A Heideggerian approach to non-indicative moods

Richard Polt

Panel: Heidegger and Grammar

Heidegger Circle, Boston University, 2023

I’d like to make a few remarks today on grammatical moods from a Heideggerian point of view. What do moods such as the subjunctive and optative disclose? And has the indicative mood been unjustly privileged over these so-called “irrealis” moods?

Heidegger gives us a provocative opening to the topic in a passage from the Black Notebooks of the early 1930s (GA 94, 51):

The full essence of being should be interrogated—presence (the “is”) is positively absorbed into it, and at the same the predominance [of presence] is beaten back into its limits.

Being must develop its horizon again, and now in full (time) ...

Not the “it is” [es ist], but the “so be it” [es sei] (thrown projection)—the “so be it” of original reticence.

The verb form sei is either imperative or subjunctive. The idea in this note seems to be that Dasein’s projection of possibilities is more primordial than any experience of what actually is, or any assertion about it. Traditional metaphysics, which takes being as presence, or more specifically as presence-at-hand, reality, or actuality, has elevated the “is”—the third-person singular present indicative—beyond its proper limits.

Now, we might try to turn this point against Heidegger and ask him, “Aren’t you making an assertion about how Dasein is?” But his note is not actually an assertion; it’s not in the indicative mood. It’s a proposal, a project. By 1936, in his “Running Notes on Being and Time” in GA 82, he has radicalized this es sei approach to the point where he takes the very concept of Dasein as a possibility he’s projecting for human beings, not a description of how they are.

But we don’t have to take such a radical stance to see that there’s an interesting grammatico-ontological issue here. As early as his 1915 habilitation, Heidegger writes that “sentences that command, wish, question, and doubt have not been sufficiently clarified and distinguished from each other” (GA 1: 385).1 He adds that we should investigate “the indicative, optative, and imperative moods [and] the act qualities of wishing and commanding” (GA 1: 395; trans. 152-53).

There’s a related passage in the 1925-26 Logic course, where Heidegger comments on “the rich multiplicity of ways of discourse” other than assertion, such as “wishing, commanding, and questioning.” Are these “non-objectifying acts” capable of truth and falsehood? He implies that the answer is yes, if we take truth more broadly than the mere correctness of a statement. He notes that “the optative, the imperative, and so on cannot be grasped conceptually in distinction from the indicative until this question is cleared up” (GA 21, 130-31).

But the analysis of grammatical moods is one of many issues that, as far as I know, Heidegger leaves to us to work out. He has bigger fish to fry. How would we follow up on his suggestion?

First, we could look into what linguistics has to say about the issue. Many linguists group all non-indicative moods in the category “irrealis.” This category covers twenty-some verbal forms in many languages, including exotic constructions such as the dubitative, apprehensive, presumptive, and benedictive.

Now, some linguists have criticized the term “irrealis” for its vagueness and its inconsistent manifestation across the languages of the world, while others have tried to define it consistently. So is it a valid category? If so, what exactly does it designate?

Roughly speaking, irrealis moods are used in speech acts that differ from simple statements of fact about what seems to be directly given. They may express uncertainty, speculation, desire, or aversion. They may communicate that the speaker has no firsthand experience of the matter under discussion. They may also express that an act is permitted, forbidden, or possible.

In Spanish, for instance, the subjunctive is routinely used in situations where there is even a slight deviation from a direct report on what is the case. For Spanish speakers, the subjunctive forms emerge unbidden in response to the nuances of a situation. A professor of Spanish recently confirmed for me that this is one of the hardest features of the language for non-native speakers to learn.

In general, we can say that while the indicative mood offers, or purports to offer, a disinterested report on what is the case, irrealis moods may add an awareness of the position and attitude of the people involved, and often refer to events or situations that are possible or impossible alternatives to what is actually given.

But attempts to define the concept more precisely remain controversial in linguistics—and I would argue that the science of linguistics as such is unable to resolve the controversy. Linguists discussing the irrealis often appeal to fundamental ontological concepts such as reality, actuality, fact, time, world, presence, possibility, and contingency. Their debate raises ontological questions about language and Dasein. So in a sense, linguistics depends on philosophy. Of course, philosophy as such isn’t dedicated to providing empirical data on languages, but the concepts used by linguists and grammarians build on traditional philosophical thought and may call for fresh philosophical reflections.

We could begin those reflections by thinking about the relation between language and being. By “being” here I simply mean how beings are—that is, existing entities’ distinctive ways of existing, of being something rather than nothing.

One extreme position would hold that language simply mirrors the features of being as they are given. Linguistic structures can then be taken as guides to ontological ones. For example, Aristotle relies on “what is said of” an underlying thing as a main clue to the kinds of being (Categories I, 2-5). The “saying” here is a predicating that typically employs the copula esti or “is”: one can say that Fido (a “primary substance”) is a dog (“secondary substance”), is the father of Rover (relation), is furry (quality), is two feet tall (quantity), and so on. But not all languages have a word that functions like einai or “be”; the common use of einai in Greek may involve systematic distortions of the phenomena; and the choice of esti, the third-person singular present indicative of einai, may also involve a certain prejudice. What would we learn by beginning with the second person singular (with Martin Buber), or with the optative mood (with Ernst Bloch and his principle of hope)?

Another extreme position would claim that, far from mirroring being, language imposes itself on experience and locks us into certain conceptions of being. This is a strong form of the Sapir-Whorf hypothesis, named after the work of linguists Edward Sapir and Benjamin Lee Whorf in the 1920s to 1940s. In its most exaggerated versions, this view becomes a defeatism that rules out any translation, or any attempt to describe the language of one community in the language of another. The extreme hypothesis even seems to invalidate itself, inasmuch as it pretends to make a trans-linguistic claim.

If we look for a balanced approach to the relation between being and language, we can benefit from the phenomenological point of view, which emphasizes the correlation between the meaningful act and its object. For every act, including linguistic signifying, a phenomenon is displayed. This point of view suggests a third way between the extreme positions we considered: there is a productive loop between experience and language. Language highlights the situation as it appears; it emphasizes, sharpens, and even influences this appearing, but it does not one-sidedly create the appearance. Every situation exceeds what language expresses, inviting ever more language without being exhausted by it; the productive loop can continue indefinitely. Words are correlated to situations that emerge as significant; but words also originate new significations, reordering our priorities. Mattering makes chattering, which in turn makes new mattering.

Different languages, then, recognize and highlight different aspects of being; some languages are better attuned than others to some dimensions of experience. For instance, the Californian language Central Pomo uses suffixes to mark whether an assertion is supported by vision, hearing, hearsay, or inference. Many a teacher of introductory philosophy classes might wish that English had such a system. It’s certainly not impossible to express these nuances in English, but our language doesn’t constantly prompt its speakers to attend to them. For similar reasons, foreign words may be adopted into English because we have no equally clear and compact equivalent; Schadenfreude is a well-known example. But this is not to say that it is untranslatable; translating, in both broader and narrower senses, happens constantly as experience and meaning adjust and readjust to each other.

I think this position is compatible with both early and late Heidegger, as long as we take his statements on language in a nuanced way. As he puts it in Being and Time, “to significations, words accrue” (SZ 161): language adjusts to the needs of our experience of what is. But it’s also true that, as he says in “Letter on ‘Humanism,’” “language is the house of being”: significations are preserved in words, and experience is, to some degree, shaped by linguistic tradition. Language brings with it—to use a Husserlian term—the “sedimented” experience of millennia.

Of course, it follows that our language, our linguistics, and our philosophical interpretation of language can be challenged in the name of a fresh experience of being. In texts such as Introduction to Metaphysics, Heidegger emphasizes that traditional Western grammar depends on the restricted ancient understanding of being. For the Greeks, to be paradigmatically means to be fully actual and present; other phenomena, such as becoming and potentiality, are typically seen as inferior and incomplete, or even (for the Eleatics) wholly nonexistent.

If the founding philosophical inquiries into being have been prejudiced in favor of present things, then linguistics would tend to privilege the thing (res) and what pertains to it (realis), defining all else as a negation of it (irrealis). Anything other than a description of the “actual” would be unjustly characterized as a derivative and perhaps inessential mode of speech.

Accordingly, the indicative mood is privileged in Western metaphysics—along with an attunement oriented to being as presence in the horizon of the “now,” especially the mode of presence Heidegger calls presence-at-hand or reality. This mode tends to be expressed in the present indicative, or else in the infinitive, which appears to stand eternally in its purity, free of accretions such as tense and mood. Likewise, the tradition tends to view truth as contained in indicative assertions.

If we accept this traditional conception of being and truth, it is natural to see the irreal as subordinate to the real, and the language of the irreal as a sort of frill, a luxury, a parasite on talk of the real. Could we replace the optative, for instance, with indicative statements of the fact that a desire is present? Can a contrary-to-fact conditional (“If I were an eagle, I could fly”) be replaced by a statement of fact (“All eagles fly”)? Perhaps all questions can be replaced by statements. But if the irreal has its own form of being, irreducible to the real, we may not be able to eliminate irrealis moods. We may even have to give the irreal a certain priority over the real.

I’m sure you see where I’m going: the irreal—the es sei—does have priority, at least from a Heideggerian point of view, and at least for Dasein. Our understanding and our existence necessarily project beyond the indicative into the possible. According to Being and Time, “possibility … is the most primordial and ultimate positive ontological characteristic of Dasein” (SZ 143-44). Our experience of the “it is” is made possible by the realm of the “so be it.” We can ascertain facts only because we care about what we can and cannot do—a range of possibilities rooted in fundamental experiences of ability and disability, enabling and disabling. Life is not simply about the is, but also about the might, could, and would. Our world is a network of possibilities, and being-in-the-world requires the ability to “project” various possible ways of operating in a significant context (SZ 145). Ultimately, we have to pursue a possible way to exist—an interpretation of who we are—in the face of the constant possibility of our own nonbeing. That possibility is far from a matter of indifference to us, even though it is supremely irreal.

In short, human beings care about the irreal, and its importance calls for linguistic expression. Spanish slips into the subjunctive, English uses modal verbs, and Central Pomo marks certain verb forms as hearsay, because the lives of their speakers demand these usages. Words come to our aid, or “accrue” to significations, when important phenomena are calling for articulation. Irreal beings need the language of the irreal.

Non-linguistic interpretation also responds to the irreal as a legitimate and important domain. Without a sense of possibilities that can be adopted or foregone, we could hardly understand ourselves as choosing. Without a sense of what we might have done but did not, we can’t take responsibility for our own past acts. All of these aspects of practical life elicit moods—not just in the grammatical, but in the emotional sense—that are part of our interpretive activity. There is hardly any mood that doesn’t extend past the real and respond to the possible or the counterfactual.

I’ll end by recalling that according to Being and Time, it’s not only Dasein that’s more than a present-at-hand entity to be addressed in “apophantic,” indicative assertions (SZ 158). We understand all beings in terms of their possibilities (SZ 144-45). Presence-at-hand is a reduced, impoverished mode of being. Can any entity’s way of being be reduced to presence-at-hand? Don’t all beings call for language that exceeds the indicative?

I have to leave it there. If you’d like to see my further thoughts on this topic, you can find them in my recent article in the Journal of Continental Philosophy titled “The Language of the Irreal.” Now I look forward to our discussion.


1. Duns Scotus’s Doctrine of Categories and Meaning, trans. Joydeep Bagchee and Jeffrey D. Gower (Bloomington: Indiana University Press, 2022), 147, translation modified.

Richard Polt - A Heideggerian approach to non-indicative moods
Original PDF version.