§ 0.

What does it take to understand oneself? I argue that it involves grasping coherence-making relations between constituents of a body of information about oneself. As such, self-understanding is best characterized as a kind of holistic understanding of a subject domain, often called objectual understanding. What is special about self-understanding is the domain — oneself. This characterization enables us to explain what it means to deepen one’s self-understanding and why self-understanding can neither be reduced to self-knowledge nor self-narrative.

The paper is divided into three parts.

§ 1.

I start by analyzing which of the kinds of understanding commonly discussed in contemporary epistemology fits best for an account of self-understanding. After rejecting the views that self-understanding is a kind of understanding-that, understanding-why, or understanding-how, I argue that we should characterize self-understanding as a kind of objectual understanding (for influential accounts of objectual understanding see, e.g., Kvanvig, 2003; Riggs, 2003, 2009; Grimm, 2011; Elgin, 2017). To wit, self-understanding is more like understanding marine biology than understanding why humpback whales migrate annually or how to stimulate coral reef restoration.

Given that objectual understanding is commonly characterized as consisting of grasping relations between constituents of a body of information, characterizing objectual understanding of any subject matter — oneself not being an exception — requires answering two questions: “What do we mean by grasping?” and “What is the relevant information?”

To answer the first question, I follow many contemporary epistemologists in claiming that grasping is a type of ability or a set of abilities (e.g., Wilkinfeld, 2013; Hills, 2016; Elgin, 2017). On my account, understanding oneself essentially involves possession of an ability (or a set of abilities) to represent and manipulate a body of information about oneself.

But what is the relevant information? Here, I appeal to the psychological notion of “self-concept,” understood as “a mental representation…what we bring to mind when we think about ourselves” (Neisser, 1997, p. 3), and argue that self-understanding is grasping coherence-making relations between constituents of one’s self-concept.

§ 2.

The view outlined above allows us to offer a fine-grained account of what it means to deepen one’s self-understanding and explain why self-understanding is not reducible to self-knowledge.

Almost all we do, from simply living our lives and letting things happen to us, through reading novels and engaging with art, to interacting with other people including our friends and therapists, may lead to deepening our self-understanding. I argue that it happens along three main dimensions. We can increase: (1) the richness of our self-understanding by increasing the amount and scope of information we possess about ourselves; (2) factivity by increasing the truths-to-falsehood ratio in the body of information, and (3) coherence by recognizing more connections between the pieces of information we have about ourselves.

Bits of self-knowledge contribute to the body of information whose grasping constitutes self-understanding. Therefore, knowing more about oneself typically results in deeper (richer and more factive) self-understanding. At the same time, self-understanding has important features that distinguish it from knowledge. It is compatible with having at least some false information about oneself and essentially consists of an ability to grasp coherence making relations between relevant pieces of information. Moreover, we typically do not consider knowledge to be gradable.

Given that non-factivity (e.g., Riggs, 2009; Elgin, 2017), gradeability (e.g., Zagzebski, 2001; Kvanvig, 2003; Riggs, 2003; Pritchard, 2009), and the essential role of grasping (Kvanvig, 2003; Riggs, 2003; Grimm, 2014; Elgin, 2017) are usually enumerated as what makes objectual understanding irreducible to knowledge, I conclude that self-understanding is irreducible to self-knowledge.

§ 3.

In the last part of the paper, I argue that while building a self-narrative is a way of developing self-understanding, it is not the only way.

Narratives afford (some level of) understanding of their protagonists by establishing coherence-making relations between their life events; the protagonist of a self-narrative is oneself; thus, a self-narrative can afford (some level of) self-understanding.

Nevertheless, as argued, by Strawson (2004), many people think about themselves in a predominantly episodic fashion, in which case “one does not figure oneself, considered as a self, as something that was there in the (further) past and will be there in the (further) future” (2004, p. 430). The view I defend enables us to explain why both predominantly episodic and predominantly diachronic individuals can attain a certain level of self-understanding and further deepen it. To understand oneself is to grasp coherence making relations between constituents of a body of information about oneself. Predominantly diachronic individuals tend to do it through self-narrative; while predominantly episodic individuals are less inclined to use this particular tool.

In short, there are more ways of grasping relations between elements of self-concept than just by putting them into a narrative. Thus, not all self-understanding is narrative self-understanding.

References

Elgin, C. Z. (2017). True Enough. Cambridge, MA: MIT Press.

Grimm, S. (2011). Understanding. In D. Pritchard & S. Bernecker (Eds.), The Routledge Companion to Epistemology. Routledge.

Grimm, S. (2014). Understanding as knowledge of causes. In Virtue epistemology naturalized (pp. 329–345). Springer.

Hills, A. (2016). Understanding Why. Noûs, 50(4), 661–688.

Kvanvig, J. (2003). The Value of Knowledge and the Pursuit of Understanding. Cambridge University Press.

Neisser, U. (1997). Concepts and self-concepts. In U. Neisser & D. A. Jopling (Eds.), The conceptual self in context: Culture, experience, self-understanding (pp. 3–12). Cambridge University Press.

Pritchard, D. (2009). Knowledge, Understanding and Epistemic Value. Royal Institute of Philosophy Supplement, 64, 19–43

Riggs, W. D. (2003). Understanding “Virtue” and the Virtue of Understanding. In M. DePaul & L. Zagzebski (Eds.), Intellectual Virtue:34 Perspectives From Ethics and Epistemology (pp. 203–226). Oxford: Oxford University Press.

Riggs, W. D. (2009). Understanding, Knowledge, and the Meno Requirement. In A. Haddock, A. Millar, & D. Pritchard (Eds.), Epistemic Value. Oxford: Oxford University Press.

Strawson, G. (2004). Against Narrativity. Ratio, 17(4), 428–452. https://doi.org/10.1111/j.1467-9329.2004.00264.x

Wilkenfeld, D. A. (2013). Understanding as representation manipulability. Synthese, 190(6), 997–1016. https://doi.org/10.1007/s11229-011-0055-x

Zagzebski, L. (2001). Recovering Understanding. In M. Steup (Ed.), Knowledge, Truth, and Duty: Essays on Epistemic Justification, Responsibility, and Virtue. Oxford University Press.

Outline. In this talk, I do three things. First, I examine a kind of objection to the account of the attitudes given in Lewis 1979 which criticizes the notion of self-ascription as a putative unexplained primitive of the theory. I argue that it’s not so much primitiveness per se that is objectionable, but diagnose two consequences of this primitiveness which are objectionable. (Cf. Stanley 2011; Capellen and Dever 2013; Holton 2015) Second, I consider recent defenses of Lewis’s theory (due to Jackson and Stoljar 2020; Openshaw 2020) which contend that self-ascription is not primitive and offer explanatory strategies. In light of my diagnosis, I argue that the more important question is not whether self-ascription is a primitive, but whether these strategies can meet the objections I diagnose. My take on this question is salutary—these strategies can defuse much of the objections. Third, I argue that there is still a lingering problem regarding self-ascription which these explanatory strategies bring into relief.

Lewis’s property theory. In Lewis 1979, Lewis advocates a novel theory of the attitudes motivated in part by the fact that the more traditional, propositionalist, possible worlds theory of the attitudes he previously endorsed was unable to represent the kinds of de se attitudes discussed by Castaneda (1966), Perry (1979), etc., nor to distinguish them in a satisfactory way from their de dicto counterparts. Lewis’s proposal was novel in two ways. First, it allowed attitudes to be relations to properties instead of propositions, construed as sets of centered worlds (formally, world-center pairs, understood as world-bound time-slices of individuals). Second, it construed the relation a subject has to content in terms of self-ascription. Taken jointly, this allows Lewis’s theory to capture both de se and de dicto attitudes in a general theory of the attitudes. Lewis also leveraged this framework for an account of de re attitudes, which I will not address directly in this talk.

First: primitive self-ascription—what’s the objection? While Lewis’s theory has proved tremen- dously influential, and the centered worlds framework based on it very productive, a number of objections to Lewis’s theory have emerged in prominent venues in the last decade. These objections are motivated by a common complaint—that the self-ascription relation, which is fundamental to Lewis’s theory and provides the linch-pin for the uniform account, is an unexplained primitive. This primitiveness is unmotivated and leaves the whole theory on shaky footing. In examining this criticism, I find that it is not the apparent primitivism that is problematic, but some consequences that flow from it. We do better, I argue, to understand the primitiveness complaint in terms of two kinds of objections. The first (of which Stanley 2011 is a representative) contends that Lewis’s account evades a description of what cognitive contact to the self amounts to. The second (which I associate with Capellen and Dever 2013) has to do with a kind of arbitrariness at the heart of self-ascription. This objection contends that there is no basis for the central technical innovation of Lewis’s account.

Second: do the explanations defuse the objections? Recent defenses of Lewis (e.g., by Jackson and Stoljar (2020); Openshaw (2020)) have appeared, objecting to the characterization of self- ascription as a primitive. Their main aim is to show how self-ascription is explained by other resources in Lewis’s work. The explanatory strategy in Jackson and Stoljar 2020 appeals to Lewis’s functionalist commitments. The strategy in Openshaw 2020, to some of Lewis’s foundational remarks on semantics, drawing on the concept of initialization. (The term is not Lewis’s but can be given a Lewisian gloss.) I argue that, in light of the diagnosis of the primitiveness complaint above, simply pointing out that self-ascription is not primitive is insufficient. We need to see how the explanation for self-ascription can deal with these deeper objections. Though the proposals on offer are different, they are actually complementary. Jackson and Stoljar’s functionalist strategy can help deal with the evasion objection, and Openshaw’s initialization strategy helps deal with the arbitrariness objection. This is a welcome result for the friend of the property theory.

Third: a remaining wrinkle for self-ascription I then argue a final point. Notwithstanding the promising explanatory strategies, there is a remaining wrinkle when it comes to the objections to self-ascription. The initialization strategy provides a pretty promising response to the arbitrariness objection, but it makes an empirical prediction about the nature of the attitudes and attitude content. It commits one to what Recanati (2007) calls the “reflexive constraint”. (Cf. also Pagin 2016 for a similar point.) In terms of the initialization strategy, it is the commitment that the initializing indices, (Lewis’s “original indices”; Lewis 1983) are those of the context of utterance/ thought. Appeal to the semantics of natural language can make the reflexive constraint seem plausible, but the move is questionable when applied to the attitudes. Recanati gives grounds for questioning it unrelated to the dispute about self-ascription. I offer some additional reasons: there are non-doxastic attitudes that have contents not plausibly described being bound by the reflexive constraint. I conclude by reflecting on where this leaves the explanation of self-ascription vis-a`-vis the aforementioned objections.

References

Capellen, H. and J. Dever (2013). The Inessential Indexical. OUP.

Castaneda, H. (1966). ‘He’: A study in the logic of self-consciousness. Ratio 8, 130–157.

Holton, R. (2015). Primitive self-ascription: Lewis on the de se. In B. Loewer and J. Schaffer (Eds.), The Blackwell Companion to David Lewis. Blackwell.

Jackson, F. and D. Stoljar (2020). Understanding self-ascription. Mind and Language 35(2), 141–155.

Lewis, D. (1979). Attitudes de dicto and de se. Philosophical Review 88(4), 513–543.

Lewis, D. (1983). Postscripts to General Semantics. In Philosophical Papers, Volume 1, pp. 230–232. OUP.

Openshaw, J. (2020). Self-ascription and the de se. Synthese 197(5), 2039–2050.

Pagin, P. (2016). De se communication : Centered or uncentered? In M. Garc ́ıa-Carpintero and

S. Torre (Eds.), About Oneself: De Se Thought and Communication. OUP.

Perry, J. (1979). The problem of the essential indexical. Nouˆs 13(December), 3–21.

Recanati, F. (2007). Perspectival Thought. OUP.

Stanley, J. (2011). Know how. OUP.