questions speak to some of the most pressing issues within my area of concern. To answer them, I will draw on insights from a wide range of philosophical fields, such as the epistemology of self-ascriptions, the overall metaphysics of experience, or theories of perception, action and emotion. When appropriate, I will also highlight empirical findings and theories which support my claims. As the title of the book suggests, there are three main topics – namely action, identification, and experience – which will be discussed in order to answer the two central questions. Before going into further detail, I will offer a brief overview of how I approach these topics.

Since the 1960s, there has been almost complete agreement that there must be something unique about first-person thought – and indexical thought in general. In a series of highly influential articles, a few philosophers (Castañeda 1966, 1967; Lewis 1979; Perry 1977, 1979), who would become the most distinguished authors in their fields, presented a variety of thought experiments which convinced the majority of philosophers that indexical thinking is an irreducible element of human cognition that, moreover, calls for a special metaphysical treatment, since it cannot be easily accommodated in our standard theory of singular thought and other propositional attitudes. The cases presented in support of this assumption have typically appealed to the essential role of first-person thinking in the context of action explanation. Until recently this was the common view of things, and no one really questioned the foundations of these assumptions. However, a couple of years ago, some doubts began to rise (Cappelen and Dever 2013; Devitt 2013; Magidor 2015; Millikan 1990) and so-called De Se Skepticism (Torre 2016) was born. De Se Skeptics claim that there is nothing especially interesting about first-person thought and that there is no tight connection between being able to entertain these types of thought and our ability to perform specific actions.

The first two chapters herein are committed to this debate and address the question of whether there is anything special about first-person thought. I will answer this question with a wholehearted “Yes”. The argument that I will provide in support of this position can be divided into two parts. The first is reactionary in the sense that it only tries to rescue the original idea that first-person thought really is essential for a wide range of actions. The second attempt to counter De Se Skepticism takes something of a novel path. I will lay out why first-person thought also involves a special kind of concern for the object thought of and thus an immediate motivation to act on the basis of it. In this, I elaborate on a recent proposal set out by Mark Textor (2018) who claims that, alongside its double reflexive nature, first-person thought contains an evaluative component and is in this respect thick. In addition, I will explain why first-person thought’s essential role for action and the concern accompanying it are both grounded in its double reflexive structure mentioned at the beginning of this book.

The third and fourth chapter elucidate the role that identification plays in first-person thinking. The main reason for this is that quite a few philosophers have forcefully suggested that first-person thought stands in a unique relation to identification and that therefore a theory of it should offer an explanation of this presumed characteristic. Many have argued that certain types of first-person thoughts are special because they exhibit the phenomenon of the so-called Immunity to Error through Misidentification (henceforth also “IEM”). It is not perfectly clear what most people have in mind when they talk about this alleged phenomenon. However, there are mainly two central ideas underlying this concept. One is that of a partial infallibility. The other, let’s call it the idea of Immunity to Misidentification, is that certain thoughts are necessarily gained without any false identification in their epistemic base. Most authors believe that self-ascriptions based on introspective experience possess IEM. The central intuition is that when we experience a pain, for instance, and self-ascribe that state, it could not happen that the self-ascribed state actually belongs to someone else or that we need any criteria – that could turn out to be false – in order to identify the subject having this state. In contrast to those outlined claims, I will argue that IEM should not be regarded as especially relevant for a theory of first-person thought. Moreover, I will maintain that it might not be as interesting as many people think. However, in chapter four, I will focus on a regress problem identified by Sydney Shoemaker (1968) and propose that someone concerned with first-person thought should have to say something about identification after all. Yet, the reason to do so is not that first-person thought stands in a unique relation to identification, but rather because the object of first-person thought is exceptional.

In chapter five, I will examine the relation between first-person thought and self-consciousness more broadly construed. In order to do so, I will take up the issues discussed in chapters one and two, and explore what is needed for first-person thought to include this double reflexive awareness. This question might also be framed in terms of identification: How are thinkers able to identify themselves as the objects of their first-person thoughts? A very common answer to this question is that there must be prior non-conceptual forms of self-consciousness which enable us to perform this kind of task. In order to justify this claim, many people appeal to what I call the argument for non-conceptual self-consciousness based on the meaning of “I” which stresses the fact that the first-person concept requires contextual information to determine the referent of an occurrence of it. And I agree that the argument serves to warrant that first-person thought requires some kind of prior self-presentation, i.e. in those cases the awareness of being the thinker. But in my view, the argument by itself does not justify the claim that there is any self-consciousness required for that. Nevertheless, I believe that there is some hope for the argument to work. That is why I will make some general remarks about the cognitive function of consciousness itself. Of course, these considerations will not be comprehensive and conclusive, but they will nevertheless provide some evidence for the claim that non-conceptual conscious contextual information is actually required for genuine first-person thought.

The last main chapter discusses the metaphysics of these forms of self-consciousness. If we assume that there is something like a non-conceptual conscious awareness of ourselves – as most contemporary researchers believe – and not merely some non-experiential self-presentation, then it seems that this object of experience has to be quite extraordinary from a phenomenological perspective. I will borrow Uriah Kriegel’s (2009) notion of a subjective character of experience3 to denote such self-consciousness which involves a feeling of what it is like to be in it (Nagel 1974). In the more recent literature, there are three main approaches to this component of experience: The Self-representationalist, the Pre-reflective, and the Mode of Presentation account of subjective character. I will argue that a theory which understands the subjective character as grounded in the modality of experience, and more precisely in a mental relation, is superior to the two other proposals, because it can provide the right sort of compromise between representationalism and anti-representationalism. I will demonstrate that the most promising version of the mode of presentation account also presupposes a specific form of representationalism, namely one which has been recently called impure intentionalism (Crane 2009, 2013), impure representationalism (Chalmers 2010) or intermodal intentionalism (Speaks 2015). I will show that there are good reasons to accept impure intentionalism and claim that once we accept it, we have the right conceptual tools to adequately understand the subjective character of experience.


After all, that is why we call them “first-person thoughts”. It is very often presupposed (also by myself) that there is a tight connection between how we speak, think, and are aware of ourselves. Yet, this is not to say that first-person thought or even any kind of self-awareness can only occur because there is a specific language which has these types of expressions or that these kinds of thoughts are especially tied to a particular set of languages. We can surely express the same type of thought in different languages and these thoughts could still exist without there being any spoken or written language. But I guess classifying thought types in virtue of the sentences they dispose thinkers to utter probably remains the best way currently available of doing this.


Later on, we will get a clearer idea of the motivations which lead philosophers to believe that first-person thought has this double reflexive structure.


The reason for this is that I think that this name is quite useful in denoting some subjective part of the phenomenal character of experience without suggesting any further metaphysical constraints, such as being representational, being non-propositional, being non-conceptual or being non-relational, etc.

  • Collapse
  • Expand