“Sketches of landscapes”: On philosophical poems

Is it possible to do philosophy by writing lyric poems—or reading them? How can a poem be genuinely philosophical, and can a philosophical poem do something that straightforward philosophical writing cannot do? Some have suggested that poems and philosophical writings have different aims and are subject to different and conflicting demands, which would render it difficult if not impossible to write a successful philosophical poem. I suggest that while this is true with respect to the aims of the standard academic philosophical journal essay, there is a different way of doing philosophical work—one that pays close attention to actual thought processes and that dramatizes the interplay of ideas—that lyric poetry is quite well suited to take on. Such work may be significant not only in helping us better understand human consciousness, but in letting us grasp insights and aspects of our experience of the world which the philosophical demand for simple, unified theories might otherwise tempt us to minimize or ignore altogether. Subjects: Arts; Humanities; Language & Literature


PUBLIC INTEREST STATEMENT
This paper considers whether lyric poems can be viewed as a kind of philosophical writing. Some have suggested that the goals, techniques, and standards of assessment for philosophical writing are deeply different from those that apply to poems, so that the very idea of a philosophical poem is problematic if not self-contradictory. This may be true if by "philosophical writing" we mean the sort of writing that features in the sort of argumentative essay standardly published in academic philosophical journals. However, we should not overlook the possibility that poems can do philosophical work in a different sense, both by providing detailed accounts of human experiences and thought processes, and by dramatizing ideas in a way that acknowledges and helps us explore a world whose inherent complexity sometimes seems to resist being captured in the sort of linear, maximally coherent theories philosophical theories often favor.
Start with the obvious: Philosophical journal articles are written in prose, which looks very different from (most) poetry on the page, and reads differently as well. Most lyric poetry, after all, is lineated: the lines are broken off before they reach the right-hand margin of the page. (Some poems go further, scattering phrases and even individual words about the page with abandon, sometimes isolating them in a field of blank white space.) Moreover, great liberties are sometimes taken in poems with capitalization, punctuation, indeed with all manner of things. For these reasons, it is more often than not possible to tell, even at a glance, whether one is dealing with a poem or with an academic philosophical essay. And if we go past the glance, and actually begin to read, we will find further differences. Poems may include narrative elements, and will often mention and describe particulars, which may be either real or fictional. A philosophical article, for the most part, will not include narratives or mention particulars unless they are examples used to illustrate a general thesis. There are, of course, exceptions. But for the most part, these generalizations hold.
Other differences between poetic and philosophical writing go deeper. One obvious but nevertheless significant difference is that poems are literary texts, and in such texts the language is supposed to be an object of interest in itself. One is meant to notice the word choices, the sounds, even perhaps the shapes of the words. The formal properties of the language are thought to be relevant both to the function of the text (which is, among other things, to provide a certain sort of sensuous pleasure in the reading experience) and to the meaning (the repeated use of certain sounds, for instance, may signal a mood or attitude or suggest a state of mind that is not stated or otherwise alluded to in the poem's language as straightforwardly interpreted). Prosaic language, on the other hand, is subservient to its subject matter: such language strives to be a transparent method of communicating ideas or information. 1 We might take, as a paradigmatic example of prosaic language, a recipe or set of directions or instructions. Bits of prosaic language are not thought of as having style-or, if they do, the style is taken to be entirely irrelevant: any set of words that allows you to reach your destination, to assemble the IKEA bookshelf, or to bake a delicious cake, has done the intended job. A given recipe or set of directions can be completely and without loss paraphrased or translated into another language; assuming you speak both languages, the English language version and French language version of a recipe are functionally equivalent and therefore the same in every respect that matters. Poetry, on the other hand, has been said, with suggestive overstatement, to be precisely what is lost in translation. 2 Indeed-I will return to this point soon-there is an important sense in which, in poems, the way something is said can be said to affect what is being said; meaning, then is not independent of style.
Another difference we might note is that poetry, unlike philosophy, does not ordinarily aim to convince. The author of a philosophy article wants, presumably, that the reader will come, through reading, to agree with her, to share her view. But what is it, then, that the author of a poem wants? She, too, may want to share something; but what is to be shared here is less a proposition or a doctrine than an experience, a sequence of mental events, a set of images, and the resonances that occur between these particular images. Or in some cases it might be a perspective, a way of seeing: if the poem is successful, the reader will come to see its subject in the way the poet sees or saw it. 3 At any rate, it does not seem generally correct to describe whatever sort of sharing goes on, or is meant to go on, when a work of art is offered and experienced in terms of agreement.
It seems safe to say that because they are to be evaluated in different terms, what counts as a flaw in a poem need not count as a flaw in philosophical prose. Ambiguity, for instance, which is nearly if not always a flaw in a prosaic text that is striving to deliver information in as clear a matter as possible, very often does not count as a flaw in a poem or other literary text. (If the aim of a philosophical article is that the reader comes to share the author's view of things, it is necessary that the article be as clear as possible about what the author's view of things is.) Similarly, asserting falsities is presumably always a flaw in philosophy, whereas the appearance of false statements in a poem is not typically viewed in that way, even if the poem seems to be structured as a series of assertions. Conversely, those things that are commonly seen as flaws in a poem-the use of clichés, the failure to provide musical pleasure or to paint a vivid or memorable image, and so forth-are not, in general, the sort of things that lead us to think less of a philosophy paper. At any rate, these are not considered failures in this context: an editor of a philosophy journal would not reject a paper because it did not achieve these things, and I would guess that a great many authors of academic philosophical arguments put little if any effort into achieving them.
The idea that poetic language and philosophical prose aim at different ends, and that different things therefore count as flaws in each case, seems to be at the heart of Jarrell's objection to the idea of philosophical poetry: The habit of philosophizing in poetry-or of seeming to philosophize, of using a philosophical tone, images, constructions, or having quasi-philosophical daydreams-has been unfortunate … Poetry is a bad medium for philosophy. Everything in the philosophical poem has to satisfy irreconcilable demands: for instance, the last demand that we should make of philosophy (that it be interesting) is the first we make of a poem. (Jarrell, 1999, p. 116) Perhaps poets will be more likely than philosophers to accede to this. There are, perhaps, too many philosophy papers dedicated to dull, trivial subjects; but it isn't hard to find poems on dull subjects either, and many of those poems fail (if they even try) to make their subjects interesting. Still, when the work in question is a poem we can presumably say that the poem, if it fails to be interesting, has failed; at the very least, that a poem bores its readers automatically counts as an esthetic flaw. That a philosophical essay is boring, on the other hand, might entitle us to complain about having had to read it, but it is-or so it might be suggested-no mark against its quality as a piece of philosophy.
But is this entirely true? If the point of philosophy is to help us come to a better understanding of the world, and so, one would presume, to say enlightening things about things that in some way matter, then we might wonder if any successful work of philosophy could fail to be interesting; for surely, unless we are simply not interested in improving our understanding, any enlightening comment on any matter that isn't entirely trivial ought to interest us. There is surely something to this thought, but it would be a mistake to make too much of it. After all, it is not only possible but important to distinguish the question of whether the conclusion of a philosophical article is interesting from the question of whether reading the paper is an interesting and rewarding experience. It might perhaps be that any philosophical claim will be of at least some interest, at least on the assumption that it is true (or likely to be true?); yet the account of the arguments that establish that conclusion might in some cases nevertheless be as dry as dust. And unfortunately, to know whether the conclusion is true, or is likely to be true, it is sometimes necessary to read and examine the dry-as-dust arguments. (To be clear, I am not suggesting that such arguments are always uninteresting in their own right, only that they may be.) One might say that philosophy aims at truth in a sense that poetry does not; or rather, that the aim of truth-telling restricts what is permissible and appropriate in a philosophical essay in a certain manner which does not seem to occur with respect to poetic writing. Suppose that in the course of writing, an author reaches a juncture where she is torn between two options for how she might go on. The demands of logic and argument require her to take the first option; but the second possibility is, in and of itself, considerably more interesting, and would make a more entertaining reading experience. If she is writing a philosophy paper then she will have to say the first thing, and forgo the opportunity to say the more interesting thing. As a philosopher, with a certain sort of responsibility toward the truth, she may only present arguments that she believes to be legitimate (even if other arguments would have been more interesting in their own right or would have allowed her to present a more interesting conclusion), she should avoid distracting extraneous material (even where such material might be entertaining in its own right), and she may only present conclusions that she believes to be established, or at the very least strongly supported, by the best arguments she knows. Philosophy, in this way, bears a certain resemblance to the empirical sciences: an experimenter might hope that her experiment yields unexpected, and therefore interesting, results, but is bound to report what actually happens whether it is interesting or not. She might hope that the results will be interesting, but she is not permitted to do anything but hope that this turns out to be the case.
The poet, by contrast, would seem to labor under no such constraint; at every decision point he can opt for the most surprising, entertaining, or otherwise interesting option, whether or not it corresponds to reality or indeed has any substantial relationship to reality at all. A poet may choose a word or phrase because of its music ("it just sounds right") or its appearance on the page ("it just looks right"), or include an image that is justified by nothing more than a gut feeling ("it just feels right"). Of course if most of its words or phrases are chosen for only one of those reasons, the poem is almost certainly less than fully successful; the ideal poem, we might say, is one in which each word is doing everything that a word can do in a poem: it sounds right, it looks right, it feels right, it contributes to a larger grammatical unit that says something interesting and perhaps even profound, and so forth. Relative to the stringent demands of producing a poem that is fully successful in this way, the poet is perhaps no more free than the philosopher. But ideal poems are rarely if ever achieved; and the point for the moment is simply to notice that, whether or not one sort of writer has more freedom than the other, the forms of freedom available to the poet are quite distinct from those that apply to the writer of philosophical prose.
If this is true then it seems a strong case for agreeing with Jarrell that "poetry is a bad medium for philosophy." A philosophical poem would, indeed, be subject to conflicting demands: to the extent that it was philosophically responsible, a poem would have to constantly choose not to be as surprising, interesting, or entertaining as it might be. It would also often have to choose not to give the kind of sensuous literary pleasure readers expect from poetry, since it would often be the case that the words that most clearly express the correct idea in the proper place would not be the words that fit the structure of the poem. Some poems-sonnets, sestinas, villanelles-wear their form on their face; but every poem has a structure of some sort, and there are words that will fit that structure in a particular spot, and words that won't. One might occasionally find that one and the same word or phrase is called for by both sets of criteria. But one cannot reasonably expect to experience this sort of luck very often, let alone consistently.
However, we should be careful not to conclude too much, or to understate the constraints that might bind the poet during the composition of some particular poem. Writing poetry is not a freefor-all, an "anything goes" game. If a poem has a certain form, for instance-if it is a sonnet, a sestina, or a villanelle-then it will be subject to certain formal constraints regarding number and length of lines, rhyme scheme, rhythmic patterns, and so forth. More interesting for our purposes, however, are constraints involved in the poet's attempt to get something right, to be true to something. Consider, for instance, a narrative poem based on a certain historical event. Such a poem may take a certain degree of liberty with the facts, but this liberty will be subject to serious limitations; there is a fairly clear sense in which a poet who sets out to provide a narrative account of a real event and who ends up offering a clearly inaccurate or misleading account of that event-one that would be expected to lead readers away from rather than toward a better understanding of the event in question-has failed. One might, of course, suggest that non-fictional narrative poetry constitutes a special case. But we should not overemphasize the degree of specialness. The most common subject matter for contemporary poetry to take on is, it seems safe to say, human experience; and it seems to me that one of the criteria for success that must apply to any poem (or other literary work or artwork) that takes human experience as its subject must be a criterion of accuracy: such a poem must be true to human experience, it must get human experience right.
These remarks perhaps suggest that poetry, at least a great deal of the time, might well be characterized in precisely the same way in which we are apt to characterize philosophy: as an attempt to get at, display, or articulate the truth. It is in this connection that I want to cite a comment made by the poet A.R. Ammons. The editor of Set in Motion, a collection of Ammons's prose writings, reports the following: When I suggested that we collect and publish a selection of his prose, Archie (Ammons) was not in favor of the project. The proliferation of prose by poets only seemed to sharpen what he saw as a turning away from the honest complexity of poetic action toward the reductive simplicity of proposition. (Ammons, 1997, p. ix) "Proposition" here is not explicitly identified with philosophy, nor with academic philosophical writing, but I think we can and should make that connection. The worry, then, is that the academic philosopher's way of seeking out truth or coming to understanding involves a certain form of pressure-a pressure that one's conclusions take the form of a theory, perhaps, or that they consist of propositions that are, wherever possible, general and perhaps even universal-that can warp our reports and perceptions of the world as actually experienced, encouraging us and perhaps even requiring us to interpret away, or ignore altogether, phenomena that do not fit with or exhibit some tension toward the theory toward which we are pushing. Just as the empirical sciences tend to neglect whatever is not public and repeatable, academic philosophy is too often constrained by the pressure to ignore whatever does not fit nicely into a theory, to disregard "honest complexity" in favor of "reductive simplicity." As the philosopher Robert Nozick once wrote, "A person may feel that she and her life are richer than any theory." (Nozick, 1989, p. 298). A philosophy too doggedly focused on expressing its points in terms of theories might risk missing that richness; whereas poetry, which tends to avoid the aspiration toward theory, might perhaps do better.
Of course, lyric poems too can be reductively simple, and academic philosophy articles can recognize honest complexity. The suggestion, though, is that there is something in the form of lyric poetry that makes this failing less likely, so that reductive simplicity will be more likely to be recognized as a flaw in a lyric poem while being praised as a virtue in a piece of academic philosophical prose. After all, lyric poetry, with its characteristic focus on the conscious experience of its speaker-and its commitment to treating this experience not as something static but as something that moves and changes through time-very often attempts to capture or depict life as it is actually experienced, to present a processed thought as it actually occurred. This is one of the places where the formal devices deployed by poets matter a great deal, for the choices a poet makes here-where to break a line, how long a certain description or passage should be-give the reader considerable information about how the poem is to be read; and this can allow poetic descriptions of conscious thought to gesture toward and even replicate many aspects, especially temporal aspects, of the thinking process itself.
Standard academic philosophical writing tends to approach its subjects, whatever they may be, from a somewhat detached perspective; typically it presents the outcome of a thinking process without, for the most part, recording or exploring the details of the process that led to that outcome. Of course, objections that occurred to the philosopher during the process might be acknowledged and responded to, but only if they are seen as relevant to the conclusion's being taken as true; only, that is, if it is judged likely that those objections will also occur to others, and hence need to be dealt with. But other forms of thought-false starts, speculations, even moments of sudden understanding, or insight-will not be expected to make an appearance in the philosophical article that results, even where it is true that the process would have been quite different, and the conclusion might never had been reached, had those mental events not occurred. Lyric poetry, by contrast, often takes a deep interest in changes of mind that take place during thinking. And because lyric poetry is interested in the activity of thinking itself, and not just in its outcomes, a lyric poem can happily devote itself to a process of thinking that does not lead to any specifiable or concrete outcome. This, too, can be of enormous value in coming to understand ourselves as thinking beings. A poem such as, for instance, Robert Hass's poem, "On Squaw Peak"-to take just one of a vast number of possible examples-can serve as a rich and finely detailed description of the process of thinking. 4 And precisely because such poems are not constrained by an ideal of "reductive simplicity," they can help us notice things about our own thinking, and other aspects of our conscious experience, that we are otherwise liable to minimize, if not miss altogether. 5 One thing we might notice, for instance, is that our conscious experience is considerably more complex than we often acknowledge, with a great many elements going on simultaneously in ways that often do not seem at all systematic or orderly and at times may feel entirely arbitrary and chaotic. The poet C.K. Williams writes that "The mind I experience myself seems to be a much more chaotic and turbulent phenomenon than those I've heard about: there is much more happening in it, much more happening at once, and much more happening in a nondetermined way." (Williams, 1998, p. 3). A second and related observation is that our consciousness is not at all constant, but is constantly changing. To the extent that our perceptions are responsive to the world, they change each time the world changes; but even when things around us appear to be holding still our mental lives refuse to do so. And it is not just the content of our perceptions and ideas that is in constant flux, but also our judgments as to which of those ideas are compelling, which ones are plausible or implausible, which ones we cannot bring ourselves to accept, which ones we can't bring ourselves not to accept. We are constantly changing our minds. (We might also say that our minds are constantly changing themselves.) One of lyric poetry's great advantages over standard philosophical writing, in terms of reproducing the actual feel of the thinking processes that lead to the ideas philosophers propose and defend, is its ability to recognize and reflect this sort of constant change in both the content and the structure of a given poem.
All of this would seem to suggest a philosophically interesting role for lyric poetry-if, at any rate, the subject matter of our inquiry is consciousness itself; one is, perhaps, likely to get a better idea of what human consciousness is actually like by reading certain lyric poems (or, for that matter, certain novels) than by reading theoretical philosophical writings about consciousness. This is not intended as a slight against the latter; the point is simply that such writings need to be complemented with other sorts of writings if we are to get an adequate picture. We might agree with Jarrell to this extent: poetry is in general a bad place for theorizing. And Jarrell might well have taken theorizing, thinking in its most abstract forms, to be the very essence of "philosophizing." But there is more to philosophy than this. To capture what human experience is really like-the way, for instance, our flow of conscious thoughts accommodates an immense plurality of ideas and feelings, many of which seem incompatible with each other and do not seem connected in any apparent logical manner even from one moment to the next-we need to go beyond theory. Poetry can be a very useful place to go. I do not mean to suggest, however, that consciousness is the only subject with respect to which lyric poetry should be expected to be a useful and powerful philosophical tool. One of the reasons for desiring a linguistic instrument that is capable of capturing the flow and play of ideas that are not reducible for or cleanly compatible with one another is the very real possibility that the world itself might be pluralistic, that the world might contain and be composed of elements that cannot all be brought together to fit smoothly into a single comprehensive representation or conceptual scheme. If this is the case-and a good deal of human experience must surely be taken as supporting the view that it is-then poems may provide ways of getting at some truths, or even at an overall depiction of "the truth" (which, we now begin to see, is probably precisely the wrong phrase to deploy here), that the philosophy journal article, as typically formulated, cannot approach. In a philosophy article, the first person, whether or not the article is written in the first person, is implied, in the sense that it is the author who is to be understood as making the claims and arguments contained in the article. These claims are to be understood as assertions, and the person asserting is the author. The reader is to proceed on the assumption-which is very often though not always the case-that the author sincerely believes that the claims she is asserting are true and that the arguments are convincing. And, as noted above, what is presented is not the entirety of the thinking process, or anything approaching that, but only the conclusions of that process-properly interpreted, cleaned up, and checked for tensions and inconsistencies-along with whatever arguments, responses to objections, etc. are thought necessary to render those conclusions convincing. The assumption, that is, is typically that the philosophical author has a theory to offer, and that anything in apparent tension with that theory has been accounted for or, ideally, explained away or otherwise eliminated. By the time the work reaches the reader, the "honest complexity of poetic action" has all too often been smoothed and manicured so as to maximize "the reductive simplicity of proposition." This is of course a generalization; some philosophers are far less interested than others in cleaning up our accounts of the world in this way. Wittgenstein's aphoristic approach is a fairly obvious and well-known example of how to write philosophically while resisting the urge to force every insight or truth into the Procrustean bed of a consistent and comprehensive theory. In the preface to Philosophical Investigations he writes, After several unsuccessful attempts to weld my results together into such a whole, I realized that I should never succeed. The best that I could write would never be more than philosophical remarks; my thoughts were soon crippled if I tried to force them on in any single direction against their natural inclination. -And this was, of course, connected with the very nature of the investigation. For this compels us to travel over a wide field of thought criss-cross in every direction. -The philosophical remarks in this book are, as it were, a number of sketches of landscapes which were made in the course of these long and involved journeyings. (Wittgenstein, 2009, p. 5e) Another well-known formal alternative to the journal article, and one that has at times found a place within the discipline of philosophy, is the dialog form. The ideas a dialog engages with are attached to and expressed by the dialog's characters, who are considered to be distinct from the author. This allows for the presentation, articulation, exploration, and defense of a variety of distinct arguments and positions, while avoiding the necessity to come to an overall conclusion that bears the mark of finality. This is to say that at the end of the standard academic essay it ought to be possible to answer the question, what does the author think?, whereas at the end of a philosophical dialog it is possible, and perhaps quite proper, to remain in considerable doubt about what the author thinks. (After all, in some cases the author herself might not know what to think.) Since the standard work of this form contains only dialog-the words uttered by the characters-plus perhaps minimal stage directions, there is no place for the author to assert herself, to express her own views, or even to designate one of the characters as speaking for her.
Rather than simply being asserted, as in standard philosophical essays, the dialog form allows ideas to be, in an important sense, dramatized. Although a certain amount of exploration and contemplation can be achieved in a philosophy journal article, the dialog form often allows for a greater degree of these, both because a dialog allows for the natural presentation of conflicting viewpoints and because a dialog is permitted to end on an uncertain, inconclusive note, and need not strive for the kind of closure we expect from journal articles. Moreover, in attaching ideas to characters a dialog is frequently able to demonstrate the implications and consequences of the ideas it treats, not only for thinking but also for life. It can therefore embody abstract ideas in concrete particulars in a way that the academic essay cannot. In Plato's Symposium, for instance, a great deal of the discussion that composes the dialog is concerned with the character of Socrates (in both of the main senses of this phrase). Socrates, as both a participant in the dialog and an object of attention and scrutiny by his fellow participants, serves not only as an investigator into the nature of beauty and love, but also as an example of a person whose thought, behavior, and perhaps very being are guided by these things and, at the same time, express or exemplify them in certain forms.
Of course, the typical lyric poem is often told from the perspective of a single person-the "speaker" of the poem-and can be seen as representing, in that sense, a single perspective. But as C.K. Williams reminds us, the perspective of a single person is often anything but unified, and it is precisely that complexity and lack of unity-the inner struggle and conflict of making up one's mind, or of having a mind that cannot be cleanly and conclusively made up-that the lyric poem is ideally suited to capture. Such poems are quite capable of capturing competing viewpoints; it is just that these viewpoints will typically represent multiple perspectives that arise and compete within a single person. Moreover-to return to a point mentioned earlier in the paper-the style of the poem, the way its ideas are expressed, affects its possible meanings and tends to disrupt any attempt to force on the poem a single, determinate meaning. Although a poem is a linguistic object, composed of words, it is a mistake to think that asking what a poem means is like asking what a road sign in a foreign country means, what a certain clause in a legal contract means, or, for that matter, what a passage in a philosophy paper means. Rather, it is far more like asking what a song or a painting means. (If people sometimes forget this, it is because, while songs and paintings may contain words, they are also made of non-linguistic materials. Poems, like songs, are musical, and their musical aspects, including performative aspects, are relevant to their meanings: compare a performance of "Stand By Your Man" by Tammy Wynette with a performance of the same song by Lyle Lovett. But because the music of a poem is entirely embodied in and arises entirely from the linguistic materials, it is more tempting here to forget the relevance of the music to the poem's meaning, and mistakenly think that the literal meaning of the poem's words must constitute, in a fairly straightforward manner, the poem's meaning.) This phenomenon-we might state it, briefly, as the fact that what a poem says is not necessarily what it means-supports the claim that poetry's way of treating ideas is at least as radically divergent from standard philosophical essays as is that of the dialog form. For here, rather than being tied to particular characters as they are in a dialog, ideas are freed from such limitations and can appear, compete, and interact in their own right. We might think of a lyric poem such as "On Squaw Peak" as a philosophical dialog in which, rather than being expressed by the characters, the ideas themselves are the characters. At the same time, the lyric poem has an advantage over the philosophical dialog in that it need not strive to end itself by reaching a definitive conclusion or leave its reader with the sense that whatever issue was under consideration has been satisfyingly resolved. This suggests that the lyric poem, as a genre of writing, is especially well suited for the depiction of certain types of experience: those in which we feel strongly pulled in more than one direction, those in which our various perceptions or ideas refuse to come together to form a clean and unified set of propositions; those, generally, in which the world presents itself as a deeply pluralistic universe rather than as a monistic whole. Thus, to recognize the poem's capacity for simultaneously dramatizing diverse competing ideas without attempting to reduce them to a single theoretical account is not to impugn its ability, as a literary form, to serve as an instrument for getting at the truth about things. In fact, if certain things are true about the world-if the world, by its nature, is in certain ways complex and pluralistic-these facts suggest that poetry may be an especially powerful instrument for getting at truth.
If the world is in a deep way genuinely pluralistic then we can understand why dealing with ideas in the straightforward manner in which academic philosophical writing tends to deal with them-a method that takes its final goal to be the clarification of ideas and the resolution of all apparent conflict into a single, conflict-free truth-is not always to be preferred over approaches that are more willing to consider the possibility that there might be something highly valuable to be learned from irreconcilable conflicts, some insight to be mined from the clash of powerful ideas that exist in tension with one another. Terry Eagleton, in his comments on the first stanza of Yeats's "Sailing to Byzantium," points out how, in capturing the multiple, irreconcilable moods a person can be subject to at one particular moment as captured in a poem, the poem moves between multiple, mutually irreducible judgments, and even inhabits mutually incompatible perspectives in the same moment: The poet is telling us that he must abandon the perishable domain of human love, sexuality, death and reproduction for some more enduring kingdom, one less carnal and fugitive. Yet even though the opening demonstrative already places this perishable domain at arm's length ('That' rather than 'This'), the imagery which portrays it is tender and mutedly sensuous. And this grants the natural, human world of the dying generations a grace and preciousness which makes it hard to abandon. Yeats is refusing to make things easy for himself by setting up a convenient straw target of the fleshly world he is leaving behind […] Instead, he pays homage to what he is repudiating. […] We suspect that he believes that the profane realm of the dying generations is pretty degenerate anyway, but he is in elegiac mood, and thus tactful enough not to say so outright. (Eagleton, 2007, p. 84).
Because a poem can dramatize ideas rather than merely state them, and because, as Eagleton notes, a poem can strongly suggest and in a sense express things they do not say "outright"-it can, as I put it before, mean things that it does not say-poetry can do a kind of philosophical work, making significant contributions to the ongoing project of achieving a better understanding of the world as we encounter it. In particular, by providing finely discriminating and carefully articulated descriptions of human experience it can help us judge, evaluate, and appreciate values as they are manifested in the world. And it can help us think about how we should live, and what the various possibilities for living are, in light of uncertainty, ambiguity, and the apparently irreconcilable conflicts that can arise between competing values and between other elements of a world that at least at times presents itself as irreducibly pluralistic.
Of course some people will deny that this is proper work for philosophy, insisting on the traditional view that philosophy's goal is the development of a single, consistent, comprehensive theory. Philosophers and scientists of a certain mindset have always hoped that we would eventually find our way to a Grand Unified Theory of Everything, and the possibility that this might one day be achieved cannot be conclusively ruled out. That said, it is important to acknowledge that our actual human experiences of the world-and this is something that lyric poetry can help us not to forgetsurely suggests, at least a good deal of the time, that no such overarching consistent account of the universe is in the offing. If nothing else, continuing philosophical perplexities about apparent tensions between the first-personal and third-personal view of ourselves-questions of consciousness, personal identity, agency, and free will-should surely at least make us wonder whether a total reconciliation of everything that seems to be true about human beings can ever take place.
For that matter, the idea of fully understanding anything-of knowing literally everything there is to know about something or other-seems elusive, no matter how modest an object we might choose as our "something or other." And of course the objects we most want to understand (human beings being the foremost example I have in mind here) are not in this respect modest at all. The idea of a genuinely total theory of human beings, or human behavior, seems unlikely, perhaps ludicrously so; yet no one who is at all familiar with the history of attempts to understand humanity through literature, psychology, the empirical sciences, and a great many other approaches will doubt that progress has been made and that a great many insights have been achieved. There are many different ways of looking at human beings, and each of these ways yields insight; but it is likely that there is no way to cleanly fit all of these ways of understanding into a single comprehensive intellectual framework. It may well be that, as Nietzsche suggested, understanding is inherently perspectival: the best way to work toward a complete understanding is not via some sort of magical omniscient eye, but by multiplying our understandings and viewing our subject from as many different mutually irreducible perspectives as possible: "There is only a perspective 'seeing,'" Nietzsche writes, "only a perspective 'knowing' … the more affects we allow to speak about one thing, the more eyes, different eyes, we can use to observe one thing, the more complete will our 'concept' of this thing, our 'objectivity,' be." (Nietzsche, 1989, section 3:12). Sometimes, after all, the most useful thing is not to be able to see which of two apparently competing claims is actually true, but to be able to see just what each of them has managed to get right. 6 Philosophical thinking is not only a matter of checking our positions for coherence and checking our arguments for possible fallacies. There are also valid roles for contemplation, for speculation, for opening up rather than ruling out possibilities. And the possibility that the world might be too complex and various ever to be fit into a single unified framework of understanding must always be acknowledged, at least up to the point when we have the Grand Unified Theory of Everything in our hands. Until that theory arrives, a certain sort of lyric poetry will continue to be not only a potent spur to philosophical insight but a useful and likely indispensable way of actually doing philosophy, in large part because of its attentiveness to detail and fact, its insistence on attempting to capture and reflect things as they happen, no matter how odd, unexpected, or incomprehensible they might appear to be, as accurately as human beings can. Free from the distorting lens of an imposed theory, acknowledging no constraint to be defensible or, for that matter, coherent, the lyric poem is able fully to acknowledge whatever plurality and complexity it encounters in the context of human beings and their experiences of the world, and it is precisely this that allows it to make significant contributions to our understanding of the world and of ourselves.

Funding
The author received no direct funding for this research.

Citation information
Cite this article as: "Sketches of landscapes": On philosophical poems, Troy Jollimore, Cogent Arts & Humanities (2016), 3: 1123081. Notes 1. Cf. Roger Scruton: "In an important sense the poetic utterance is to be understood non-instrumentally. Not only are the words and the allusions contained in them to be weighed and appreciated for their own sakes; the same is true of the subject matter too. [ …] [I]f it is real poetry then much that it says will be lost when told in another way. The effect of poetry depends on the way of telling, or rather (to adapt a well-known distinction) on the way of showing what is told.
[…] The sound of the words, their associations, the things they call to mind, and all the other ways in which words and things can be connected, begin to compete in the reader's or listener's mind with the semantic content." (Scruton, 2015). 2. The quote is frequently attributed to Robert Frost, though it appears likely he never put it in quite those words. A somewhat similar statement appears in (Frost, 1961). 3. Some poets may think this sounds a bit too coercive.
They might prefer to say something like, "I just want the reader to have some sort of strong, memorable experience-I don't care what it is; any experience will do, so long as it is vivid." But of course not any experience will do; hating a poem, or finding it tremendously boring or pretentious, might be a vivid experience, but it is not the sort most poets aim at. I suspect that most poems, at least, intrinsically aim at something like what I suggest in the main text; though it might be that many of the poets who write them are willing to be satisfied with less. (Hass, 1989) and reprinted in his The Apple Trees at Olema: New and Selected Poems (Hass, 2010). The original version of this paper contained a fairly detailed discussion of this poem. Unfortunately the highly restrictive permissions policies of this journal prohibited me from quoting passages from the poem, and so made it necessary to cut the entire discussion. 5. Cf. Peter Lamarque: "What a poem offers is not, usually, a single thought on its subject but a thought-process. The value of poetry is at least partially the value of following a thought-process through the perspective of a poetic speaker." (Lamarque, 2015, p. 31). 6. One might compare F. Scott Fitzgerald's well-known statement, "The test of a first-rate intelligence is the ability to hold two opposed ideas in mind at the same time and still retain the ability to function.".