Penumbral Thinking Revisited: Metaphor in Legal Argumentation

J. Christopher Rideout*

The whole problem of the relation between parent and subsidiary corporations is one that is still enveloped in the mists of metaphor.

Judge Benjamin Cardozo1

Science, not to mention everyday thought, is influenced by metaphors. Why shouldn’t law be?

Judge Richard Posner2

[A]ll thinking . . . is metaphorical.

Robert Frost3

Starting with Aristotle, the place of metaphor in human language has long been double-edged. Metaphors are well recognized, often even recommended, and yet their use is controversial and at times criticized.4 Aristotle acknowledged that metaphor was a common feature of language,5 but he also regarded metaphor as a deviation from ordinary uses of language.6 Accordingly, he warned that metaphors should be used with a sense of their appropriateness for the occasion.7 On the other hand, Aristotle pointed out that anyone who achieved the appropriate use of metaphors might be a “master of metaphor“ and, hence, exhibit the signs of genius. Why? “[A] good metaphor implies an intuitive perception of the similarity in dissimilars.”8 Aristotle, like so many to follow, found a dual nature to metaphors: they are something to be mastered and used well—to good or even brilliant effect—but with caution and with an understanding of their peculiar nature.

This acknowledgement of the place for metaphor, accompanied on the other hand by the controversy over the appropriate use of metaphor, has extended to modern discussions of metaphor, including the place of metaphor in legal discourse. Judge Cardozo evoked a metaphor, “the mists of metaphor,” to complain about the figure’s effect on the logical clarity of “familial” relationships in corporate law.9 In the same opinion, he warned that although metaphors in the law can “liberate thought, they end often by enslaving it.”10 And yet Judge Posner, recognizing the place of metaphor in both science and in everyday thinking, notes that law can be no exception to metaphor’s influence.

In the modern jurisprudence of the United States Supreme Court, the controversy over the place of metaphor came directly into the spotlight in Griswold v. Connecticut.11 Justice Douglas, writing for the majority and relying in part on metaphoric reasoning for his argument, located a right to privacy in the penumbral area formed by emanations from specific guarantees in the Bill of Rights.12 Just as confidently, in his concurrence, Justice Harlan found that an inquiry into a right to privacy could not depend on any metaphorical radiations from the Bill of Rights.13 Justice Black, admonishing the Court for not sticking to the simple language of the Constitution, advised that he could “get nowhere in this case by talk about a constitutional ‘right of privacy’ as an emanation from one or more constitutional provisions.”14 And Justice Stewart flatly rejected the idea of finding any general right to privacy, penumbral or otherwise, whether in the text of the Constitution, in the Bill of Rights, or in any case decided by the Court.15 Justice Stewart was no doubt predisposed to his lack of sympathy for Justice Douglas’s penumbra metaphor. Three years earlier, he had rejected outright another metaphorical usage by the Court, asserting that the “wall of separation” between church and state was a phrase “nowhere to be found in the Constitution.”16 Metaphorical penumbras hardly fared any better than metaphorical walls with Justice Stewart.

The various opinions in Griswold represent a divide, then, regarding the place of metaphoric reasoning in legal argument. Justice Douglas employs metaphoric reasoning, while several of his fellow justices either avoid it or reject it. Because the case has been pivotal in establishing a right to privacy, this division about the Court’s use of metaphoric reasoning has extended into the many commentaries on the case.17

Somewhat ironically, at about the same time that a number of commentators began to analyze, and disagree about, Douglas’s use of the penumbra metaphor, another group of thinkers began reexamining the place of metaphor altogether, both in language and in thought. This latter group asserted not only that metaphor was a common and ordinary feature of language, but that metaphor was also central to human thinking—part of the very stuff of thought.18 For them, metaphor is conceptual, and it lies at the heart of how we understand the world. At the level of thought, we can hardly avoid using metaphor.

Why has Douglas’s penumbra metaphor created such a stir?19 Despite the contemporary views regarding metaphor, one possibility is that Douglas organized his opinion around a metaphor. As this article will discuss below, metaphors have a long history of being viewed with suspicion, or of being regarded as untrue. No doubt part of the criticism of Douglas’s metaphor has to do with this general distrust of metaphor, but that does not fully explain the uneasiness with the penumbra metaphor in Griswold. Other Supreme Court justices have built their opinions around a central organizing metaphorical concept without calling such attention to their metaphorical usage.20

Another possibility is that Douglas’s metaphor is overly novel—that he simply invented it, fancifully, to arrive at his desired result. But Douglas was not the first judge to use the penumbra metaphor, and historical and precedential evidence exists that Douglas’s penumbra metaphor accords with an established line of usage.21

Yet another possibility is that the objections are triggered not by the fact that Douglas used a metaphor in his reasoning, but rather by how he used that metaphor. The conceptual soundness of Douglas’s use of the penumbra metaphor in Griswold has been called into question by several earlier authors.22 It may be that metaphors have some unspoken “rules” for usage and that Douglas violated some of them. Aristotle hints at this when he indicates that mastery of metaphors is a sign of genius. But Aristotle never offers clear guidelines for achieving that mastery.

This article takes a step in that direction, by revisiting and further exploring some of the ways in which metaphors work. Starting with classical rhetoric, metaphors have traditionally been regarded as a matter of style, but recent research indicates that their role in our understanding of the world lies much deeper. Metaphors work through complex systems of correspondences, involving the mapping of conceptual domains onto each other and relying on the entailments that arise from that mapping. The “rules” for metaphorical usage, then, no doubt run deep as well—into this mapping and its coherence.

If such unspoken “rules” for metaphorical usage exist, then Douglas may have violated some of them with his penumbra metaphor—sound as his conceptual use of the metaphor may have been in other respects. Understanding the strengths, and perhaps the shortcomings, of the penumbra metaphor, then, can be a way of learning how to use metaphoric reasoning effectively in legal argument. That understanding of the penumbra metaphor must begin, however, with some theoretical understanding of how metaphors work.

I. The Great Divide: Metaphors as Deviation and Metaphors as Fundamental

Like many of the terms used in classical Greek rhetoric, the word “metaphor” is itself a metaphor,23 taken from the Greek verb “metapherein,” meaning “to transfer.”24 The transference involves giving one thing a name that belongs to something else.25 Empedocles, a 5th century B.C. philosopher and orator who taught the noted rhetorician Gorgias, may have been the first to become known for his conscious, deliberate use of metaphor.26 Aristotle took note of this usage, additionally commenting that Empedocles was the first to discover rhetoric.27 Argumentative use of metaphors goes back a long way.

In his own Rhetoric, Aristotle introduced metaphor in Book 3, the section of the treatise that deals with prose style. That he did so is fitting. For Aristotle, as for the rest of the classical commentators on rhetoric, metaphor was largely a matter of style. For the Rhetoric, Aristotle appears to have adopted the definition of metaphor that he also presented in the Poetics, his one other work that discusses the topic.28 Indeed, in Section 3.2 of the Rhetoric, Aristotle sets metaphor apart from some of the other stylistic devices as being poetic. But he also terms metaphor one of the most useful of the stylistic devices because of its ability to make what is said seem unusual, and thus more striking.29

After introducing metaphor in Section 3.2, Aristotle resumed his discussion in Section 3.12. Regarding prose style, Aristotle valued clarity above all else, but viewed distinction of style as also very important. For this reason, metaphor was, in his opinion, especially valuable because, of the poetic devices that lent both clarity and distinction, metaphor was the only poetic device available to the writer of prose.30

The other major classical commentators on rhetoric followed Aristotle’s lead, all of them discussing the stylistic uses of metaphor, including its uses in forensic persuasion:31 Cicero in De Oratore,32 Quintilian in Institutio Oratoria,33 and the anonymous author of the Rhetorica ad Herennium.34 Like Aristotle, these commentators gave stylistic reasons for the usefulness of metaphors, regarding them as helpful to explanation, description, or persuasion,35 or as being concise or subtle.36 Cicero noted that metaphors are pleasurable because they suggest a wholeness of insight, or a picture of the whole thing.37

Underlying all the classical discussions of metaphor, however, was the assumption that metaphor is a matter of language, not of thought.38 And, as a matter of language, metaphorical expressions were viewed as being novel or poetic—a way of taking the words for one concept outside of their ordinary conventional meaning in order to make a statement about a similar concept.39 The subsequent treatment of metaphor, not only among the classical rhetoricians but for the centuries to follow, extends this assumption,40 as does the consequent status accorded to metaphors: metaphors were regarded as a deviation from ordinary language. They were regarded as figures of speech, rather than as structures of thought. Both of these views have survived to the present and thus warrant further examination.


In the opening sentence of their ground-breaking book on metaphor, George Lakoff and Mark Johnson note how the classical assumptions about metaphor have continued to the present:41

Metaphor is for most people a device of the poetic imagination and the rhetorical flourish—a matter of extraordinary rather than ordinary language. Moreover, metaphor is typically viewed as characteristic of language alone, a matter of words rather than thought or action. For this reason, most people think they can get along perfectly well without metaphor.42

Aristotle first set the stage, most fully in his Poetics, for viewing metaphor as extraordinary or deviant. There, after defining metaphor as the giving of the name of one thing to another, he terms the transposed name allotrios—“the alien name.”43 The original name he terms “ordinary” or kurion—”current”—the name used generally, or in general usage.44 He then defines metaphor as deviant—para to kurion.45 Aristotle did not regard metaphor’s deviation from ordinary usage as a problem. On the contrary, the deviation lent metaphor a certain distinctiveness, as a consequence of its strangeness. Using a metaphor was a way of going beyond the prosaic in one’s style,46 and in this way the deviation itself could lend clarity.47

Aristotle’s definition of metaphor as a deviation also sets up what has later been called the substitution theory of metaphor: that the metaphor is a substitute for the ordinary or literal meaning.48 The substitution, and the displacement of meaning, is double. The metaphorical term is, first, a borrowed word and, second, a substitute for an ordinary way of expressing the same thing.49 The substitution theory of metaphor, and its opposition between the extraordinary or deviant meaning of a metaphor and the ordinary meaning for which it substitutes, has a long history. Quintilian, for example, described metaphors as “a purposeful deviation in sense or language from the ordinary simple form . . . [or] that which is poetically or rhetorically varied from the simple and immediately available means of expression.”50

The idea of metaphor as deviant survives well into the present. In his handbook of rhetorical terms, Richard Lanham, a respected modern commentator on rhetoric, defines metaphor as “[c]hanging a word from its literal meaning to one not properly applicable but analogous to it”51 and notes how commentators have treated metaphor as a special case in the relationship between words and meaning.52 Like Aristotle, he then uses this extraordinary status of metaphorical language to suggest its attractiveness, although in modern terms:

Perhaps it is metaphor’s intrinsic instability that has attracted so much recent attention: to appreciate the metaphoricality of a metaphor we must posit a nonmetaphorical, normative ‘reality’ against which to project the metaphorical transformation. The oscillation of the two reality states, normative and transformative, provides the essential bounded instability of a bistable illusion.53

Nevertheless, metaphor remains deviant and opposed to the “normal.” Indeed, even though metaphor’s deviation from ordinary language lent it a usefulness and a striking effect, this effect was regarded as only stylistic and had the consequence as well of relegating metaphor to a secondary linguistic status—a stylistic fiction, made in necessary opposition to any truthful statement about reality.

As Lakoff and Johnson point out, the idea that metaphors lack truthfulness has engendered a long history of resistance to them, starting with Plato and his Allegory of the Cave.54 In Plato’s allegory, truth is absolute and art a mere illusion. Metaphors, being poetic and extraordinary features of language, belonged to the realm of art. Aristotle, who praised metaphors for their ability to lend something of the poetic to prose, inadvertently also condemned them to the status of the secondary. This suspicion of metaphors continued into the modern period in Western civilization. Hobbes regarded metaphors as absurd and misleadingly emotional.55 Locke, relegating metaphors to the “deceitful” realm of rhetoric, noted that their use could insinuate “wrong ideas” and “mislead the judgment.”56 This continuing criticism of metaphor as a “deviation from [the] normal mode” of language did, however, open the way for a later reevaluation of its place.57

B. Classical Rhetorical Theory: Metaphors as Stylistic Ornament

For the classical rhetoricians, metaphor was additionally the dress of thought rather than the stuff of thought. That is, metaphor was a stylistic figure of speech, but only that.58 Again, Aristotle set the stage with his definition of metaphor as being a deviation from ordinary language usage. The idea of metaphor as a figure of speech is a corollary to the idea of metaphor as deviation. If metaphor is a deviation from ordinary usage, and if ordinary usage is proper usage, then metaphor is also a deviation from proper usage. This logical progression sets up the distinction between literal meanings and figurative meanings. Metaphor, being a deviation from proper usage, must always only be a figure of speech.59

As a figure of speech, rather than an ordinary statement about reality, metaphor also can only be an ornament. Quintilian called metaphor an “ornament of oratory”60 and, in forensic discourse, the “supreme ornament.”61 Later thinkers continued this classical tradition of viewing metaphor as mere linguistic ornament: for example, Hobbes,62 Samuel Johnson,63 and Lord Mansfield.64

Ricoeur further argues that metaphor’s treatment as an ornamental figure of speech follows from its grounding in a classical substitution theory of metaphor:

It is the idea of substitution that appears to bear the greatest consequences: for if the metaphorical term is really a substituted term, it carries no new information, since the absent term (if one exists) can be brought back in; and if there is no information conveyed, then metaphor has only an ornamental, decorative value. These two consequences of a purely substitutive theory characterize the treatment of metaphor in classical rhetoric.65

Because metaphor has so commonly been viewed as a figure of speech and a matter of style, many writing treatises follow the lead of Aristotle in commenting on its stylistic uses, including treatises on writing for the law. These commentaries almost always come with cautions as well. Henry Weihofen warns that metaphors, like all figures of speech, must be used with restraint.66 He also appears to have implicitly adopted Aristotle’s substitution view of metaphor—which conceives of metaphor as another way of stating the same thing. “Lawyers write for persons who can usually understand a metaphor. . . . But . . . [ask] whether the reader is likely to misunderstand your reference. If there is any danger that he may, better make it clearer.”67 “Clearer” would presumably be in ordinary, nonfigurative language. Even Bryan Garner is cautious in his advice about metaphors, citing them as a cause of purple prose and admitting that “[f]igures of speech can be extremely effective, but only when sparingly used.”68 This kind of advice is not without warrant, but nevertheless is better understood in the context of a more fully developed theory of metaphor, as will be demonstrated below.

C. Modern Theories of Metaphor: Thought, not Language

Modern theories of metaphor turn the classical hierarchy upside down. Metaphor, far from being a matter of language and style, is instead a matter of thought. And rather than belonging to the realm of the extraordinary, metaphors are in fact ordinary, a central part of our everyday lives. The very way in which we think, “our ordinary conceptual system . . . is fundamentally metaphorical in nature.”69 Lakoff calls the classical theory of metaphor just that—a theory.

As a cognitive scientist and a linguist, one asks: what are the generalizations governing the linguistic expressions referred to classically as ‘poetic metaphors?’ When this question is answered rigorously, the classical theory turns out to be false. The generalizations governing poetic metaphorical expressions are not in language, but in thought: they are general mappings across conceptual domains.70

Metaphor, in the view of Lakoff and others, offers a way to understand how we conceptualize the world. We do so by mapping some of the features of one domain onto another domain, what is often termed “crossdomain mapping.”71

Central to the modern theory of metaphor is the understanding that cross-domain mapping occurs at the level of thought, not language. This understanding, in turn, makes it important to distinguish between the metaphor itself—a conceptual matter, occurring at the level of thought— and the words that express the metaphor, better referred to as the “metaphorical expression.”72 Thus, according to this view, the language of the metaphor, the metaphorical expression on which the classical rhetoricians had focused their attention, is merely the surface manifestation of the underlying conceptual metaphor.73

Our very language reveals the underlying metaphorical structures that organize our thoughts. One often-cited example, relevant to law, is the conceptual metaphor ARGUMENT IS WAR. In this metaphor, the domain of WAR is mapped onto the domain of ARGUMENT, as revealed in our common ways of talking about argument:

Your claims are indefensible.
He attacked every weak point in my argument.
His criticisms were right on target.
I demolished his argument.
I’ve never won an argument with him.
You disagree? Okay, shoot!
If you use that strategy, he’ll wipe you out.
He shot down all of my arguments.74

According to modern theories of metaphor, these statements about argument as war are not only figures of speech or matters of linguistic expression, but also ways of understanding argument by means of the mapping of some of the features of war (physical attack, strategy, victory and defeat) onto it. The metaphorical structure and its mapping are fundamental to how we understand argument.

A modern view of metaphor is the obverse of classical theories of metaphor, then. Metaphors are ordinary, not extraordinary, and conceptual, not ornamental. The nuances of metaphor, understood in light of these two features of contemporary theory, may in turn provide helpful insights into how metaphors work in legal argumentation.

1. Modern Theory: Metaphors as Ordinary and Fundamental

The first feature of contemporary theory, that metaphors are ordinary, found full statement in the 1930s, with the work of I.A. Richards.75 Richards criticized the idea that metaphor was “something special and exceptional” in the use of language,76 declaring instead that metaphor was the “omnipresent principle of language.”77 In his view, it was not possible to use language without using metaphor. Richards had strong words regarding the traditional view of metaphor, which he regarded as superficial: 78

The traditional theory noticed only a few of the modes of metaphor; and limited its application of the term metaphor to a few of them only. And thereby it made metaphor seem to be a verbal matter, a shifting and displacement of words, whereas fundamentally it is a borrowing between and intercourse of thoughts, a transaction between contexts.79

In Richards’s view, meaning is always contextual, a claim that he used to debunk the notion of metaphor as a deviation from proper usage. The meaning of words derives not from their unitary attachment to an idea for which they stand, but rather from their attachment to the full context within which that idea resides. Any use of language, whether a metaphorical expression or otherwise, is an abridgement of that fuller context. Thus, constancy of meaning, a favored argument for those who viewed metaphor as a deviation from ordinary language use, is “never anything but the constancy of contexts.”80 Metaphors obey this same principle of contexts, and the constancy of meaning—being contextual, not lexical—applies to them perhaps even more than to words alone. Richards further derided as a form of magical thinking the idea that certain forms of language usage, but not others, would be proper, ordinary, and attached to the true meaning of those words.81 Far from being a divergent principle for how language works, metaphor is the constitutive form of language and the very principle by which language operates.82

Richards also based his understanding of metaphor on the idea of interaction: “when we use a metaphor we have two thoughts of different things active together and supported by a single word, or phrase, whose meaning is a resultant of their interaction.”83 Later theorists have adopted this final term, interaction, to characterize any modern theory of metaphor as an interaction theory.84 The interaction theory of metaphor is important in two ways. First, it moves metaphor away from the idea of comparison, and toward a view in which one system or domain is a “lens” for understanding another.85 And second, interaction theory replaces substitution theory as the account of how we arrive at metaphorical meaning.86 In doing so, interaction theory moves metaphor away from the realm of the extraordinary or deviant—something capable of being restated more literally (i.e., capable of being restated by that for which it is a substitution)—and into the realm of the cognitive.

To describe more fully the interaction of metaphor, Richards introduced terms for the two parts (or domains) of a metaphor. The principal subject, or original idea, of the metaphor he called the “tenor”;87 and the figure, or what is often regarded as the source of the metaphorical transference that carries over, he called the “vehicle.”88 In the metaphorical expression “the fog of war,” war is the tenor and fog is the vehicle. Understanding the nature of the relationship between these two terms, tenor and vehicle, is central to an interaction theory of metaphor. Richards noted that “in many of the most important uses of metaphor, the copresence of the vehicle and tenor results in a meaning (to be clearly distinguished from the tenor) which is not attainable without their interaction.” 89 The relationship between the tenor and the vehicle is larger than simply the meaning of the tenor plus that of the vehicle.90 And this relationship, or interaction, is part of what distinguishes metaphor from analogy. With metaphor, the whole can be greater than the sum of its parts, unlike for analogy, as Richards points out: “[T]here is no whole to any analogy, we use as much of it as we need; and, if we tactlessly take any analogy too far, we break it down. There are no such limits to the relations of tenor and vehicle. . . .”91

As mentioned earlier, this interaction between tenor and vehicle has more recently been called “cross-domain mapping.”92 It also results in what is referred to as the “entailment” of a metaphor.93 The mapping of some features of the vehicle onto the tenor “entails” certain metaphorical meanings. Black noted that “[i]n this ‘connection’ resides the secret and the mystery of metaphor. To speak of the ‘interaction’ of two thoughts ‘active together’ . . . is to use a metaphor emphasizing the dynamic aspects of a good reader’s response. . . .”94 Hence the metaphorical expression “the fog of war” entails, among other things, that in times of battle things can be hazy and indistinct, seeing clearly is difficult, and thus—by extension— sound judgment can be elusive. Berger calls the interaction, or mapping, of metaphors “cognitively efficient” and relies on it to compare metaphor to other cognitive schemes that make a concept meaningful: schemata, analogies, and narratives.95 Cross-domain mapping, and its entailments, is also central not only to a cognitive theory of metaphor, but also to an understanding of “penumbral reasoning,” the investigation of this article. For these reasons, we will return to the idea of entailment several more times below.

2. Modern Theory: Metaphors as Conceptual

The second central feature of the modern theory of metaphor is that metaphors are conceptual. This is the claim of the leading modern theorists on metaphor, Lakoff and Johnson:96

[M]etaphor is not just a matter of language, that is, of mere words. . . . [O]n the contrary, human thought processes are largely metaphorical. This is what we mean when we say that the human conceptual system is metaphorically structured and defined. Metaphors as linguistic expressions are possible precisely because there are metaphors in a person’s conceptual system.97

The very way in which we think depends on metaphors. If metaphors are fundamental to how we think, that is because they are deeply embedded into the way in which we understand the world. This embedding takes place through “long, constant, and unconscious experience.” 98 Part of the embedding occurs through our physical experience of the world, and part of it occurs through our shared cultural experience of the world.99 Lakoff and Mark Turner (another co-author) write that we could call metaphors “commonplace models” because they derive in part from our shared experience of the world, or we could call them “cultural models” because they derive in part from our cultural knowledge.100 Either way, as part of our conceptual system, they are cognitive models.

Many people, most notably Steven Winter, have observed that, if metaphor is central to how we reason in general, then it must also be central to legal reasoning.101 In a symposium exploring the role of metaphor in law, Ritchie writes that “metaphors in legal discourse are fundamental to the way we understand and use legal concepts” and are a “fundamental . . . way of forming and expressing our intellectual imagination in the context of legal reasoning and communication.”102 In the same symposium, Johnson writes that a cognitive science of the law, based on a modern theory of metaphor, has great potential for transforming legal theory.103 As this article will argue, it also has potential for helping to understand legal theory.104

As mentioned earlier, the conceptual system of metaphor relies upon an understanding of one domain in terms of another, an understanding that relies upon the mapping of the one onto the other.105 The mapping is based on conceptual correspondences, not superficial similarities or comparisons.106 These conceptual correspondences operate like other cognitive schemas: we use them largely unconsciously and, because of their power, effortlessly.107 And the conceptual mapping itself is the metaphor; the linguistic form of that mapping is the metaphorical expression.108 Lakoff uses the terms “target domain” and “source domain” in place of Richards’s “tenor” and “vehicle” as the terms for the two sites of the mapping.109 And as mentioned earlier, Lakoff, like others, notes the “entailment” that is associated with “cross-domain mapping.”

The mapping borrows the language of the source domain (the vehicle) but, more importantly, it borrows the inference patterns of the source domain as a way of understanding the target domain (the tenor).110 Each mapping—that is, each metaphor—relies upon a fixed pattern of conceptual correspondences that extends across the domains.111 The patterns for the mapping are often conventionalized.112 This conventionalized knowledge is usually rich and well established cognitively, which is why metaphors can seem so innate, powerful, and persuasive.113 The mappings also preserve the cognitive structure of the source domain (or, what Lakoff also terms the “image-schema” structure of the source domain), but in a way that is consistent with the inherent structure of the target domain.114 The mapping (and the metaphor) work only if there are correspondences between our knowledge of the highlighted features of the source domain and our knowledge of the corresponding typical features of the target domain.115

In all of these ways, metaphor obeys deeply engrained cognitive rules of usage, and thus not every expression will be a successful instance of metaphor. “The flames of passion” is a metaphor—and metaphorical expression—that works;116 “the green beans of passion” does not. The mapping, then, depends not only upon our embedded understandings of the world—whether experiential or cultural—but also upon a certain structural topology for the inference patterns.

The “internal systematicity“ of metaphorical mapping and its entailment is important for the coherence of the metaphor.117 Any mapping will induce a set of structural similarities between the two domains of the metaphor, and those similarities will give rise to potential entailments.118 The target domain, through those structural similarities, will fit at least some of the potential entailments. In this way, the entailment of any metaphor achieves a kind of metaphorical coherence. In addition to this internal metaphorical coherence, metaphors can also be coherent among each other: if they are all special cases of the same general or underlying metaphor; if they all map onto the same target domain; or if they have the same grounding in everyday experience or cultural knowledge.119 The coherence of metaphors is another example of their underlying cognitive rules of usage and, like entailments, will become important when we look at Justice Douglas’s penumbra metaphor.

II. Penumbral Reasoning

As mentioned at the beginning of this article, one of the most prominent metaphorical expressions in a modern United States Supreme Court opinion occurs in Griswold v. Connecticut, where Justice Douglas located a right to privacy in the penumbral area formed by emanations from the Bill of Rights.120 Justice Douglas’s invocation of the penumbra metaphor triggered controversy, some of it only lukewarm but some of it outwardly critical of his reasoning.121 The case even became the focus of an ideological dispute. Some liberals, who may have agreed with Douglas’s outcome, nevertheless were uncomfortable with his methodology.122 And some conservatives, outwardly critical of both the methodology and the outcome, regarded the opinion as an unprincipled example of resultsoriented jurisprudence.123

Other commentators have come to the defense of Justice Douglas’s reasoning, however, citing it as an example of “penumbral reasoning.”124 In their view, Justice Douglas used a form of reasoning that the court has used many times, including in some of its earliest cases. As Denning and Reynolds note, penumbral reasoning involves “reasoning-by-interpolation,” a matter of “drawing logical inferences by looking at relevant parts of the Constitution as a whole and their relationship to one another.”125 In their view, Griswold offers a standard example of penumbral reasoning:

Justice Douglas looked at various provisions of the Bill of Rights, including those that protect assembly, freedom from self-incrimination, and the right not to have troops quartered in one’s home. From his survey, he inferred that there was a common thread throughout that government could not intrude into the privacy of individuals absent fairly compelling circumstances.126

Douglas found the common thread in the overlap of penumbras from the enumerated rights.127

But Griswold is only one of many examples of penumbral reasoning by the Court. Denning and Reynolds point to McCulloch v.Maryland128 as perhaps the “quintessential” example of penumbral reasoning.129 In that case, Justice Marshall, like Justice Douglas, reasoned from the text and structure of the Constitution and supplemented his use of specific provisions with references to the overarching principles that bind the text and its structure together.130 Justice Marshall had to answer questions about the power of Congress and the power of the states in the absence of any specific textual provision in the Constitution. He used a form of penumbral reasoning similar to Justice Douglas’s: “Though the literal text of the Constitution provided no definitive answer . . . Marshall formulated a persuasive answer based on the whole structure of the Constitution, its text, the interrelatedness of its provisions, and constitutional ‘first principles.’ ”131

Penumbral reasoning can be thought of as a form of textualism,132 in that it relies on the text and structure of the Constitution as the basis for its inferences. And arguably, the Ninth Amendment sets the stage for penumbral reasoning when it states that the “enumeration in the Constitution, of certain rights, shall not be construed to deny or disparage others retained by the people.”133Without the type of reasoning-by-interpolation that characterizes penumbral reasoning, it would be difficult to avoid “denying or disparaging” rights that the Constitution does not explicitly enumerate.134

So why the controversy regarding Griswold? Part of the answer lies in the ideological reactions, noted above, both to the idea of a right to privacy as an unenumerated Constitutional right and to the Court’s move toward “results-oriented jurisprudence.” As a part of these political reactions, the metaphorical nature of the reasoning in Griswold also came under fire. But the metaphorical reasoning of the case is conceptually sound, although Justice Douglas may have skewed his expression of it. And the underlying conceptual structure of the penumbra metaphor in Griswold is the same as that of other judicial instances of penumbral reasoning. The penumbral reasoning in Griswold has a basis that is both conceptual and historical. Its conceptual basis lies in the fact that it is an instance of a spatial metaphor, and its historical basis lies in its judicial usage in constitutional law.

A. Penumbra as a Spatial Metaphor

Some of our most fundamental concepts are organized as spatial metaphors,135 and, as other authors have pointed out, the judicial use of penumbra is one of them.136 Spatial metaphors, like all conceptual metaphors, possess an inherent systematicity.137 For example, the spatial metaphor HAPPY IS UP138 reflects a coherent system that metaphorically organizes part of the way in which we think about the emotion of happiness. Individual expressions of this metaphor, such as “I’m feeling up,” “my spirits rose,” or “you’re in high spirits,” are part of a coherent metaphorical system, rather than isolated or random statements. In addition, an external systematicity exists among the various spatial metaphors. So the spatial metaphor GOOD IS UP gives UP an orientation to general well-being, and HAPPY IS UP is in turn externally coherent with this second metaphor GOOD IS UP.139 Furthermore, this coherence is deepseated, because the two spatial metaphors are rooted in our physical and cultural experience. They are not random, nor is the coherence of their metaphorical systematicity.

Spatial metaphors are, in turn, an instance of an even more fundamental metaphorical system—that of container metaphors.140 This relationship becomes important for understanding penumbral reasoning. Container metaphors are fundamental because they derive, in the most basic sense, from our physical experience of the world.141 As physical beings, we experience the world as outside of us, beyond the boundaries of our skin. In this sense, each of us is a container, with a surface that bounds the “container” that is our body and that defines what is “inside” of us and “outside” of us.

The conceptual system of container metaphors relies upon boundaries. In our experience of the rest of the world, we perceive physical boundaries, but even when we don’t—and, more importantly, when we are conceptualizing in terms of container metaphors—we still impose boundaries:

A clearing in the woods has something we can perceive as a natural boundary—the fuzzy area where the trees more or less stop and the clearing more or less begins. But even where there is no natural physical boundary that can be viewed as defining a container, we impose boundaries—marking off territory so that it has an inside and a bounding surface—whether a wall, a fence, or an abstract line or plane. There are few human instincts more basic than territoriality.142

Many of our notions of law are container metaphors, expressed in spatial terms.143 When I teach metaphor in a law and literature course, I ask the students to help me find examples of metaphorical expressions about the law that reveal the underlying metaphor of LAW AS A TERRITORY OR DEFINED SPACE. They have little trouble coming up with examples:

areas of the law / areas of law practice

“X” falls within that case / “X” falls within that statute

“X” falls within the four corners of the case

“X” falls within her rights

the case doesn’t cover “X’s” situation (a related metaphor)

the long arm of the law (that drags a defendant into its territory); the long arm statute

standing / the complainant lacks standing (in the territory of the law)

jurisdiction—the area in which a court has power (more literally, where the law is declared)

As these examples reveal, the concept of boundaries is central to spatial metaphors regarding the law.

Steven Winter has written about the relationship between container metaphors and the logical concept of categories, a concept essential not only to rationalist notions of logic144 but also to legal decision making.145 The entailment of the container metaphor, that things have boundaries, when applied conceptually to formal models for logic, leads to the inference that logical categories have boundaries, ones that are well defined.146 In one of the originating statements of formal logic, a thing is either P or not-P; a thing is either inside the category, or it is not. Winter notes, however, that this standard logical use of categories can be overly totalizing and limited by its all-or-nothing quality:147

The conventional view treats the notion of a category as an a priori conceptual structure that, consistent with the logic of P or not-P, is homogenous in content . . . . But, it turns out, categories are both varied in kind and complex in structure. On the standard view, categories are descriptive, definitional, and rigidly bounded. The empirical evidence, in contrast, presents a picture of categorization as an imaginative and dynamic process that is flexible in application and elastic in scope.148

The idea that categories are less rigidly defined accords with our own physical experience of the world. For example, we understand the states of day and night as categories. An event can occur “in” the day or “at” night. But our experience of these categories is that they lack rigidly defined boundaries, and we even give names to the transitional period when we “go” from one to the other: dawn and dusk. One could argue that the concepts of dawn and dusk are simply new categories, but even if they are, the need for them as categories indicates the slippage in the boundaries between day and night. And these new categories do not really solve anything, for the slippage continues—for example, more slippage occurs in asking when night ends and dawn begins.

If, in our experience, not all categories have rigid boundaries, and if, as Winter claims, the well-bounded logical categories that derive from the CONTAINER metaphor do not always work, then how can the idea of categories be useful as a logical concept? The answer is that we employ more than one type of structuring metaphor for logical categories. Winter, citing Lakoff, points out that we also commonly employ what he calls “radial categories”—categories that are more flexible and elastic.149 “A radial category consists of a central model or case with various extensions that, though related to the central case in some fashion, nevertheless cannot be generated by rule.”150 Perhaps radial categories are easier to understand by means of their underlying metaphorical schema: CENTER-PERIPHERY.151

Winter asks which is the better representative of the category “bachelor”: the Pope, James Bond, or Tarzan?152 Technically, as unmarried men, all three are bachelors. But under the CONTAINER metaphor, the Pope, and perhaps even Tarzan, do not quite seem to fit inside the boundaries. James Bond clearly does. Under the CENTER-PERIPHERY metaphorical schema, however, the category of “bachelor” makes more sense, with James Bond being near the center and the Pope more at the periphery. The “reach” of the “bachelor” category is more dynamic, capable of expansion or adjustment.

More recently, Mark Johnson suggested that the idea of radial categories, employing the CENTER-PERIPHERY metaphor, is important for law:

If a principle, rule, or law consists of a set of classically structured concepts, then that law would apply in a certain clear fashion solely to those situations where the defining conditions for the concepts were satisfied in our experience. If “thou shalt not murder,” and if you know the necessary and sufficient conditions that define murder, then your only problem in evaluating a proposed course of action is to determine precisely whether it involves an act of murder. If a certain act manifests the requisite properties that constitute murder, then it is prohibited, period. This classical objectivist view of categories, if it were true, would make law a neat little process of strict rule application. How a legal concept might grow without completely redefining the concept, and how legal judgment might change in a rational, stable manner, could never be explained using this view.153

Johnson proposes that we should, instead, think of most legal concepts as having complex radial structures. Like all metaphorical concepts, these concepts are not free-wheeling, but rather are grounded in our shared experience.154 Through their entailments, these concepts can be applied to novel cases, and in some instances can even expand the concept. But if legal concepts possess radial structures, they are not exceptional in doing so. “[M]any of our most basic concepts, from those for simple objects like cups and beds, all the way to abstract concepts in morality, politics, science, religion, and law, have complicated internal radial structures. . . .”155

The metaphorical use of penumbra by Justice Douglas and others seems to implicitly acknowledge this complex structure of legal categories. We think of law in spatial terms, a subsidiary of the logical categorization that is so central to legal reasoning. Certain activities are protected, for example, by lying “within” the First Amendment (association), the Fourth Amendment (protection from search and seizure), or the Fifth Amendment (freedom from forced surrender).156 Certain rights clearly lie within the center. But not all the categories for rights are so precisely defined or presented. For privacy, a right he was confident existed, Douglas had to look in the metaphorical periphery. Douglas may have expressed the concept in an incoherent fashion,157 as will be examined below. But conceptually, his use of the metaphor makes sense and, metaphorically, it is logical.158 It also has historical precedent in the U.S. judicial system, including in the reasoning of the Supreme Court.

B. Penumbra in Constitutional Law: Historical Usages

Judicial usage of the penumbra metaphor in U.S. courts goes back at least as far as the nineteenth century.159 In Montgomery v. Bevans,160 ruling on a claim to the property of a person who had disappeared and was presumed dead, the court referred to the time that had passed as “the penumbra, or death period, of seven years.”161 The ruling does not expound on the metaphorical usage, however.162

Justice Oliver Wendell Holmes made the most prominent early uses of penumbra as a metaphor.163 In each instance, his use of penumbra relied upon a spatial metaphor for the underlying conceptual structure. He first used the penumbra metaphor explicitly in a law review article, The Theory of Torts, in 1873.164Writing about the need to distinguish between two lines of cases that were taking the law in opposite directions,165 Holmes recommended drawing a line at some point. “The distinction between the groups [of cases], however, is philosophical, and it is better to have a line drawn somewhere in the penumbra between darkness and light, than to remain in uncertainty.”166 The penumbral area between the two lines of cases is spatial, its shadowy territory marked by the partial illumination that exists between full light and the shadow of darkness.167 Although he seems to prefer bright lines, Holmes resorted to the metaphorical penumbra as a way of dealing with the slippage between those lines.

Once he was on the bench, Holmes invoked the penumbra metaphor a number of times, beginning with a Massachusetts case in 1901. Here, again, he acknowledged that the categories of the law, in this case constitutional rules, lack boundaries that can be rigidly delineated:

Perhaps the reasoning of the cases has not always been as sound as the instinct which directed the decisions. It may be that sometimes it would have been as well not to attempt to make out that the judgment of the court was consistent with constitutional rules, if such rules were to be taken to have the exactness of mathematics. It may be that it would have been better to say definitely that constitutional rules, like those of the common law, end in a penumbra where the Legislature has a certain freedom in fixing lines, as has been recognized with regard to the police power.168

Holmes also acknowledged that these areas of penumbral uncertainty grant more freedom and flexibility to lawmakers.169

While on the U.S. Supreme Court, Holmes continued his use of the penumbra metaphor. Writing of the state boundaries of empowering jurisdictions, he noted the desirability of drawing clear categorical lines. “If this view be adopted we get rid of all questions of penumbra, of shadowy marches where it is difficult to decide whether the business extends to them. We have sharp lines drawn upon the fundamental consideration of the jurisdiction originating the right.”170 But Holmes also knew, and acknowledged, that such lines were not always possible. For example, in Schlesinger v. Wisconsin,171 Holmes wrote:

[W]hile I should not dream of asking where the line can be drawn, since the great body of the law consists in drawing such lines, yet when you realize that you are dealing with a matter of degree you must realize that reasonable men may differ widely as to the place where the line should fall . . . . But the law allows a penumbra to be embraced that goes beyond the outline of its object in order that the object may be secured.172

Here, again, Holmes made use of the penumbra as a spatial metaphor. This time he uses it to acknowledge that, although statutes may circumscribe an outline within which they fall, they may also reach beyond that outline into a penumbral area in order to secure their object.173 Holmes’s use of the penumbra metaphor, and the accompanying jurisprudence, was referred to after Schlesinger as his “penumbra doctrine.”174

Holmes’s next two uses of the penumbra metaphor apply the spatial metaphor to constitutional law. In 1928, Holmes noted the importance of penumbral shading between extremes:

The great ordinances of the Constitution do not establish and divide fields of black and white. Even the more specific of them are found to terminate in a penumbra shading gradually from one extreme to the other. . . . When we come to the fundamental distinctions it is still more obvious that they must be received with a certain latitude or our government could not go on.175

And in Olmstead v. United States,176 he suggested that certain provisions of the Bill of Rights create a penumbra of protective rights:

While I do not deny it, I am not prepared to say that the penumbra of the Fourth and Fifth Amendments covers the defendant, although I fully agree that Courts are apt to err by sticking too closely to the words of a law where those words import a policy that goes beyond them.177

In this particular case, he was reluctant to extend the penumbral protection of the Bill of Rights to the defendant, and so he decided the case on other grounds. But he admitted that the Fourth and Fifth Amendments contain penumbras that would cover other cases.178

Holmes was not the only federal judge to use the penumbra metaphor.179 Learned Hand used the metaphor eleven times, primarily to refer to the indistinct borders of words180 or concepts.181 Benjamin Cardozo, who replaced Holmes on the Supreme Court, had used the penumbra metaphor early in his career: “There is in all such controversies a penumbra where rigid formulas must fail. No test more definite can then be found than the discretion of the court, ‘to be carefully and guardedly exercised’ in furtherance of justice.”182 Once on the Supreme Court, Cardozo used the metaphor at least three more times. For example, in 1937, he invoked the metaphor in a fashion very much like Holmes, as a way of acknowledging that sharp categorical lines cannot always be drawn and that an area of uncertainty exists where judges have discretion:

The conception of the spending power advocated by Hamilton and strongly reinforced by Story has prevailed over that of Madison, which has not been lacking in adherents. Yet difficulties are left when the power is conceded. The line must still be drawn between one welfare and another, between particular and general. Where this shall be placed cannot be known through a formula in advance of the event. There is a middle ground or certainly a penumbra in which discretion is at large.183

For Cardozo, as for Holmes, the penumbra metaphor offered a way of reasoning about the slippage that often occurs in rigid categorical thinking.

Not only judges, but also legal commentators employed the penumbra metaphor. Karl Llewellyn used the metaphor to point out that the edges of the Constitution are not sharp but rather “penumbra-like.”184 H. L. A. Hart used the metaphor twice, both times in a manner that accords with the conceptual metaphor of CENTER-PERIPHERY.185 In Hart’s usage, “[t]here are penumbras of uncertainty, but also core areas where the distinctions are clear.”186 Henly notes that Hart’s use of the penumbra metaphor in the late 1950s and early 1960s indicates how accepted the metaphor had become.187

Before Griswold, Douglas used the penumbra metaphor eight times.188 In six of these cases, he used penumbra as a way of expressing his view that the outer reach of the law cannot be rigidly delineated. For example, he wrote in 1956 that “[t]he problem lies in the penumbra of Louisiana law, making all the more difficult a prediction as to what the Louisiana courts would hold,”189 and in 1957 that “[t]he Labor Management Relations Act expressly furnishes some substantive law. It points out what the parties may or may not do in certain situations. Other problems will lie in the penumbra of express statutory mandates.”190 In a third case, Douglas used the metaphor in a manner similar to Holmes’s use of penumbra in Olmstead, as an area that is protective—here protecting managerial practices from judicial scrutiny:191

Here it is plain that the stockholder and those who manage the corporation are completely and irrevocably opposed on a matter of corporate practice and policy. A trial may demonstrate that the stockholder is wrong and the management right. It may show a dispute that lies in the penumbra of business judgment, unaffected by fraud.192

In each of these cases, Douglas used the penumbra metaphor spatially and as a way of referring to a peripheral area or region.193 His use of the metaphor seems fixed by 1964, and the stage is set for Griswold. But in finally using the penumbra metaphor in that case, he did so in a way that he had already employed in prior cases, and his use followed several decades of judicial usage.

C. Penumbra in Griswold v. Connecticut

Griswold v. Connecticut began with the arrest of Estelle T. Griswold, the Executive Director of the Planned Parenthood League of Connecticut, and Dr. C. Lee Buxton, in 1961.194 Griswold and Buxton were arrested for giving medical advice to married couples who wished to avoid conception.195 Both were found guilty of violating two Connecticut statutes, one prohibiting the use of “any drug, medicinal article, or instrument for the purpose of preventing conception,”196 and the other punishing any person who assists another person in committing an offense.197 The judgments against them were affirmed by Connecticut’s Appellate Division of the Circuit Court and by its Supreme Court of Errors.198 Griswold and Buxton appealed to the U.S. Supreme Court, which granted them standing to raise the constitutional rights of the married people with whom they had a professional relationship.199

At the Supreme Court, the appellants offered three arguments for overturning their convictions. First, they argued that the statutes violated the Fourteenth Amendment by depriving them of their right to liberty and property without due process of law.200 Second, they argued that the statutes violated due process as an unwarranted invasion of privacy.201 And third, they argued that the statutes violated the First Amendment.202 Douglas, writing the opinion of the Court, found the second argument convincing.203 After discussing a line of cases in which the Court had previously found rights not explicitly mentioned in the Constitution or the Bill of Rights—for example, the right of association,204 Douglas then made his well-known statement regarding a right to privacy:

The foregoing cases suggest that specific guarantees in the Bill of Rights have penumbras, formed by emanations from those guarantees that help give them life and substance. . . . Various guarantees create zones of privacy. The right of association contained in the penumbra of the First Amendment is one, as we have seen. The Third Amendment in its prohibition against the quartering of soldiers “in any house” in time of peace without the consent of the owner is another facet of that privacy. The Fourth Amendment explicitly affirms the “right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures.” The Fifth Amendment in its Self-Incrimination Clause enables the citizen to create a zone of privacy which government may not force him to surrender to his detriment.205

Noting that the Constitution did not explicitly mention a right to privacy, Douglas used a combination of case precedent, analogy, and inferential reasoning—penumbral reasoning—to assert that one existed.

Douglas’s use of penumbral reasoning draws some attention to the concept of “penumbra” itself. Greely remarks that “penumbra” originates with Johannes Kepler, who invented the term in 1604 from the Latin words paene (“almost”) and umbra (“shadow”).206 Kepler had observed that during full lunar eclipses the earth produced a sharply defined deep shadow, or umbra, but that a less distinct shading surrounded the umbra—the penumbra.207 Recent definitions of “penumbra” embody this eclipse configuration: “the partially shaded outer region of a shadow (contrasted with the umbra) when the light comes from a source of some size,”208 or “the partially shaded outer region of the shadow cast by an opaque object.”209

Douglas’s use seems to rely on an eclipse configuration for the central spatial metaphor of his argument, the variant of the container metaphor known as CENTER-PERIPHERY. Not finding a right to privacy “within” the center of the Constitution or Bill of Rights, he has to look to the penumbral area, the shading, formed by specific guarantees. There, he finds it: the penumbra of the First Amendment, for example, “contains” a right of association. This right is spatial—a “zone of privacy.” Other Amendments create other zones of privacy as well. Consistent with the underlying metaphor, Douglas notes that the privacy rights derived from the First Amendment are “peripheral rights,”210 meaning that they radiate from the core rights of the Amendment. “Penumbra” seems synonymous with “peripheral.” A few sentences later, Douglas restates that “the First Amendment has a penumbra where privacy is protected from governmental intrusion.”211 Two pages later, he summarizes the spatial metaphor: “The present case, then, concerns a relationship lying within the zone of privacy created by several fundamental constitutional guarantees.”212

The entailment of Douglas’s penumbra metaphor is consistent with the historical usage of that entailment by prior federal and Supreme Court judges, as well as with Douglas’s own prior usage. The categories of the law, whether those formed by legal rules or those consisting of constitutional provisions or constitutional protections, are not always sharply delineated. In the penumbral area of less certainty, on the periphery, legal decision makers have more discretion. Stephen Kanter calls the penumbra metaphor used by Douglas the “Basic Core/Penumbra Model”213 and finds it unexceptional:

The central idea is that each textually explicit core right also has a protective shell, and a set of corollary or derivative rights. The Court describes these necessarily implied or peripheral rights as penumbras; that is, each core right casts a shadow or has a “penumbra” that contains the implied rights. This is a well-established approach to constitutional text, and hardly a novel or inadmissible proposition conjured out of the æther by Justice Douglas.214

Kanter also observes that the Court is acting “responsibly and legitimately when it finds penumbral rights that flow naturally from one of the textually explicit core rights. . . .”215

It would seem, then, that the penumbral reasoning in Griswold could be readily accepted. Many have argued for its conceptual integrity, whether in terms of metaphorical reasoning or Supreme Court jurisprudence,216 and it has roots in historical usage. But the penumbral reasoning in Griswold has not been fully accepted. One reason has to do with the larger ideological debate that surrounded the right to privacy.217 Another has to do with a general distrust of metaphor as a deviant form of expression. Douglas’s own colleagues evince some of this distrust. Justice Harlan, who was better convinced by the Fourteenth Amendment due process argument, rejects the metaphorical emanations that Douglas saw: “While the relevant inquiry may be aided by resort to one or more of the provisions of the Bill of Rights, it is not dependent on them or any of their radiations.”218 Harlan, however, relies on his own underlying spatial metaphor when summarizing his concurrence, using “infringement”: “the proper constitutional inquiry in this case is whether this Connecticut statute infringes the Due Process Clause of the Fourteenth Amendment. . . .”219 And in rejecting the radiations, he invokes his own, somewhat comical, metaphor for the stability of the Due Process Clause: “[t]he Due Process Clause of the Fourteenth Amendment stands, in my opinion, on its own bottom.”220 Justice Black, dissenting, implicitly conceptualizes privacy in spatial terms with his use of the verb “invade”: “I like my privacy as well as the next one, but I am nevertheless compelled to admit that government has a right to invade it unless prohibited by some specific constitutional provision.”221 Like Harlan, he rejects emanations from those provisions, and so presumably his interest in a right to privacy would originate elsewhere. Justice Stewart, as mentioned at the beginning of this article, seemingly rejects metaphors altogether.222

A third reason for Griswold’s difficulty, however, may have to do with Douglas’s own expression of the penumbra metaphor. The underlying spatial metaphor that he invokes seems conceptually sound, but his expression of it violates the structure of the underlying concept. In fact, it violates the rules of metaphorical coherence. Lakoff and Johnson write that metaphors have the power to define reality through “a coherent network of entailments that highlight some features . . . and hide others.”223 In his expression of the entailment of the penumbra metaphor, Douglas violates this coherence in two ways. First, he fails to render the entailments of the penumbra metaphor coherently. And second, he applies the metaphor (to his target domain—constitutional rights) in a way that renders it inconsistent with another underlying metaphor that he invokes in his expression, IDEAS ARE LIGHT SOURCES.224

First, the problems with Douglas’s expression of the entailment. To repeat a quote, Douglas wrote that “specific guarantees in the Bill of Rights have penumbras, formed by emanations from those guarantees that help give them life and substance.”225 To begin with, if the specific guarantees in the Bill of Rights have penumbras (a peripheral region), then those specific guarantees should lie in the umbra, or shaded portion (the core region next to which the penumbra is peripheral). But the shaded portion could not at the same time have emanations. The emanations would more properly come from the light source. Others have noted this problem: “Penumbras—areas of half light and half shadow—are not caused by the areas of full light which they abut. Rather the light of penumbras and the light of areas of full illumination have the same source.”226

I teach Griswold as a part of a metaphor component in one of my courses. My students regularly conclude that if the right to privacy is created by emanations from the light source (the specific guarantees of the Bill of Rights), then the right to privacy more properly lies, not in the penumbra, but in the corona (the outermost part of a luminous body, such as the sun).227 That is, the right to privacy, as an emanation from light, should lie in the light, not in the dark. In attempting to spell out the entailment by referring to the penumbra as being created by emanations from the same body that has the penumbras, Douglas has violated the spatial systematicity of the metaphor.

By the time Douglas was using the penumbra metaphor, it appears to have been fully conventionalized;228 so he would not have chosen a corona metaphor, which lacked a tradition of usage and violated the convention. He also may not have been aware that he was violating the underlying image-schema; he would not, at that time, have had this technical language for such a thing. Nevertheless, that image-schema seems to be very much present, and violated, for those who have thought about the metaphor.

In discussing the metaphor in class (and setting aside the problem of the corona), I also ask students to diagram the light-source, the blocking body (the counterpart to the moon, e.g., in a solar eclipse), and the umbra and penumbra, assigning the appropriate terms to the elements of the eclipse.229 Although they can almost always come up with a diagram, the students also admit that their diagrams do not seem to work (again, setting aside the problem that Douglas’s emanations have their own penumbra). The following are representative of diagrams that come the closest to working.

Although these schemas fill in the blanks for the penumbra image, they still fail to represent Douglas’s expression of the metaphor, because these schemas necessarily locate the right to privacy on the opposite side from its source, in the Bill of Rights or its specific guarantees. If the right to privacy is a penumbral right in the sense of a right peripheral to the core rights, then it should be located on the left side of the diagram, next to the region of those core rights. Also, as a right, privacy should be located affirmatively, next to other rights—and not next to the negative entities (violation of the statutes, government intrusion) located in the diagrams above.230 These types of problems always arise when my students and I discuss the diagrams.

Second, Douglas’s metaphorical expression violates not only the metaphor’s internal systematicity, but also its external coherence with another metaphor. This occurs because, by adding “emanations” to his expression of the entailment, Douglas has added a second metaphorical scheme: IDEAS ARE LIGHT SOURCES.231 This is a common metaphorical scheme, and it makes perfect sense for Douglas to use it. In his entailment, the ideas, specific guarantees of the Bill of Rights, are light sources, and thus they have emanations, resulting in the right to privacy. The right to privacy is itself another idea, and one that we associate with light. The problem lies with the cultural entailment of IDEAS ARE LIGHT SOURCES. Light is good, and darkness and shadow are not. In addition, we “find” things in the light, but we do not find things in darkness. So, if in Douglas’s telling, the right to privacy lies in the penumbral shading, this location violates the entailment of his second metaphorical schema. We would expect to find the object of his inquiry, the right to privacy, in the light, not in the shade.232

Douglas employed a metaphor that had become conventionalized and that could have worked, conceptually, for his argument. But his effort to spell out the entailment of the metaphor made it confusing and, more importantly, violated both its internal and external coherence—its systematicity. To go back to an earlier part of this article, metaphors are powerful conceptual devices, but they work most effectively when they rely upon correspondences between our knowledge of the features of the source domain that are being highlighted and our knowledge of typical features of the target domain.233 In the entailment of the penumbra metaphor in Griswold, as spelled out by Douglas, these correspondences confuse at best or, worse, fail to persuade. Additionally, when metaphors become conventionalized, they become even more powerful because they are unconscious and seem innate to our understanding of things. Douglas, however, made the entailment of his penumbra metaphor conscious, invoked a second conceptual metaphor, but then was unable to express the entailment in a way that respected the underlying metaphorical coherence.


Douglas’s penumbra metaphor in Griswold is a metaphor that should have worked, but for many it did not. Somewhere between “should have” and “did not” lies a lesson in the use of metaphor in legal argument. First of all, the “should have.”

Metaphors are not only important to legal thinking, they are commonplace—not just as the “dress” of legal thought but also as the “stuff” of it. To take umbrage234 at Douglas’s use of a metaphor in a judicial opinion is to fail to recognize that, at a fundamental conceptual level, all judicial thinking relies on metaphors in some fashion and that often these metaphors rise to a kind of expression that consciously guides the course of the law. Think of the “marketplace of ideas,” the “wall of separation” between church and state, or the “chilling effect” on constitutional freedoms.235 These are simply the overt examples. In much more fundamental and unconscious ways, given how deeply they are woven into both our thought and our language, metaphors help to create the terms for legal argument.

Second, metaphorical usages can also become conventionalized, and in that way become a part of legal thinking and legal argumentation. By the time of Griswold, the penumbra metaphor seems to have been generally accepted and useful. Both federal judges and legal scholars employed it. These conventionalized metaphorical usages become important because, over time and with continued use, they acquire a history of meaning. Metaphorical usages, like any other linguistic usage, achieve a stability and, through their conventionality, become a part of our ways of thinking.

And finally, metaphors can allow the law to change and evolve over time. Johnson asks, “[H]ow can law preserve its integrity over time, while managing to address the newly emerging circumstances that continually arise throughout our history?”236 One answer is through metaphoric reasoning and its capability for allowing innovative and imaginative thought. But, in response to those who might fear it or regard it as deviant, metaphoric reasoning is not uncontrolled and subjective. On the contrary, it is highly constrained—both by its internal systematicity and coherence and by the social contexts in which the meaning of its systems of correspondences is grounded.237 The penumbra metaphor, in its historical usage, allowed some flexibility to certain legal categories. But the very fact that it has been controversial in Griswold, where its usage was skewed, demonstrates how constrained metaphors actually are.

Now for the “did not.” A group of critics objects to Douglas’s penumbra metaphor.238 Metaphor, given its long treatment as being viewed as a figure of speech only or as being deviant, takes yet another beating. I suspect that in many instances, however, these objections are motivated, not by Douglas’s metaphor, but by his expression of the metaphor.239 Douglas made an unfortunate choice when he spelled out the entailment of his penumbra metaphor. Rather than elucidating what he meant, Douglas made it more confusing. The more closely one reads his metaphorical expression, the more puzzling its details become. Even worse, Douglas’s effort to elaborate on the entailment violated hidden “rules” of metaphorical coherence—both internal and external. The metaphor does not appear to work. But what fails, in my view, is his expression of the metaphor, not the underlying metaphor—itself a common instance of one kind of container metaphor. Griswold’s critics have not, to my knowledge, made that distinction.

Here, then, lies another lesson on the use of metaphor in legal argument—a lesson that goes back to Aristotle. Whether in law, poetry, or anywhere else, the expression of a metaphor must be done appropriately. That appropriate expression does not mean merely following stylistic rules, but rather expressing the metaphor in a way that conveys its correspondences, or mappings, coherently. If metaphors, conceptually, are constrained by their own systematicity, then the expression of those metaphors must in turn reveal, not muddle, that systematicity. It is not that metaphors are figures of speech and, like any poetic device, must be used cautiously and with restraint. Rather, metaphors, and their expression, can be powerful and persuasive, and by virtue of those features they must be expressed carefully.

Regarding metaphors in the law, as elsewhere, the challenge is not whether to use them, but rather how to understand their rich and complex nature. Metaphors are central to legal thinking, and, by adding flexibility, they help law accommodate complexity and change in human social experience. Unfortunately, for some, metaphoric thinking seems to violate the formal constraints of legal argument.240 To believe this, however, is to fail to understand how metaphors work, including the workings of their internal structures and constraints. Lawyers and legal writers—those who construct arguments in the law—would benefit from a fuller understanding:

[W]e need to understand first, that human rationality is not linear and criterial to begin with, but imaginative and adaptive (that is, involving metaphor, image-schemas, metonymies, and radial categories); second, that imaginative thought (including metaphor) is systematic and regular rather than arbitrary and unconstrained; and third, that innovation (whether via metaphor or otherwise) is itself a contingent and, therefore, highly constrained phenomenon. Legal metaphor presents neither the problems perceived by the realists nor those feared by conventional scholars.241

Thus understood, the realm of metaphor belongs not just to the poets, and not just to the scholars, but—through the law and the law’s central place in human social and political life—to everyone.

* © J. Christopher Rideout 2010. Professor of Lawyering Skills and Associate Director of Legal Writing, Seattle University School of Law.

1 Berkey v. Third Ave. Ry. Co., 155 N.E. 58, 61 (N.Y. 1926).

2 The Problems of Jurisprudence 395 ( Harv. U. Press 1990).

3 Education by Poetry, in Collected Poems, Prose, & Plays 717, 720 ( Lib. of America 1995).

4 Aristotle commented on metaphor both in his Rhetoric and in The Poetics. See Aristotle, On Rhetoric: A Theory of Civic Discourse (George A. Kennedy trans., Oxford U. Press 1991) [hereinafter Aristotle, On Rhetoric]; Aristotle, The Poetics (Ingram Bywater trans., The Modern Lib. 1954) [hereinafter Aristotle, The Poetics].

5 Aristotle, On Rhetoric, supra n. 4, at 222.

6 Aristotle, The Poetics, supra n. 4, at 254.

7 Aristotle, On Rhetoric, supra n. 4, at 222, 228.

8 Aristotle, The Poetics, supra n. 4, at 255.

9 Berkey, 155 N.E. at 61.

10 Id.

11 381 U.S. 479 (1965).

12 Id. at 484.

13 See id. at 500 (Harlan, J., concurring). Justice Harlan does offer a non-metaphorical analysis, employing the Due Process Clause of the Fourteenth Amendment. Id. See also Stephen Kanter, The Griswold Diagrams: Toward a Unified Theory of Constitutional Rights, 28 Cardozo L. Rev. 623, 668 (2006).

14 Griswold, 381 U.S. at 509–10 (Black, J., dissenting).

15 Id. at 530 (Stewart, J., dissenting).

16 Engel v. Vitale, 370 U.S. 421, 445–46 (1962) (Stewart, J., dissenting).

17 See e.g. Robert H. Bork, The Tempting of America: The Political Seduction of the Law 99 (Free Press 1990); Brannon P. Denning & Glenn Harlan Reynolds, Comfortably Penumbral, 77 B.U. L. Rev. 1089 (1997); John Hart Ely, The Wages of Crying Wolf: A Comment on Roe v. Wade, 82 Yale L.J. 920, 929 n. 69 (1973); David Luban, The Warren Court and the Concept of a Right, 34 Harv. Civ. Rights-Civ. Liberties L. Rev. 7 (1999); Glenn H. Reynolds, Penumbral Reasoning on the Right, 140 U. Pa. L. Rev. 1333 (1992); Mark Tushnet, Two Notes on the Jurisprudence of Privacy, 8 Const. Commentary 75, 75–80 (1991); Harry H. Wellington, Common Law Rules and Constitutional Double Standards: Some Notes on Adjudication, 83 Yale L.J. 221, 292–94 (1973).

18 Most prominent in this area of research is the work of George Lakoff and Mark Johnson. See Mark Johnson, The Body in the Mind: The Bodily Basis of Meaning, Imagination, and Reason (U. Chi. Press 1987); George Lakoff, Women, Fire, and Dangerous Things: What Categories Reveal About the Mind (U. Chi. Press 1987) [hereinafter Categories]; George Lakoff, The Contemporary Theory of Metaphor, in Metaphor and Thought 202 (Andrew Ortony ed., 2d ed., Cambridge U. Press 1993) [hereinafter Theory]; George Lakoff & Mark Johnson, Philosophy in the Flesh: The Embodied Mind and Its Challenge to Western Thought (Basic Books 1999) [hereinafter Philosophy]; George Lakoff & Mark Johnson, Metaphors We Live By (U. Chi. Press 1980) [hereinafter Metaphors]; George Lakoff & Mark Turner, More than Cool Reason: A Field Guide to Poetic Metaphor (U. Chi. Press 1989). Among the work in metaphoric reasoning in legal discourse, see the writings of Steven L. Winter, including A Clearing in the Forest: Law, Life, and Mind (U. Chi. Press 2001) [hereinafter Clearing]; Re-Embodying Law, 58 Mercer L. Rev. 869 (2007) [hereinafter Re-Embodying]; Transcendental Nonsense, Metaphoric Reasoning, and the Cognitive Stakes for Law, 137 U. Pa. L. Rev. 1105 (1989) [hereinafter Transcendental Nonsense]; and The Metaphor of Standing and the Problem of Self-Governance, 40 Stan. L. Rev. 1371 (1988). On metaphor as “the stuff of thought,” see Steven Pinker, The Stuff of Thought: Language as a Window into Human Nature 235–78 (Viking 2007).

19 The opinion also became the focus of an ideological controversy. For one of many discussions of this controversy, see Luban, supra n. 17. This article will deal with questions about Douglas’s metaphorical usage.

20 For a study of some of the more prominent ones, see Haig Bosmajian, Metaphor and Reason in Judicial Opinions (S. Ill. U. Press 1992).

21 Two earlier articles have traced this line of usage by the Court. See Henry T. Greely, A Footnote to “Penumbra” in Griswold v. Connecticut, 6 Const. Commentary 251 (1989); Burr Henly, “Penumbra”: The Roots of a Legal Metaphor, 15 Hastings Const. L.Q. 81 (1987). This article retraces the line of usage noted by Henly in particular. Henly offered a convincing argument that the penumbra metaphor fit within a well-established line of usage, but he left somewhat unanswered the questions that he raised about the rhetorical effectiveness of the metaphor. Turning to more recent work on the cognitive bases for metaphor, this article updates the Henly discussion and offers some answers regarding the rhetorical effectiveness of Douglas’s penumbra metaphor.

22 See Henly, supra n. 21, at 96–99; Richard D. Mohr, Mr. Justice Douglas at Sodom: Gays and Privacy, 18 Colum. Hum. Rights L. Rev. 43, 62 (1986); Luban, supra n. 17, at 32; Kanter, supra n. 13, at 640–711; Greely, supra n. 21, at 251.

23 See George A. Kennedy, A New History of Classical Rhetoric 59 (Princeton U. Press 1994).

24 The New Shorter Oxford English Dictionary 1756 (Lesley Brown ed., Oxford 1993).

25 See Aristotle, Poetics, supra n. 4, at 251.

26 See Kennedy, supra n. 23, at 18. Empedocles was active in Sicilian political life in the earlier 5th century B.C. and, in the process, acquired a reputation as a skilled orator.

27 Id. at 18. See also Paul Ricoeur, The Rule of Metaphor: Multi-Disciplinary Studies of the Creation of Meaning in Language 10 (Robert Czerny trans., U. Toronto Press 1977).

28 See Ricoeur, supra n. 27, at 13.

29 See Kennedy, supra n. 23, at 59.

30 See George A. Kennedy, Classical Rhetoric and Its Christian and Secular Tradition from Ancient to Modern Times 78 (U. North Carolina Press 1980).

31 For further discussion of the forensic uses of metaphor by the classical rhetoricians, see Michael H. Frost, Greco-Roman Analysis of Metaphoric Reasoning, in Introduction to Classical Legal Rhetoric 85 (Ashgate Publg. 2005).

32 Cicero, De Oratore (E.W. Sutton & H. Rackman trans., Harv. U. Press 1942).

33 Quintilian, Institutio Oratoria vol. 3, (H. E. Butler trans., Harv. U. Press 1921).

34 Rhetorica ad Herennium (H. Caplan trans., Harv. U. Press 1954).

35 See Frost, supra n. 31, at 85. Frost notes that the classical rhetoricians also valued them for their reliance on metaphoric reasoning, but I do not think he is employing the term in the same way that this article does. See id.

36 See id. at 88.

37 See id. at 88, 104.

38 See Lakoff, Theory, supra n. 18, at 202.

39 See id.

40 See id.

41 In his later article, Lakoff notes that the classical theory of metaphor, which he terms a theory only, became definitional for metaphor. See id.

42 Lakoff & Johnson, Metaphors, supra n. 18, at 3.

43 See Ricoeur, supra n. 27, at 18.

44 Id.

45 Id.

46 See id.

47 Aristotle, Poetics, supra n. 4, at 254.

48 See Max Black, Models and Metaphors: Studies in Language and Philosophy 31–32 (Cornell U. Press 1962).

49 See Ricoeur, supra n. 27, at 19. According to Max Black, when the metaphorical expression calls explicit attention to the comparison between its terms, as in a simile (using the word “like” or “as” to express the comparison), then that metaphorical expression exhibits a special form of the substitution theory of metaphor called the “comparison view.” See Black, supra n. 48, at 35–37. Black calls this view of metaphor the most popular one. Most people regard metaphor as a comparison between two things, and they have the easiest time expressing the comparison as a simile. See id.; see also Ricoeur, supra n. 27, at 85–87. Even in the comparison view, however, the metaphorical expression is a substitute for an ordinary way of stating the same thing.

50 Linda L. Berger, Of Metaphor, Metonymy, and Corporate Money: Rhetorical Choices in Supreme Court Decisions on Campaign Finance Regulation, 58 Mercer L. Rev. 949, 954 (2007) (quoting Quintilian, Institutio Oratoria, vol. 4, bk. I, 10–14).

51 Richard A. Lanham, A Handlist of Rhetorical Terms 100 (2d ed., U. Cal. Press 1991). See also Encyclopedia of Rhetoric 493 (Thomas O. Sloane ed., Oxford U. Press 2001) (defining metaphor as a “substitution of similarities”).

52 Lanham, supra n. 51, at 100-01.

53 Id. at 101 (internal emphasis omitted).

54 See Lakoff & Johnson, Metaphors, supra n. 18, at 190.

55 See id.

56 See id. at 190–91.

57 See I.A. Richards, The Philosophy of Rhetoric 90 (Oxford U. Press 1936). This article will turn to this re-evaluation in Section C below.

58 Technically, the classical rhetoricians would have called metaphor a trope (because it generally involves only a single word, unlike a figure, which would involve two or more words). But this distinction, first propounded by the Stoics, was lost on many classical commentators on rhetoric. See Kennedy, supra n. 23, at 91–92.

59 See Ricoeur, supra n. 27, at 19. The step appears to have been taken by later rhetoricians; no evidence exists that Aristotle himself explicitly tried to establish metaphor as a figure only.

60 Frost, supra n. 31, at 90 (quoting Quintilian, Institutio Oratoria, vol. 3, 253). I should note that Frost offers an argument that the classical rhetoricians saw metaphors as useful not only as stylistic figures, but for a form of metaphoric reasoning. Frost’s use of the term “metaphoric reasoning” seems different from the way it is used by contemporary theorists of metaphor.

61 Id. at 87 (quoting Quintilian, Institutio Oratoria, vol. 3, 199).

62 See David T. Ritchie, Who Is on the Outside Looking In, and What Do They See?: Metaphors of Exclusion in Legal Education, 58 Mercer L. Rev. 991 (2007) (citing Thomas Hobbes, Leviathan, Pt. I, Ch. 4 (John Plameatz ed., Prometheus Books 1962)).

63 See Richards, supra n. 57, at 93.

64 See Ritchie, supra n. 62, at 992.

65 Ricoeur, supra n. 27, at 20.

66 Henry Weihofen, Legal Writing Style 119 (2d ed., West 1980).

67 Id.

68 Bryan A. Garner, The Elements of Legal Style 196 (Oxford U. Press 1991).

69 Lakoff & Johnson, Metaphors, supra n. 18, at 3.

70 Lakoff, Theory, supra n. 18, at 202–03.

71 See id. at 203.

72 Id. See also I.A. Richards, who observed that metaphor in language is a matter of “explicit verbal metaphors” that are super-imposed upon underlying conceptual metaphors. Richards, supra n. 57, at 108–09.

73 See Lakoff, Theory, supra n. 18, at 244.

74 Lakoff & Johnson, Metaphors, supra n. 18, at 4.

75 See Richards, supra n. 57. On the centrality of Richards to modern metaphorical theory, see Berger, supra n. 50, at 955.

76 Richards, supra n. 57, at 90.

77 Id. at 92.

78 See id. at 98–112.

79 Id. at 94 (italics in original).

80 See Ricoeur, supra n. 27, at 78.

81 See id.

82 See id. at 80.

83 Richards, supra n. 57, at 93.

84 See e.g. Black, supra n. 48, at 38–44; Ricoeur, supra n. 27, at 65–66; James E.Murray, Understanding Law as Metaphor, 34 J. Leg. Educ. 714, 715 (1984). Murray describes yet another theory (a second modern theory), the “tension” theory. See id. at 715–16. Other theorists treat tension—between the two domains of the metaphor—as a feature of interaction theory. See e.g. Ricoeur, supra n. 27, at 247.

85 Berger, supra n. 50, at 955 (citing Black, supra n. 48, at 41).

86 See Ricoeur, supra n. 27, at 88.

87 Richards, supra n. 57, at 97.

88 Id. Richards himself describes the vehicle as a “figure,” but he quickly points out that he does not mean the word “figure” in the superficial sense in which it has traditionally been applied to metaphor. See id. at 97–98. As will be seen below, this application of the vehicle to the tenor has more recently been referred to as “cross-domain mapping.”

89 Id. at 100.

90 Id. at 134.

91 Id. at 133.

92 See Lakoff, Theory, supra n. 18, at 203.

93 See e.g. Lakoff & Johnson, Metaphors, supra n. 18, at 152.

94 Black, supra n. 48, at 39 (internal emphasis omitted).

95 Berger, supra n. 50, at 956.

96 See Lakoff & Johnson, Metaphors, supra n. 18, at 3.

97 Id. at 6 (internal emphasis omitted).

98 Berger, supra n. 50, at 956 (citing Lakoff & Johnson, Philosophy, supra n. 18, at 47).

99 See id.

100 Lakoff & Turner, supra n. 18, at 67.

101 See generallyWinter, Clearing, supra n. 18.

102 Ritchie, supra n. 62, at 993–94.

103 See Mark L. Johnson, Mind, Metaphor, Law, 58 Mercer L. Rev. 845 (2007).

104 For an application of cognitive metaphor theory to legal advocacy, see Linda L. Berger, What is the Sound of a Corporation Speaking? How the Cognitive Theory of Metaphor Can Help Lawyers Shape the Law, 2 J. ALWD 169 (2004).

105 See Lakoff, Theory, supra n. 18, at 204.

106 See id. at 207.

107 Lakoff & Turner, supra n. 18, at 65–66. It is possible, in most instances, to make them conscious if we so choose.

108 See Lakoff, Theory, supra n. 18, at 209.

109 Id. at 207. Black used his own terms, “principal” subject and “subsidiary” subject. See Black, supra n. 48, at 44. Ricoeur rejected any use of the term “principal subject”—or of “original idea”—for the tenor, citing the potential implication that the second term might then be a deviance from ordinary usage or meaning. Ricoeur, supra n. 27, at 80–81.

110 See Lakoff, Theory, supra n. 18, at 208. These inference patterns borrowed from the source domain are the heart of what could be called “metaphorical reasoning.” See Lakoff & Turner, supra n. 18, at 65.

111 See Lakoff, Theory, supra n. 18, at 210.

112 But not always so—hence the potential for new metaphors. See id. at 211.

113 See Lakoff & Turner, supra n. 18, at 60–63.

114 Lakoff, Theory, supra n. 18, at 215.

115 See Lakoff & Turner, supra n. 18, at 39.

116 The metaphorical expression itself is also highly conventionalized and familiar.

117 See Lakoff & Johnson, Metaphors, supra n. 18, at 91.

118 See id. at 149–52.

119 See Lakoff & Turner, supra n. 18, at 88–89.

120 Michael Smith would categorize Douglas’s “penumbra” metaphor as a “legal method” metaphor—part of the process of legal analysis by which Douglas created the doctrinal right to privacy. As such, it is distinct from a “doctrinal” metaphor—a legal rule—and a “stylistic” metaphor. See Michael R. Smith, Levels of Metaphor in Persuasive Legal Writing, 58 Mercer L. Rev. 919 (2007).

121 See Reynolds, supra n. 17, at 1336.

122 See e.g. Ely, supra n. 17, at 929; Wellington, supra n. 17, at 292–94.

123 Most prominently, Judge Robert Bork. See Bork, supra n. 17, at 187–221; Robert H. Bork, Neutral Principles and Some First Amendment Problems, 47 Ind. L.J. 1, 8 (1971).

124 See Reynolds, supra n. 17, at 1333; Denning & Reynolds, supra n. 17, at 1090.

125 Denning & Reynolds, supra n. 17, at 1092.

126 Id.

127 Reynolds, supra n. 17, at 1335.

128 17 U.S. 316 (1819).

129 Denning & Reynolds, supra n. 17, at 1093. See also Reynolds, supra n. 17, at 1336–37. Marshall did not use the word “penumbra” in his opinion, but, according to Denning and Reynolds, he did use a form of judicial reasoning that could also be called “penumbral,” in that he relied on inferences based on the textual structure of the Constitution.

130 Denning & Reynolds, supra n. 17, at 1093.

131 Id.

132 For a discussion of the various forms of textualism in Griswold, including Douglas’s “strong textualism,” see Anthony Musante, Black and White: What Law and Literature Can Tell Us About the Disparate Opinions in Griswold v. Connecticut, 85 Or. L. Rev. 853 (2006).

133 U.S. Const. amend. IX. See also Reynolds, supra n. 17, at 1345.

134 Reynolds, supra n. 17, at 1345.

135 See Lakoff & Johnson, Metaphors, supra n. 18, at 17.

136 See Henly, supra n. 21, at 83 (identifying Justice Douglas’s penumbra metaphor as the “most important and puzzling spatial metaphor in American constitutional law”).

137 See id.

138 Remember that, in contemporary metaphor theory, this kind of phrasing names the mapping of a source domain (“UP”) onto a target domain (“HAPPY”). See Lakoff, Theory, supra n. 18, at 207.

139 See Lakoff & Johnson, Metaphors, supra n. 18, at 18.

140 See id. at 29–32; Ritchie, supra n. 62, at 1003.

141 See Lakoff & Johnson, Metaphors, supra n. 18, at 29.

142 Id.

143 See Ritchie, supra n. 62, at 1003. Burr Henly also notes that spatial metaphors in the law make it possible to visualize change. “Doctrines can be extended, narrowed, expanded, and circumscribed. Branches are separated but not hermetically sealed off. Legal rules develop frontiers, cores, and cutting edges. Without splitting hairs, fine lines get drawn. Without using protractors, lawyers are able to define spheres of influence.” Henly, supra n. 21, at 82.

144 See Winter, Clearing, supra n. 18, at 62.

145 See id. at 101.

146 See id. at 62.

147 Id. at 64.

148 Id. at 69.

149 Id. at 71 (citing Lakoff, Categories, supra n. 18, at 79–114).

150 Id.

151 See id. at 75–76.

152 Id. at 85–86.

153 Johnson, supra n. 103, at 848.

154 Id. at 852.

155 Id. at 848–49.

156 See Griswold, 381 U.S. at 484.

157 By “incoherent,” I mean that in spelling out the details of the penumbra metaphor, his expression of these details violates the underlying structure of the metaphor.

158 As Steven Winter explains: Logic is not a truth about the world whose validity is coincidentally supported by metaphor; the ”truth” of logic is, rather, a function of human cognitive processes. As a product of human minds, logic is structured in just the same way that other human concepts are structured—that is, imagistically and metaphorically. Winter, Clearing, supra n. 18, at 58.

159 Greely, supra n. 21, at 251.

160 17 F. Cas. 628 (1871).

161 Montgomery, 17 F. Cas. at 632.

162 One other 19th Century use of the word was clearly not metaphorical. In 1898, a federal court ruled on a case involving a United States Marshal named Penumbra Kelly. See United States v. Kelly, 89 F. 946 (1898). See also Greely, supra n. 21, at 253 n. 10. This author cannot help but note that Kelly was, at the time, the United States Marshal for the District of Oregon, a peripheral territory.

163 Henly traces the development of the metaphor by collecting and describing its use by Justice Holmes, Justice Benjamin Cardozo, Justice Felix Frankfurter, Judge Learned Hand, Justice Douglas, and Prof. H.L.A. Hart. Henly, supra n. 21, at 83–92.

164 Oliver Wendell Holmes, Jr., The Theory of Torts, 7 Am. L. Rev. 652, 654 (1873), reprinted in 44 Harv. L. Rev. 773 (1931). See also Henly, supra n. 21, at 83–84.

165 I am compelled to point out the spatial metaphors that underlie this common way of talking about the law.

166 Holmes, supra n. 164, at 654, reprinted at 775.

167 See Henly, supra n. 21, at 84.

168 Danforth v. Groton Water Co., 178 Mass. 472, 476–477 (1901). See also Henly, supra n. 21, at 84.

169 See Henly, supra n. 21, at 84.

170 Hanover Star Milling Co. v. Metcalf, 240 U.S. 403, 426 (1916) (Holmes, J., concurring). See also Henly, supra n. 21, at 85.

171 270 U.S. 230 (1926).

172 Schlesinger, 270 U.S. at 241 (Holmes, J., dissenting). See also Henly, supra n. 21, at 85.

173 See Henly, supra n. 21, at 85.

174 See id. at 85 (citing Black, The “Penumbra Doctrine” in Prohibition Enforcement, 27 Ill. L. Rev. 511 (1933)).

175 Springer v. Govt. of the Philip., 277 U.S. 189, 209–10 (1928) (Holmes, J., dissenting). See also Henly, supra n. 21, at 86.

176 277 U.S. 438 (1928).

177 Olmstead, 277 U.S. at 469 (Holmes, J., dissenting). See also Henly, supra n. 21, at 87.

178 See Henly, supra n. 21, at 87.

179 See id. at 87–90; Greely, supra n. 21, at 256–59.

180 Commr. v. Ickelheimer, 132 F.2d 660, 662 (1943) (Hand, J., dissenting).

181 United States v. Dennis, 183 F.2d 201, 212 (1950). See also Greely, supra n. 21, at 256–58.

182 Norwegian Evangelical Free Church v. Milhauser, 252 N.Y. 186, 191 (1929) (internal citation omitted). Henly argues that this use is not accidental, given the great influence that Holmes had over Cardozo. See Henly, supra n. 21, at 88.

183 Helvering v. Davis, 301 U.S. 619, 640 (1937). See also Henly, supra n. 21, at 88.

184 Karl Llewellyn, The Constitution as an Institution, 34 Colum. L. Rev. 1, 26–28 (1934).

185 See H. L. A. Hart, The Concept of Law 119 (1961); H. L. A. Hart, Positivism and the Separation of Law and Morals, 71 Harv. L. Rev. 593, 606–14 (1958).

186 Henly, supra n. 21, at 91.

187 See id.

188 See Federal Power Commn. v. Texaco, Inc., 377 U.S. 33, 39 (1964); Wilson v. Schnettler, 365 U.S. 381, 392 n. 5 (1961); Panama Canal Co. v. Grace Line, 356 U.S. 309, 317 (1958); Textile Workers Union v. Lincoln Mills, 353 U.S. 448, 457 (1957); Smith v. Sperling, 354 U.S. 91, 97 (1957); General Box Co. v. United States, 351 U.S. 159, 169 (1956) (Douglas, J., dissenting); General Comm. of Adjustment v. Missouri-K-T Rwy., 320 U.S. 323, 336 (1943); United States v. Classic, 313 U.S. 299, 331 (1941) (Douglas, J., dissenting).

189 General Box Co., 351 U.S. at 169.

190 Lincoln Mills, 353 U.S. at 457.

191 See Henly, supra n. 21, at 91.

192 Smith, 354 U.S. at 97.

193 See Greely, supra n. 21, at 260.

194 Griswold, 381 U.S. at 480.

195 Id.

196 Id.

197 Id.

198 Id.

199 Id. at 481.

200 See Musante, supra n. 132, at 880.

201 See id.

202 See id.

203 See Griswold, 381 U.S. at 482.

204 See id. at 482–83.

205 Id. at 484.

206 Greely, supra n. 21, at 252.

207 Id. The less distinct shading produced the phenomenon of a partial lunar eclipse, when the moon was not fully in the center of the earth’s shadow. Id.

208 The New Shorter Oxford English Dictionary, supra n. 24, at 2152.

209 The New Oxford American Dictionary 1267 (Elizabeth J. Jewell & Frank

210 Griswold, 381 U.S. at 483.

211 Id.

212 Id. at 485.

213 Kanter, supra n. 13, at 627.

214 Id. at 626.

215 Id. at 630. Abate eds., Oxford U. Press 2001).

216 See Luban, supra n. 17, at 7.

217 See supra nn. 122-23 and accompanying text.

218 Griswold, 381 U.S. at 500 (Harlan, J., concurring).

219 Id. (emphasis added). “Infringe” means “to break into” or “to encroach upon”—that is, a breaking into or encroachment onto a territory or space; from the Latin IN + frangere, to break. See The New Shorter Oxford English Dictionary, supra n. 24, at 1365.

220 Griswold, 381 U.S. at 500 (Harlan, J., concurring).

221 Id. at 510 (Black, J., dissenting). Henly pointed out that even though Justice Black objected to the emanation view from which the right of privacy emerged, Justice Black nonetheless viewed the First Amendment in spatial terms, referring to his not being able “to stretch” it to cover the defendants’ conduct. See Henly, supra n. 21, at 92–93.

222 See supra n. 15 and accompanying text.

223 Lakoff & Johnson, Metaphors, supra n. 18, at 157–58.

224 SeeWinter, Clearing, supra n. 18, at 173. Winter also comments directly on the IDEAS ARE LIGHT SOURCES metaphor in Transcendental Nonsense, supra n. 18, at 1172 n. 212.

225 Griswold, 381 U.S. at 484.

226 Mohr, supra n. 22, at 62 (emphasis in original). See also Luban, supra n. 17, at 32.

227 Henly also notes this. See Henly, supra n. 21, at 99.

228 Greely calls the metaphor “dead” by the time Douglas used it. Greely, supra n. 21, at 260.

229 Others have tried to diagram the penumbra image-schema as well. See Henly, supra n. 21, at 96–99; Kanter, supra n. 13, at 640–711.

230 My students almost always attach a negative entity to the umbra, or deep shadow, for—I believe—cultural reasons, as noted in the next paragraph.

231 SeeWinter, Clearing, supra n. 18, at 173; Lakoff & Johnson, Metaphors, supra n. 18, at 48.

232 Greely also observes that shadows and darkness are “unfortunate connotations for affirmative rights.” Greely, supra n. 21, at 251.

233 See Lakoff & Turner, supra n. 18, at 39.

234 “Umbrage”: originally shade, or shadow; then a shadowy appearance; then suspicion; then, from the early seventeenth century, displeasure or offence. See The New Shorter Oxford English Dictionary, supra n. 24, at 3450.

235 For a discussion of these judicial metaphors and others, see Bosmajian, supra n. 20.

236 Johnson, supra n. 18, at 845.

237 See Winter, Re-Embodying, supra n. 18, at 897.

238 I will set aside the group whose objections to Griswold rest primarily on ideological grounds.

239 See supra nn. 223–33 and accompanying text.

240 SeeWinter, Re-Embodying, supra n. 18, at 872.

241 Id.