Science & Education

, Volume 23, Issue 3, pp 589–607

Observation, Inference, and Imagination: Elements of Edgar Allan Poe’s Philosophy of Science


    • Department of Philosophy, Faculty of Arts and Social SciencesNational University of Singapore

DOI: 10.1007/s11191-012-9551-8

Cite this article as:
Gelfert, A. Sci & Educ (2014) 23: 589. doi:10.1007/s11191-012-9551-8


Edgar Allan Poe’s standing as a literary figure, who drew on (and sometimes dabbled in) the scientific debates of his time, makes him an intriguing character for any exploration of the historical interrelationship between science, literature and philosophy. His sprawling ‘prose-poem’ Eureka (1848), in particular, has sometimes been scrutinized for anticipations of later scientific developments. By contrast, the present paper argues that it should be understood as a contribution to the raging debates about scientific methodology at the time. This methodological interest, which is echoed in Poe’s ‘tales of ratiocination’, gives rise to a proposed new mode of—broadly abductive—inference, which Poe attributes to the hybrid figure of the ‘poet-mathematician’. Without creative imagination and intuition, Science would necessarily remain incomplete, evenby its own standards. This concern with imaginative (abductive) inference ties in nicely with his coherentism, which grants pride of place to the twin virtues of Simplicity and Consistency, which must constrain imagination lest it degenerate into mere fancy.

1 Introduction

Edgar Allan Poe’s standing as the inventor of both the detective story and science fiction makes him an intriguing character for any exploration of the historical interrelationship between science, literature and philosophy. Yet his relationship to science and philosophy was at best an uneasy one. His dislike for Transcendentalism, the most influential school of thought in the New England of the mid-1830s to late 1840s, is well-known and led Poe to refer to its followers as ‘Frogpondians’, after the frog pond on Boston Common, where Ralph Waldo Emerson and his fellow Transcendentalists could often be seen to go for a stroll. Later, in an 1844 letter to the poet Thomas Holley Chivers, Poe qualifies his dismissal when he writes: ‘You mistake me in supposing I dislike the transcendentalists—it is only the pretenders and sophists among them.’1 This suggests that Poe objects not to the activity of philosophizing per se, but to a perceived lack of sincerity and authenticity on the part of its practitioners. Indeed, Poe goes on to tell Chivers that ‘[m]y own faith is indeed my own’, and summarizes the key philosophical ideas of a proposed article as follows:

There is no such thing as spirituality. God is material. All things are material; yet the matter of God has all the qualities which we attribute to spirit: thus the difference is scarcely more than of words. (Poe 1966: 260)

When Fyodor Dostoevsky, in 1861, writes that ‘if there is fantasy in Poe, it is a kind of materialistic fantasy’,2 then the same may be said of Poe’s overall philosophical leanings: If there is philosophy in Poe, it is a kind of materialistic philosophy. Poe’s scattered philosophical remarks, however, are especially resistant to systematization, not least due to what has been described as ‘Poe’s peculiar espousal of apparently opposite philosophies’ (Hovey 1995: 348). Perhaps this is why attempts at unearthing a coherent philosophical system behind Poe’s ruminations have given rise to such awkward labels as ‘materialistic idealism’ (Feidelson 1953: 37) and ‘Psychal Transcendentalism’ (Carlson 1973: 10).

The focus of the present paper is much narrower. In particular, I shall focus on Poe’s views on natural philosophy and the methodology of science. For this, I shall draw mainly—though not exclusively—on Poe’s essay (or ‘prose-poem’) Eureka, which purports to offer an explanation of the origins and evolution of the physical Universe as well as a critique of traditional accounts of scientific methodology. As I hope to show, Poe claims to have identified a new method of inference, which transcends deduction and induction in that it enables its practitioners to move imaginatively from our imperfect and fallible knowledge of ‘everyday’ empirical matters to a deep understanding of the fundamental principles that govern the world of universal phenomena. This third—as I shall argue, broadly abductive—mode of inference is best developed among those ‘poet-mathematicians’ who, like Kepler, are endowed with a superior faculty of the imagination.

The remainder of this paper is organized into four sections. First (Sect. 2), I shall outline the content and argumentative strategy of Poe’s ‘prose-poem’ Eureka, published in 1848. While the text has sometimes been scrutinized for anticipations of later scientific developments, I argue that it ought to be understood as a contribution to the contemporaneous debates about the methodology of scientific inquiry. Section 3 argues that Poe’s methodological interest is also an important feature of his famed ‘tales of ratiocination’. Indeed, Poe’s hybrid figure of the ‘poet-mathematician’ must be taken seriously, precisely because, without poetic imagination and intuition, Science would necessarily remain incomplete, evenby its own standards.3 Section 4 ties Poe’s concern with imaginative (abductive) inference into his defense of coherentism about truth, which grants pride of place to the twin virtues of Simplicity and Consistency, which must constrain imagination lest it degenerate into mere fancy. The paper concludes with a brief assessment of the merits, and the novelty, of Poe’s—largely implicit—‘proto-philosophy of science’.

2 ‘Eureka’: Natural Philosophy as Cosmology

In June 1848, barely a year before his death, Edgar Allan Poe published Eureka, an ambitious work in ‘natural philosophy’, based on public lectures given earlier that year and incorporating passages and themes from earlier writings on cosmology and the science of his times. Excerpts of Eureka had been presented by Poe on 3 February 1848, in a public lecture entitled ‘On the Cosmogony of the Universe’, but his interest in related scientific topics can be traced back to reviews and comments on Peter Mark Roget’s Physiology (which Poe reviewed in February 1836) and William Whewell’s Astronomy and General Physics Considered with Reference to Natural Theology (1834), both of which were published as installments of the Bridgewater Treatises.4 As Susan Welsh has argued, ‘[t]wo kinds of scientific writing seem to speak through Eureka—natural theology writing, and the more secular and largely synthetic surveys of scientific theory or information as it stood in the 1840s’ (1991: 3). Much of the former remained indebted to traditional theistic arguments and looked for common ground between the rapidly expanding scientific knowledge and belief in an intelligent Creator and revealed religion. Thus, William Herschel, in his Preliminary Discourse on the Study of Natural Philosophy (1830: 7) could write that, while ‘the testimony of natural reason […] places the existence and principal attributes of a Deity on such grounds as to render doubt absurd and atheism ridiculous, it unquestionably opposes no natural or necessary obstacle to further progress’, and the same sentiment—or hope—was shared by other religious men of science, who hoped to ‘lead the friends of religion to look with confidence and pleasure on the progress of the physical sciences’ (Whewell 1833: vi). Yet it proved difficult to ignore the mounting tension between scientific accounts of the world and traditional beliefs based on a literal interpretation of scripture, especially concerning the formation of the Earth itself. Thus, early in the nineteenth century, natural theologians like William Paley had counted Laplace’s nebular hypothesis—which includes as its corollary ‘that the planets themselves are only cooled or cooling masses’—among ‘the guesses of those who reject an intelligent Creator’ (Paley 1809: 383). Conservative worries about the impact of scientific findings—on the religious as well as on the social and political order—were fuelled by conclusions of the sort drawn by John Pringle Nichol in his Views of the Architecture of the Heavens, who suggested that the nebular hypothesis shows that ‘the existence of the human race is an invisible speck’ in a mere ‘stage of condensation in a secondary nebula’ (1839: 151–152). Nichol, whom Poe refers to as ‘Dr. Nichols’ in The Murders in the Rue Morgue (in a passage in which Poe’s master detective, M. Dupin, outlines his chain of reasoning) and whose Views of the Architecture of the Heavens he discusses explicitly in Eureka, thus foreshadows a more ambitious scientific positivism, according to which science need not content itself with looking for natural-theological evidence, but can legitimately aim at developing a rival world view.5 Thus, as Lawrence Frank (2003: 30) puts it, Poe was writing ‘at a time in which both a resurgent evangelicalism and a conservative Natural Theology were confronted by a positivist science’, and it is against this historical backdrop that one must read Eureka, Poe’s ‘Essay on the Spiritual and Material Universe’.

Taken at face value, Eureka attempts to provide nothing less than an account of the origins of the Universe, of the principles that govern its law-like evolution, and of its eventual collapse, along with a discussion of the place of the human mind within this cosmological framework. About the success of this audacious (or, rather, megalomaniac) project, Poe had not the slightest doubt. In a letter of January 1848 to one of his followers, George Eveleth, Poe writes: ‘What I have propounded will (in good time) revolutionize the world of Physical & Metaphysical Science. I say this calmly—but I say it.’6 Puzzlingly subtitled ‘A Prose-Poem’, Eureka, in the words of one of Poe’s biographers, ‘allies itself with the thinking of mechanists, Deists, and pantheists’ (Wagenknecht 1963: 215):

In the Eureka world, God is unparticled matter, more rare than ether. Creation began when God willed into being a primordial particle, from which a shower of atoms radiated into space. […] Diffusion having occurred, the atoms immediately sought to return to their original condition through gravitation. But if gravitation had been allowed to follow its course unchecked, no universe could have come into being; at this point alone, therefore, God intervened to set up a counter-force of electrical repulsion. (Wagenknecht 1963: 216)

All cosmological evolution is the result of the interplay between the twin forces of attraction and repulsion, accounting for the large-scale structure of the ‘perceptible Universe’ as ‘a cluster of clusters, irregularly disposed’ (Eureka, p. 73).7 Eventually, repulsion will cease to counteract gravitational attraction, leading to ‘the clusters themselves, with a speed prodigiously accumulative, […] rushing towards their own general centre’ (Eureka, p. 100) and all matter returning into a state of unity. This ‘inevitable catastrophe’ (ibid.), however, may just be the beginning of ‘a novel Universe swelling into existence’ (Eureka, p. 103). Along the way, Poe discusses at length, and with varying degrees of competence, phenomena of electricity, optics, and Laplace’s nebular hypothesis. While Eureka had little measurable impact on scientific developments at the time of publication, later admirers of Poe, such as Paul Valéry, praised it as ‘an abstract poem, one of the rare modern examples of a total explanation of the material and spiritual universe, a cosmogeny.’8
Poe enthusiasts have more than once pointed to parallels between some of what Poe says and results in modern astronomy as evidence of the scientific merit of Eureka.9 Some of the similarities are indeed striking, not least Poe’s vision of the Big Bang-like origins of the Universe:

The assumption of absolute Unity in the primordial Particle includes that of infinite divisibility. […] From the one particle as centre, let us suppose to be irradiated spherically—in all directions—to immeasurable but still definite distances in the previously vacant space—a certain inexpressibly great yet limited number of unimaginably yet not infinitely minute atoms. (Eureka, pp. 23–24.)

Similarly farsighted is the link he makes between the long-term effects of gravitational attraction and the large-scale structure of the ‘perceptible Universe’ as ‘a spherical space interspersed, unequably, with clusters’ (Eureka, p. 72). As Harold Beaver (1976: 400) notes with respect to the factual claims contained in Eureka, ‘Poe scored some surprisingly clear-headed hits’, ranging from a questioning of the uniqueness of Euclidean geometry to his recognition that the light of distant galaxies shows ‘the phantoms of processes completed long in the Past’ (Eureka, p. 69), and to his insistence on a spatiotemporally finite Universe of Stars. With utmost generosity, one might even credit Poe with anticipating the centrality of the notion of symmetry in contemporary (e.g., particle) physics, for example when he describes symmetry as ‘the poetical essence of the Universe’, urging that ‘the sense of the symmetrical is an instinct which may be depended upon with an almost blindfold reliance’ (Eureka, p. 96). Less charitable readers of Poe will feel uneasy about placing their faith in what has sometimes been characterized as a haphazard piece of cosmological speculation. When approached for his opinion in 1940 by Poe’s biographer Arthur Quinn, the astronomer Sir Arthur Eddington diplomatically described Eureka as ‘the work of a man trying to reconcile the science of his time with the more philosophical and spiritual cravings of the mind.’10 It would indeed be easy, with the benefit of hindsight, to ridicule Poe’s project as overambitious, pseudo-scientific, and wildly inconsistent, and to criticize him for glaring mistakes—such as his claim that ‘our moon is strongly self-luminous, [as] we see at every total eclipse, when, if not so, she would disappear’ (Eureka, p. 63). However, when one looks at Eureka in the light of the spirit of the time, one cannot but agree with the editors of a recent critical re-issue of Eureka that, on the whole, Poe’s observations ‘are very competent’, his extrapolations ‘consistently intelligent’ (if overly speculative), and much of his presentation ‘no more rhetorically overblown than other comparable statements of the era’ (Levine and Levine 2004: xx).

Most of those who have written on Poe have tended to tactfully gloss over the bold, and often overstated, scientific assertions that are scattered throughout the text of Eureka. Thus, Edward Wagenknecht writes with thinly veiled false modesty that it would not ‘be in order for me to evaluate it [=Eureka] either as science or metaphysics even if I myself had either the science or the metaphysics necessary to achieve this’ (Wagenknecht 1963: 215). Others, perhaps in an attempt to protect Poe from his own more outlandish claims, have emphasized the ‘ironic distance’ that Poe allegedly places between his cosmological speculations and himself as author (Weissberg 1991: 115). Yet others have suggested that the whole text should be regarded ‘not so much as a theory of the known universe but as a parallel universe created by Poe’ (Peeples 2002: 188)—one that conforms to his narrative vision and the poetics of decline and rebirth that pervades Poe’s fiction. Thus, George Kelly has called Eureka ‘a remarkably ingenious effort to construct a cosmology based upon aesthetic theory.’11 Peter Ackroyd, finally, in what is perhaps the laziest and least charitable interpretation, sees in Eureka little more than another indication of Poe’s descent, towards the end of his life, into another—final—episode of excessive drinking and a ‘desire to return to some state of infantile bliss and tactility’ (2008, p. 136).

Much has been made of Poe’s statement that he wished to have his composition judged ‘as a Poem only’ after his death. At the same time, however, he emphasized that ‘What I here propound is true: —therefore it cannot die’ (Eureka, p. 5). Clearly, then, Poe’s designating Eureka a ‘prose-poem’ cannot mean that he did not wish it to be taken seriously. It has even been claimed that the subtitle ‘A Prose Poem’ may have been suggested to him by a review published in the New York Express, which stated: ‘The work has all the completeness and oneness of plot required in a poem.’12Eureka itself develops the idea that poetry, understood as an exercise of the imagination, is able to access truths that “ordinary” methods of reasoning (such as induction and deduction) must necessarily miss out on: ‘Poetry and Truth are one.’ (Eureka, p. 96) Indeed, as Kuno Schumann rightly argues, ‘it is not enough to take as the starting point for an interpretation of these [“scientific”] works of Poe’s either his journalistic desire to constantly offer his audience novelties and sensations, or his tendency to indulge in pseudo-science’ (Schumann 1958: 87; my translation). Poe was evidently proud of what he thought he had achieved in Eureka, writing in a letter to a Mrs. Clemm, shortly before his death: ‘I have no desire to live since I have done Eureka. I could accomplish nothing more.’13

It is obvious that Poe could hardly have given himself credit for those (posthumous) scientific developments that later admirers have read into his texts. In spite of his keen interest in specific physical processes and phenomena, Poe did not so much aim for scientific details, but for a transformative vision of the scientific Universe. In a letter to George E. Isbell, dated 29 February 1848, Poe laments the ‘objections of merely scientific Men—men, I mean, who cultivate the physical sciences to the exclusion, in a greater or less degree, of the mathematics, of metaphysics and of logic’; such objections ‘are generally invalid except in respect to scientific details’. What is instead called for is ‘using, generalizing, or deciding upon the facts’ which are brought to light by scientific experiments—yet those very ‘efforts at generalization’ are typically denounced ‘as “speculative” and “theoretical”’.14 Thus, in order for Poe’s cry of ‘Eureka!’ to be vindicated, his achievements—and the source of Poe’s pride in his work—must lie not in scientific details, but elsewhere. An important clue as to where one might look consists in his rejection of pseudo-scientific flights of fancy, such as Mesmerism, Swedenborgianism, and any ‘other equally delicious ism of the same species, and invariably patronized by one and the same species of people’ (Eureka, p. 36). Poe is eager to distance himself from such unworthy endeavors, ridiculing the Swedenborgians in particular for their credulity, since they fell for one of his hoaxes (the ‘Mesmeric Revelation’).15 Clearly, then, Poe believes that the argument he develops in Eureka is not just another vision of how the physical and the spiritual relate to one another, but instead captures a fundamental and novel aspect of our scientific aspirations. As I shall argue in the remainder of this paper, even in its overt scientific aspirations, Poe’s project in Eureka is deeply methodological in character: Poe claims to have identified a new method of inference, which stands alongside deduction and induction, and which allows one to make an imaginative leap from the imperfections of our finite and fallible ‘everyday’ knowledge to the fundamental principles that govern the world of ‘Universal phenomena’ (Eureka, p. 35).16

3 From Ratiocination to Abduction: The Poetics of Inference to the Best Explanation

As early as 1907, J. Brander Matthews declared Poe to be the master, and inventor, of the modern detective story.17 Poe himself does not, of course, speak of ‘detective stories’, but of ‘tales of ratiocination’, and indeed only three of these tales feature Poe’s master detective, M. Dupin. An example of a ‘non-detective’ application of the ratiocinative method may be found in Poe’s story A Descent into a Maelstrom. In this story, the narrator recounts the terrifying experience at sea of desperately trying to save himself from being drawn into a gigantic whirlpool. Realizing that he is physically powerless to extricate himself from his predicament, he observes the kinds of physical objects floating around him, analyzing which physical shapes are more susceptible to being drawn under. He infers, ‘partly from memory, and partly from present observation’, that a cylindrical object is last to be sucked up in the whirlpool, and saves himself by holding on to a cask until the maelstrom has subsided.18 Such ratiocinative inference, apart from its obvious survival value in the present example, for Poe also is a source of joy: ‘As the strong man exults in his physical ability, delighting in such exercises as call his muscles into action, so glories the analyst in that moral activity which disentangles.’19

It is a commonplace that the character of Sherlock Holmes is modeled after Poe’s M. Dupin; Nancy Harrowitz even goes so far as to say that ‘most of the principles of Dupin’s method were lifted outright by Conan Doyle and immortalized in the creation of Sherlock Holmes’ (Harrowitz 1988: 193). However, it is not Dupin but Holmes who had the greater impact on popular culture, not least thanks to the catchphrases ‘Elementary, my dear Watson’ and ‘It’s all deduction’.20 By the middle of the 20th century, Raymond Chandler, in his essay The Simple Art of Murder (1944), voices a certain fatigue when he speaks rather dismissively of the tradition of the ‘logic-and-deduction-novel’, which operates on the assumption that ‘no story is a detective story which does not pose a formal and exact problem’ (Chandler 1944: 60). However, as philosophers never tire of pointing out, when Sherlock Holmes (or Dupin, for that matter) solves a crime, the inferences involved are often neither elementary nor typically deductive, but usually involve choosing from among a number of competing explanatory hypotheses, all of which are compatible with (and, to a varying extent, supported by) the empirical evidence. In the parlance of contemporary epistemology, they are abductive inferences, i.e. inferences to the best explanation.21 What counts as ‘best’ depends on context: While we often seek the most probable (i.e. ‘likeliest’) explanation, the objective likelihood of different explanatory scenarios may not be readily obvious. In such cases, one may well decide to settle for what Peter Lipton calls the ‘loveliest’ explanation—that explanation which, if true, would explain the most.22 In one of Charles Sanders Peirce’s formulations,

Abduction is the process of forming an explanatory hypothesis. It is the only logical operation which introduces a new idea; for induction does nothing but determine a value, and deduction merely evolves the necessary consequence of a pure hypothesis. Deduction proves that something must be; Induction shows that something actually is operative; Abduction merely suggests that something may be. (CP 5: 171; quoted after Harrowitz 1988: 181)

Peirce’s original motivation was the realization that he had been ‘too much taken in considering syllogistic forms’ (CP 2: 102), and that instead of taking premises as a given (as in the case of deduction), what was needed was abduction as ‘the first starting of a hypothesis and the entertaining of it, whether as a simple interrogation or with any degree of confidence’ (CP 6: 525).23
In his Dupin stories, Poe’s use of ratiocination is more specific and is intimately connected with the figure of the detective. In the first of the three stories, The Murders in the Rue Morgue (1841), Poe begins with a discussion of the human being’s ‘mental features’, using the example of chess, urging that the ‘analytical power’ of the mind ‘should not be confounded with simple ingenuity’.24 The actual detective story, concerning the brutal murder of two women, is presented as a mere illustration of these more basic considerations of the human cognitive apparatus.25 In the story itself, Dupin is able to prove, through a series of conjectures on the basis of evidence that had been disregarded by the police, that the killings were not in fact carried about by a human, but by an escaped orangutan. Along the way, Dupin laments that E. F. Vidocq, the famous French criminalist and first director of the Sûreté Nationale, ‘impaired his vision by holding the object too close’, whereby he ‘might see, perhaps, one or two points with unusual clearness, but in doing so […], necessarily, lost sight of the matter as a whole’.26 By contrast,

Dupin’s principles are these: never assume anything, the nature of the object under scrutiny must dictate the nature of the inquiry, it is necessary to keep sight of the matter as a whole, one must prove that “crucial impossibilities” are possible (if, indeed, they are so). (Harrowitz 1988: 193)

As the narrator of The Murders in the Rue Morgue puts it, it is ‘in matters beyond the limits of mere rule that the skill of the analyst is evinced’:

He makes, in silence, a host of observations and inferences. So perhaps do his companions; and the difference in the extent of the information obtained lies not so much in the validity of the inference as in the quality of the observation.27

As Paul Grimstad notes there is a parallel between this passage and Peirce’s notion of ‘firstness’ which refers to a level of (qualitative) observation that precedes active theorizing: ‘Quality here fuses the logical with the aesthetic, in that a type of inference “beyond the limits of mere rule” moves toward the perceptual, toward pattern recognition, and toward judgments of taste.’ (Grimstad 2005: 26) It is Vidocq’s lack of attunedness to factors that cannot easily be subsumed under rules that renders his methods ineffective. Yet, Vidocq’s failure to represent to himself, in thinking, ‘the matter as a whole’ is not an isolated one. In Eureka, too, Poe criticizes Baconian inductivism on similar grounds: ‘The error of our progenitors was quite analogous with that of the wise-acre who fancies he must necessarily see an object the more distinctly the more closely he holds it to his eyes’ (Eureka, p. 11).

Similarly, in The Purloined Letter (1844), Dupin quickly recovers a stolen letter in full view of the thief, a government minister, during a social visit, whereas the police prefect had wasted precious resources on an extensive, month-long, but ultimately unsuccessful, search of the minister’s premises—even though the prefect’s inductive methods, as Dupin admits, ‘were not only the best of their kind, but carried out to absolute perfection’.28 Interestingly, in the same text, Dupin also enters into a long diatribe against over-reliance on deduction, since ‘[w]hat is true of relation—of form and quantity—is often grossly false in regard to morals’, not least since ‘[i]n the consideration of motive it fails’.29 Dupin alone is able to prove the minister’s guilt since, like him, he is a ‘poet and mathematician’30—that is, someone who is able to combine analytical rigor with vigor of the imagination.31 The best-confirmed methods are worthless if one has not imaginatively solved the ‘problem of what to look for, how to direct the inquiry, which clues are important and which are irrelevant, what “truth” is being sought after’ (Harrowitz 1988: 194). Significantly, one finds a cross-reference in Eureka (p. 40) to The Murders in the Rue Morgue, in which Poe suggests that it is the ‘roughnesses’, the ‘peculiarities’, the ‘protuberances’, and not the inductive generalizations (or universal laws of deductive logic) that contain the key to the puzzle-solving ability of reason. Without guidance by an imaginative understanding, empirical observation has no way of ascertaining which details are relevant, or which premises apply in the given case at hand.

Another example of ratiocination as (a species of) inference to the best explanation is Poe’s 1836 essay on Maelzels Chess-Player.32 Johann Nepomuk Maelzel toured the world with what he claimed was a purely mechanical chess-playing machine, and Poe witnessed one of his performances at the Dance Academy in Richmond a few years before writing his report. Like Dupin, Poe first formulates an explanation, before recounting how his observations made his explanatory hypothesis inevitable:

In attempting ourselves an explanation of the Automaton, we will, in the first place, endeavour to show how its operations are effected, and afterwards describe, as briefly as possible, the nature of the observations from which we have deduced our result. (Poe 1883: 298–299)

When Poe speaks of ‘deducing’ his explanation from ‘the nature of the observations’, he is, of course, not suggesting that a given set of observations logically entails a unique explanatory hypothesis, but only that they help make it psychologically compelling. Among a list of crucial facts and observations, Poe refers to the fact that Maelzel cancelled his performances during a period when his assistant, Schlumberger, was hospitalized, suggesting that Maelzel’s ‘automaton’ must really have been operated by Schlumberger; he concludes by appealing to the reader’s capacity to infer the best explanation: ‘The inferences from all this we leave, without farther comment, to the reader’ (Poe 1883: 310).
The discussion so far has focused on the ‘detective model’ of ratiocination as abductive inference to the best explanation; in this respect, our discussion is in broad agreement with previous accounts, notably by Nancy Harrowitz (1988), Ilkka Niiniluoto (1999), and Paul Grimstad (2005), all three of whom have linked Poe’s Dupin stories to inference to the best explanation as discussed by Peirce. Abduction, in the Peircean sense, however, has its primary domain of application in the sciences. Thus, Peirce writes:

Presumption, or more precisely abduction, furnished the reasoner with the problematic theory which induction verifies. Upon finding himself confronted with a phenomenon unlike what he would have expected under the circumstances he looks over its features and notices some remarkable character or relation among them […] so that a theory is suggested which would explain (that is, render necessary) that which is surprising in the phenomenon. (CP 2: 776)


A man must be downright crazy to deny that science has made many true discoveries. But every single item of scientific theory which stands established today has been due to Abduction. (CP 5: 172)33

However, there has been no systematic attempt so far in either the literature on Poe, or in the philosophical debate about scientific inference, to analyze Poe’s views on different inferential strategies specifically in relation to the sciences. In the remainder of this section, I want to show that Poe’s Eureka provides a connection between ‘detective-style’ ratiocination and abduction in the scientific context—where the latter is understood as encompassing both creative hypothesis-generation and inference to the best explanation. In order to avoid possible misunderstanding, it should be emphasized that, while there are several references in Peirce’s writings to other works by Poe (who is often regarded as one of Peirce’s favorite writers; see Brent 1998: 22) there is no direct evidence that Peirce read Eureka. The subsequent argument, therefore, is not meant to establish a direct causal-historical link between Poe’s Eureka and Peirce’s account of scientific inference; rather, it should be understood as an attempt to illuminate an affinity that exists between the two.34
Eureka begins with a polemic against traditional philosophy, for which Poe draws on the—somewhat awkward—literary device of a ‘letter from the future’, written in the year 2848. In this letter, which the narrator quotes at length, the writer satirizes and, in doing so, criticizes ‘Aries Tottle’ (Aristotle) and ‘Hogg’ (Bacon) as proponents of an outmoded deductivism and inductivism, respectively. Roughly in line with his criticism of deductive reasoning and the inductive method, discussed earlier in relation to his ‘tales of ratiocination’, Poe criticizes both schools of thinking ‘on account of their pompous and infatuate proscription of all other roads to Truth than the two narrow and crooked paths—the one of creeping and the other of crawling—to which, in their ignorant perversity, they have dared to confine the Soul’ (Eureka, p. 14). True scientific innovation fits neither the inductive nor the deductive model of scientific inference. In order to make his case, Poe helps himself to the—historically inaccurate, but nonetheless instructive—case of Kepler’s laws as the basis for Newton’s law of gravitation:

Newton deduced it from the laws of Kepler. Kepler admitted that these laws he guessed—[…] Yes!—these vital laws Kepler guessed—that is to say, he imagined them. Had he been asked to point out either the deductive or inductive route by which he attained them, his reply might have been—“I know nothing about routes—but I do know the machinery of the Universe. Here it is. I grasped it with my soul—I reached it through mere dint of intuition.” (Eureka, p. 15)

While it may be possible to maintain that intuition is ‘but the conviction resulting from deductions or inductions’ (Eureka, p. 15), for Poe this does not render intuition reducible in any straightforward way, since the posited underlying ‘processes were so shadowy as to have escaped his consciousness, eluded his reason, or bidden defiance to his capacity of expression’ (ibid.). The inductivists’ and deductivists’ insistence on a reduction that cannot, as a matter of principle, ever be achieved thus brings out their ideological stubbornness; thoroughgoing inductivism and deductivism, for Poe, are no more than a leap of faith.35
For Poe, what really manifests itself in such cases of theoretical creativity as Kepler’s uncovering of the laws of planetary motion, is a superior faculty of imagination, which sets itself apart from ‘mere’ deduction and induction. The point is not that deductive reasoning in science cannot do without inductively confirmed empirical contents, but rather that imagination provides a vital dimension that can neither be reduced to, nor compensated by, induction and deduction. As Poe puts it, ‘the repression of imagination was an evil not to be counterbalanced even by absolute certainty in the snail processes’—where ‘snail processes’ refers to the twin paths of inductive ‘creeping’ and deductive ‘crawling’. (Eureka, p. 14) Notwithstanding his evident worship of Kepler, Poe importantly does not simply equate imagination with an innate gift of genius. This is evident in a passage in Eureka, in which he expresses puzzlement at the failure of Leibniz to discover a unifying principle behind Newton’s Laws. Poe blames this failure on a lack of imagination on Leibniz’s part:

[I]t is almost impossible to fancy of Leibnitz that, having exhausted in his search the physical dominions, he would not have stepped at once, boldly and hopefully, amid his old familiar haunts in the kingdom of Metaphysics. Here, indeed, it is clear that he must have adventured in search of the treasure:—that he did not find it after all, was, perhaps, because his fairy guide, Imagination, was not sufficiently well-grown, or well-educated, to direct him aright. (Eureka, p. 36)

In The Murders in the Rue Morgue, this appeal to Samuel Taylor Coleridge’s distinction between ‘Fancy’ and ‘Imagination’ is even more explicit when the narrator asserts:

‘Between ingenuity and the analytic ability there exists a difference far greater, indeed, than that between the fancy and the imagination, but of a character very strictly analogous. It will be found, in fact, that the ingenious are always fancyful, and the truly imaginative never otherwise than analytic.’36

From his earlier comment on Leibniz, it is clear that Poe believes that imagination can be cultivated (‘grown’) within each of us—although, no doubt, to a varying degree, depending on the individual.37 Instead of the inculcation of method as a ‘mere mechanical skill of correctly combining the facts’ (Schlutz 2008: 204), what distinguishes the ‘truly imaginative’ is genuinely analytic ability as ‘a truly creative power’ (ibid.), capable of simultaneously entertaining different perspectives.38

Poe’s description of Kepler’s achievement as guesswork might seem to underestimate the degree to which imagination is a reliable guide in the process of generating hypotheses. However, it would be wrong to regard imagination as playing a merely auxiliary, or occasional, role in this process. Hypotheses themselves carry a great deal of weight for Poe, even if, in popular parlance, the term ‘hypothesis’ is associated with a lack of substance or significance. Responding to the objection that his cosmological scenario is “an hypothesis and nothing more”, Poe declares that ‘the word hypothesis is a ponderous sledge-hammer, grasped immediately, if not lifted, by all very diminutive thinkers, upon the first appearance of any proposition, wearing in any particular, the garb of a theory’ (Eureka, p. 48). What such loose talk about “mere” hypotheses overlooks is ‘that these conditions [of the distribution of matter, as inferred by Poe] themselves have been imposed upon me as necessities, in a train of ratiocination as rigorously logical as that which establishes any demonstration in Euclid’ (Eureka, p. 49; italics added).

In addition to its role in the generation of explanatory hypotheses, imagination has a crucial function in that it provides theoretical unification. This was already evident in Poe’s puzzlement at Leibniz’s failure to identify a unifying principle for Newton’s Laws, and it is a recurring theme in Poe’s writings on the proper goals of science. With respect to Newton’s law of gravity—‘a law whose admission as such enables us to account for nine-tenths of Universal phenomena—a law [… we] cannot help admitting as a law’—Poe laments that it has so far defied attempts at explanation. It is, he argues, a law ‘which, neither in its detail nor in its generality, has been found susceptible of explanation at all’ (Eureka, p. 49). If one could find a more fundamental principle that would render Newton’s law ‘at every point thoroughly explicable’—a principle that ‘would enable us to understand as satisfied conditions so miraculous […] as those involved in the relations of which Gravity tells us’—would it still be legitimate to brush off the theoretical formulation of such a principle as ‘the merest hypothesis’? Poe’s negative answer takes the form of a rhetorical question: ‘what rational being could so expose his fatuity as to call even this absolute hypothesis a hypothesis any longer’? For Poe, a vital function of the imagination—one that is repressed by the stubborn insistence that all scientific reasoning should be modeled after deductive reasoning or inductive inference—lies in its ability to move from a pretheoretical understanding, or ‘vision’, of a phenomenon to clearly articulable principles that bestow a sense of unity. First we find ourselves entertaining ‘unthoughtlike thoughts—soul-reveries rather than conclusions or even considerations of the intellect’, and it is through an exercise of our imagination alone that we stand a chance of ‘grasping the great principle’. Indeed, entertaining a vague pretheoretical vision of what a phenomenon “must be like” may even be a necessary precondition for one to arrive, in due course, at a well-ordered theoretical explanation. After all, ‘with such ideas—with such a vision of the marvellous complexity of Attraction [=the principle Poe is here concerned with] fairly in his mind’, it is now possible to ‘let any person competent of thought on such topics as these, set himself to the task of imagining a principle for the phenomena observed—a condition from which they sprang’ (Eureka, p. 33). Poe does not give a detailed discussion of how ‘principles’ and empirical ‘laws of nature’ relate to one another, except that the former offer a unified framework for the latter; instead, he merely alludes to ‘all that which the world now calls “principle”’ (Eureka, p. 48), as if this would settle the question. A glimpse of what he might mean may be gleaned from his discussion of simplicity, ‘which confronts us either at the end of an inductive journey from the phenomena […] or at the close of a deductive career’. In order for a principle to warrant adoption, it must be of utmost simplicity—simplicity itself being ‘a conclusion of so accurate a logicality that to dispute it would be the effort’ (Eureka, p. 50).

When Poe describes the inferential route by which he has reached his explanatory hypotheses as ‘a train of ratiocination as rigorously logical as that which establishes any demonstration in Euclid’ (Eureka, p. 49), he clearly expresses his conviction that he has identified a new mode of inference, at least on a par with—and, quite possibly, superior to—induction and deduction. What distinguishes this third—abductive—mode of inference is its psychological compellingness, which manifests itself in ‘an irresistible Intuition’ (Eureka, p. 34). If such inferences, guided by intuition and imagination, ‘according to the schools […] prove nothing’, then, as Poe sees it, this is so much the worse for traditional inductivism and deductivism. For truly ‘profound and cautiously discriminative human intellects’ there can be no doubt about the authoritativeness of such inferences: ‘To these intellects—as to my own—there is no mathematical demonstrating which could bring the least additional true proof’ of the truths that have been so inferred. Adopting an anti-skeptical tone (which is in itself significant, given his earlier endorsement of Pyrrhonian skepticism; see fn. 16), Poe asserts that ‘[f]or my part I am not so sure that I speak and see—I am not so sure that my heart beats and that my soul lives—of the rising of to-morrow’s sun—a probability that as yet lies in the Future—I do not pretend to be one thousandth part as sure—as I am’ of the fundamental truths derived from an imaginative exercise of intuition (Eureka, p. 35).

One might suspect that Poe’s defense of intuition and imagination is simply a version of the same old story, according to which ‘poetic’ knowledge has privileged access to truths that are necessarily unavailable to scientific understanding. But Poe’s take on the relation between intuition and understanding is different: Science would remain incomplete by its own standards, were it not for the role of imagination and intuition. Recall that it is a scientist, Kepler, not a poet that, for Poe, embodies the ideal of the ‘poet-mathematician’, whose intuitive understanding is matched by knowledge of the facts and an awareness of the constraints imposed on his reasoning by the world. Poe goes to some length to emphasize that gaining true understanding how the world works is not merely a matter of a (subjective) ‘aha’ experience—though, as the title, ‘Eureka’, suggests, such an experience may well accompany fundamental new insights—and he rejects popular alternative frameworks such as Mesmerism and Swedenborgianism (or, for that matter, any ‘other delicious ism of the same species’; Eureka, p. 36), whose ‘occasional fantastic efforts’ (ibid.) are marked by an unchecked credulity and a tendency to produce convoluted theoretical systems. In this sense, Poe locates his third mode of inference—a precursor, I have argued, to Peirce’s notion of abduction—precisely between the ‘snail processes’ (Eureka, p. 11) of traditional inductivism and deductivism, and the whimsical speculations of the various pseudo-scientific ‘isms’ that were prevalent in his time. As Poe sees it, for science to make progress, it needs people with a well-educated imagination—poet-mathematicians—who exercise their inferential powers in the appropriate way, guided by their explanatory intuitions.

4 Poe’s Coherentism and the Rejection of the Axiomatic Method

Much of the presentation in Eureka builds up towards the positing of ‘Simplicity’ as, in Poe’s terminology, a non-negotiable ‘assumption’, by which he essentially means a regulative principle that should guide our explanatory inferences. (See Eureka, p. 50.) At the same time, Poe sets great store by a second principle—Consistency—which, even more so than Simplicity, he regards as a touchstone of truth. Indeed, it seems not all that far-fetched to attribute to Poe, at least at an implicit level, a coherence theory of truth:

A thing is consistent in the ratio of its truth—true in the ratio of its consistency. A perfect consistency, I repeat, can be nothing but an absolute truth. (Eureka, p. 96)

Again, as in his argument for a third mode of (explanatory) inference, Poe places himself in direct opposition to the proponents of deductivism and inductivism, complaining that ‘in spite of the eternal prating of their savans about roads to Truth, none of them fell, even by accident, into what we now so distinctly perceive to be the broadest, the straightest, and most available of all mere roads—the great thoroughfare—the majestic highway of the Consistent’ (Eureka, p. 14). Consistency, along with simplicity, acts as a constraint on the otherwise free exercise of intuitive reasoning; however, Poe argues, its relation to truth has been wantonly ignored by previous philosophers: ‘Is it not wonderful that they should have failed to deduce from the works of God the vitally momentous consideration that a perfect consistency can be nothing but an absolute truth?’ (Eureka, p. 15). Perfect and complete consistency can never be attained by ‘the work of human constructiveness’ and the mere following of methodological rules: ‘in human constructions a particular cause has a particular effect; a particular intention brings to pass a particular object; but this is all; we see no reciprocity’ (Eureka, p. 88).39 It is only by attuning oneself to the totality of reciprocal relations—well beyond the reach of inductive or deductive inference—that the poet-mathematician can glimpse the true nature of the Universe. Contrasting ‘human ingenuity’ with ‘Divine adaptation’, Poe writes:

In this sense, of course e, perfection of plot is really, or practically unattainable—but only because it is a finite intelligence that constructs. The plots of God are perfect. The Universe is a plot of God. (Eureka, p. 89)

Consistency has a natural place in this vision of the Universe as the ultimate plot of a divine narrator. Truth no longer is a matter of correspondence between a proposition and reality, but is determined by a proposition’s coherence within the fabric of the overall plot.40 As Alexander Schlutz notes, there is a striking contrast between Poe’s and Coleridge’s conceptions of truth and our ways of accessing it: Whereas Coleridge ‘had argued for a common principle in which both induction and deduction, as well as materialist and idealist philosophies could be seen as united, Poe no longer bases his road to truth on correspondence to an essence, but on formal consistency’ (Schlutz 2008: 219).
Once the connection between consistency and truth is acknowledged, the road to scientific progress, as Poe sees it, is wide open:

How plain—how rapid our progress since the late announcement of this proposition! By its means investigation has been taken out of the hands of the ground-moles and given as a duty, rather than a task, to the true—to the only true thinkers—to the generally-educated men of ardent imagination. These latter—our Keplers—our Laplaces—‘speculate’—‘theorize’—these are the terms—can you not fancy the shout of scorn with which they would be received by our progenitors, were it possible for them to be looking over my shoulders as I write? The Keplers, I repeat, speculate—theorize—and their theories are merely corrected—reduced—sifted—cleared, little by little, of their chaff of inconsistency—until at length there stands apparent an unencumbered Consistency—a consistency which the most stolid admit—because it is a consistency—to be an absolute and unquestionable Truth. (Eureka, p. 15)

Poe’s suggestion appears to be that—provided scientific theorizing is properly constrained by considerations of consistency and simplicity—it will be possible not just to arrive at satisfactory explanations, but to achieve uniqueness in one’s search for a best explanation. Put crudely, the best inferred explanation ceases to be merely the ‘best of a bad lot’ (or of an otherwise incomplete set of alternatives) but turns out to be, in fact, the only—hence, according to Poe, the true—explanation:

[N]o conviction can be stronger, to my mind at least, than that with which I am impressed by an hypothesis that not only reconciles these conditions, with mathematical accuracy, and reduces them into a consistent and intelligible whole, but is at the same time the sole hypothesis, by means of which the human intellect has been ever enabled to account for them at all. (Eureka, p. 66)

Poe’s rejection of the monopoly of deductivism and inductivism occasionally exhibits all the hallmarks of a knee-jerk reaction—not least when he insults their proponents as ‘ground-moles’ or ‘wise-acres’ (Eureka, p. 15/11). However, Poe has principled reasons for his rejection, and he offers incisive criticism of inductivism and deductivism, respectively.
Against inductivism, Poe argues that empirical findings never constitute in and of themselves, without theoretical presuppositions, conclusive evidence for or against anything; in this sense, he acknowledges the theory-ladenness of observation, as when he speaks of ‘telescopic observation’, which always needs to be ‘guided by the laws of perspective’ (Eureka, p. 72). Another barrier that stands in the way of a naïve inductivism and blocks any simple inference from observed empirical findings to the confirmation of a unique theory is, of course, the problem of underdetermination. Poe is well aware of the near-impossibility, without reliance on external (non-empirical) considerations, of narrowing the set of available hypotheses:

To show that certain existing results—that certain established facts—may be, even mathematically, accounted for by the assumption of a certain hypothesis, is by no means to establish the hypothesis itself. In other words:—to show that certain data being given, a certain existing result might, or even must, have ensued, will fail to prove that this result did ensue, from the data, until such time as it shall be also shown that there are, and can be, no other data from which the result in question might equally have ensued. (Eureka, p. 66)

Poe’s rejection of pure inductivism, thus, does not merely express a general sense of impatience with the ‘Hog-ish’ (i.e. Baconian; cf. Eureka, p. 16) ‘snail processes’ (Eureka, p. 11) that are characteristic of the drudgery of everyday science; rather, he has principled reasons for emphasizing, time and again, the incompleteness of simple inductivism.
Against deductivism, Poe rejects the possibility of an axiomatic approach, regarding it as a ‘now well-understood fact that no truths are self-evident’ (Eureka, p. 9). Taking a broadly Aristotelian tradition to be his main target, he writes: ‘The simple truth is, that the Aristotelians erected their castles upon a basis far less reliable than air: for no such things as axioms ever existed or can possibly exist at all.’ (Eureka, p. 12) The basis for Poe’s rejection lies in his take on the criterion of conceivability:

[T]he principle of the Logical axiom—in other words, of an axiom in the abstract—is simply obviousness of relation. Now, it is clear, not only that what is obvious to one mind may not be obvious to another, but that what is obvious to one mind at one epoch, may be anything but obvious at another epoch, to the same mind. […] It is seen, then, that the axiomatic principle itself is susceptible of similar change. Being mutable, the “truths” which grow out of them are necessarily mutable too. (Eureka, pp. 50–51)

Outside the domain of logic, which deals in pure relations, axioms are neither self-evident, nor can they function as secure foundations of knowledge, since ‘[a]n axiom in any particular science other than Logic’ would be ‘merely a proposition announcing certain concrete relations which seem to be too obvious for dispute’, but which would not thereby, in virtue of its obviousness, be indubitable. Importantly, and in keeping with his rejection, tout court, of axioms as the basis of scientific inquiry, this proscription also extends to Poe’s cherished regulative idea (‘assumption’) of simplicity: ‘Simplicity, considered merely in itself is no axiom’ (Eureka, p. 50). If simplicity does play a special role in scientific inquiry, as Poe asserts when he calls it the ‘starting-point’ (ibid.) of all inferences, it does so not because it is an axiom in any substantive sense, but because it embodies a cognitive constraint that even the proponents of an axiomatic approach must implicitly acknowledge as methodologically prior:

It will now be readily understood that no axiomatic idea—no idea founded in the fluctuating principle, obviousness of relation—can possibly be so secure, so reliable a basis for any structure erected by the Reason, as that idea—(whatever it is, wherever we can find it, or if it be practicable to find it anywhere)—which is irrelative altogether […]. (Eureka, p. 51)

While there is a significant amount of tension in this passage—how could simplicity be regarded as altogether irrelative, if judgments about simplicity are typically (and, perhaps, by necessity) comparative?—it nonetheless indicates a desire on Poe’s part to demonstrate the consistency, and self-reflexivity, of the methodological part of his project.
Poe’s implicit endorsement of coherentism about truth, coupled with a rejection of both the idea there can be ‘neutral’ (and in this sense ‘foundational’) empirical evidence and of the axiomatic method, opens up logical space for abductive inference—constrained only by the twin demands of Simplicity and Consistency—to provide the true basis for any final scientific claims. Science—or, rather, ideal science as envisaged by Poe—welds together a creative imagination and analytical acumen; its practitioners must be guided by an imaginative understanding of the phenomena they investigate, and must be prepared to not be blocked by minor obstacles (as would inevitably be the case—at least this is what Poe implies—if one followed solely ‘the two narrow and crooked paths’ of induction and deduction). A concrete illustration of this idea is Poe’s defense, against its critics, of Laplace’s famous Nebular Theory which, Poe argues,

is by far too beautiful, indeed, not to possess Truth as its essentiality—and here I am very profoundly serious in what I say. In the revolution of the satellites of Uranus, there does appear something seemingly inconsistent with the assumptions of Laplace; but that one inconsistency can invalidate a theory constructed from a million of intricate consistencies, is a fancy fit only for the fantastic. (Eureka, p. 60)

There is a distinct echo of Poe’s earlier disdain for ‘scientific details’ (see Sect. 2), along with the familiar (implicit) distinction between ‘fancy’ and ‘imagination’ (see Sect. 3). As Poe sees it, through the proper exercise of a well-attuned imagination we can judge the Nebular Theory to be not just true simpliciter, but ‘beautifully true’—even when mechanical insistence on the rules of deductive and inductive logic, no matter how ingeniously applied, would deliver the opposite verdict. Not content with the scientific ideology of his time, Poe is thus nonetheless confident that ‘educated men of ardent imagination […]—our Keplers—our Laplaces’ will eventually carry the day, venturing ever deeper ‘into the vast halls where lay gleaming, untended, and hitherto untouched by mortal hand—unseen by mortal eye—the imperishable and priceless secrets of the Universe’ (Eureka, p. 16).

5 Conclusion

Disputes about the priority of ideas are as difficult to decide as they are typically pointless. In the present paper I have not tried to show that Poe anticipated Peirce’s theory of abductive inference, let alone that Peirce was directly influenced by Poe’s philosophical writings. After all, Peirce himself only gradually arrived at a clear understanding of how to delineate abduction from, say, merely hypothetical reasoning. By situating Poe in the historical context of debates about scientific inquiry and its methods, and by reconstructing his positions on core philosophical questions such as explanation, consistency, and truth, I have tried to convey a sense of Poe’s own historical moment. What I hope to have shown is that Poe contributed to, and actively engaged with, an intellectual climate that took discussions about the methodology of the sciences seriously, and that his own contributions are both original and, if not always clearly argued, then at least well-motivated. Moreover, Poe’s ruminations on observation, inference, and imagination, while anchored in his aesthetic and poetic outlook, should not be interpreted as a merely self-referential commentary on his literary oeuvre; instead, he aims to offer nothing less than a transformative vision of the scientific universe. Poe’s Eureka, especially, deserves to be recognized, not as an ingenious anticipation of later scientific results, but as an imaginative analysis of how to approach, in thought, the complex world of natural phenomena around us.


Quoted after Hovey (1995: 347).


Quoted in (Passage 1954: 191).


Ilkka Niiniluoto insists that, conversely, one must take ‘seriously Poe’s account [in his The Philosophy of Composition] of his poetic construction “with the precision and rigid consequence of a mathematical problem”’ (Niiniluoto 1999: 252).


On Poe’s assessment of the Bridgewater Treatises, see also Welsh (1991: 9).


For an in-depth discussion, see also Frank (2003), esp. Chapter 1 (‘“The Murders in the Rue Morgue”: Edgar Allan Poe’s Evolutionary Reverie’).


Quoted after Meyers (2000: 215).


Page numbers for Eureka refer to the 2004 edition by Levine and Levine; (see Poe 2004).


Quoted after Meyers (2000: 217).


Norstedt (1930) suggests that Poe anticipated certain of Einstein’s views (see especially p. 175). For a more recent (popular) comparison of Einstein’s, Newton’s, and Poe’s views on gravitation, see van Calmthout (1995). For a comparison of Arthur Eddington’s theories and Poe’s cosmology, see Hoagland (1939). Gelfert (2008) notes the similarity between Poe’s speculations about the origin of the physical universe and contemporary Big Bang theories.


Quoted after Meyers (2000: 217).


Quoted after Wagenknecht (1963: 219).


Quoted after Levine and Levine (2004: xv).


Quoted after Link (1968: 333–334).


Quoted after Beaver (1976: 397).


See Marginalia (No. CCIV); Griswold edition (1850), Vol. 3: p. 581.


John Tresch (2002: 121) nicely characterizes Poe’s Pyrrhonism about much of our everyday knowledge as follows: ‘Poe is fundamentally a skeptic about human knowledge. What currently passes for “reality” or “the world” is an imperfect tissue of conjectures and practices patched together as a makeshift version of the wider, ungraspable cosmos.’


See Weissberg (1991: 101).


The continuity between the Dupin stories and those stories that are told by a first-person narrator, who recounts reasoning processes under conditions of psychological pressure (A Descent into a Maelstrom, The Pit and the Pendulum, etc.), is important, for it is the latter conditions that often create the impulse for abductive reasoning (on this point, see Reichertz 1990: 309). See also footnote 21.


The Murders in the Rue Morgue; Griswold edition (1850), Vol. 1: p. 178.


Never mind that the exact phrase ‘Elementary, my dear Watson’ nowhere occurs in Conan Doyle’s writings.


For a dissenting voice, see Reichertz (1990), who argues that, at least in the case of Conan Doyle’s Sherlock Holmes, the inferences typically lack the Peircean element of guesswork and spontaneous (context-driven) hypothesis-generation that, for the later Peirce, is characteristic of abduction (as opposed to mere hypothetical reasoning); instead, Holmes always reasons from firmly established rules—even if these are only known to himself, not to others. See also footnote 18.


See Lipton (2004), esp. Chapter 4. ‘Loveliness’ and ‘likelihood’ may, of course, come apart: For example, many conspiracy theories are designed to explain irrelevant detail (making them ‘lovelier’, in the technical sense of ‘potentially explaining more facts’), but would require objectively unlikely conditions for them to be true.


Quoted after Niiniluoto (1999: 240), who also discusses historical precursors to Peirce.


The Murders in the Rue Morgue; Griswold edition (1850), Vol. 1: p. 180.


Paul Grimstad, in a lucid discussion of how Poe’s abductive preferences figure in his aesthetic choices, notes how in The Murders in the Rue Morgue ‘Poe acknowledges the novelty of this “peculiar” new generic logic in his own experimentation with the “rules” of storytelling, for his scholarly meditation [at the beginning of the story] likely disoriented those antebellum readers who expected only a murder intrigue’ (Grimstad 2005: 25).


Loc. cit., p. 193.


The Murders in the Rue Morgue; Griswold edition (1850), Vol. 1: p. 180.


The Purloined Letter; Griswold edition (1850), Vol. 1: p. 271.


Loc. cit., p. 275.


Loc. cit., p. 274.


Terry Martin (1989) claims that, when Dupin goes into a diatribe against Vidocq, he ‘impeaches himself, […] for it is, again, Dupin himself who loses sight of the matter as a whole—as an imaginative whole, that is’ (Martin 1989: 38). This accusation is unfounded, however, as is Martin’s claim that ‘Dupin’s analysis reduces the narrator to a curious machine whose inner workings are to be charted solely for the scientific interest of the activity’ (Martin 1989: 37). To be sure, Dupin is a complex character (perhaps shaped, as Martin puts it in one of his ad hominems, by ‘the unsoundness of his denial of the life of the body’; 1989: 41), but it is precisely his constitution as a ‘Bi-Part Soul’ (a ‘double Dupin—the creative and the resolvent’; Murders in the Rue Morgue, 1850: 183), which marks him out as an example of a ‘poet-mathematician’ in the making.


In line with this suggestion, Weissberg (1991: 100) classifies Maelzels Chess-Player as a ‘precursor’ to the ‘tales of ratiocination’.


Both citations follow Harrowitz (1988: 181).


A similar example of such an indirect affinity concerns the origins of the term ‘semiotics’, which makes its first appearance in an 1839 text by George Field (1777?–1854). While this text had almost certainly had no direct influence on Peirce, it nonetheless shows that, as John Deely puts it, ‘[t]he stuff of what is becoming the doctrine of signs was clearly part of the intellectual climate into which Peirce was born, and suffused the air he breathed’ (Deely 2003: 16). The same, I argue, may be said of Poe’s anticipation of a third (abductive) kind of inference.


Interestingly, Poe discusses this point twice, first as part of the introductory ‘letter from the future’, and shortly afterwards in his own voice. The first discussion openly mocks the reductionist proposal, imagining how a reductionist metaphysician might try to “cure” the ‘poor ignorant old man’ Kepler of his “misconception” that he, Kepler, had grasped the ‘machinery of the Universe’ by ‘mere dint of intuition’ (Eureka, p. 15): ‘How great a pity it is that some “moral philosopher” had not enlightened him about all this!’ The second discussion is more neutral, yet insists that, while we can only entertain an idea of intuition as the outcome of underlying ‘shadowy’ processes (of induction and deduction), nonetheless intuition manifests itself in ‘irresistible, although inexpressible’—and, by extension, irreducible—ways (Eureka, p. 22). I am grateful to one of the referees for pointing out this tension in Poe’s position regarding the reducibility, or irreducibility, of Intuition.


The Murders in the Rue Morgue; Griswold edition (1850), Vol. 1: p. 181.


The relationship between Poe and Coleridge’s thought has been characterized as one of ‘devious complexity’ (Kearns 2002: 10). On the one hand, as Floyd Stovall already argued, Poe and Coleridge agree on at least four points: ‘1. Imagination is the soul of poetry. 2. It harmonizes diverse matters and gives unity to variety. 3. It is analogous to the creative power of God. 4. Of two elements known and unlike it can create a third element different from either.’ (Stovall 1930: 111) On the other hand, Poe is keen to set himself apart from Coleridge, whom he faults for an overreliance on the traditional tools and methods of reason—much like the Prefect in The Purloined Letter and Vidocq in The Murders in the Rue Morgue, whose insistence on systematic observation blinds them to poetically attuned insight into the situation at large. As Christopher Kearns puts it, for Poe, ‘Coleridge’s overly profound perspective distorts the object of his investigation, blinding him to beauty or what might be called the poetic dimension of his subject’ (Kearns 2002: 11).


On this point see Schlutz (2008).


The importance of reciprocity in Poe’s account is highlighted by Matthew Taylor, who writes that, in Poe’s Universe, all things ‘are related not only in their common origin but also immediately in the present, across time and space, in the infinite interactions taking place between every atom in the universe as they fall back toward unity’ (Taylor 2007: 203).


See also Manning (1989).



An earlier version of this paper was presented at the 8th International History of Philosophy of Science conference (HOPOS 2010), held at the Central European University, Budapest, in June 2010; I am grateful to Cornelis Menke (Bielefeld), Michael Heidelberger (Tübingen), and Laura Snyder (St. Johns) for their insightful comments following my talk. Two anonymous reviewers for this journal provided exceptionally detailed and constructive comments. I would also like to thank my father, Hans-Dieter Gelfert, for pointing me to Poe’s Eureka, and Sophia Yap for bibliographic support.

Copyright information

© Springer Science+Business Media Dordrecht 2012