6.1 Introduction
On August 5, 1999, in the state of Washington, Kenneth Lee was stabbed in his apartment by Michael Crawford, who was accompanied by his wife, Sylvia Crawford. According to court records, Crawford believed that Lee had attempted to rape Sylvia at some earlier time, and the Crawfords were in Lee’s apartment to confront him. In an altercation, Lee was stabbed. When the police arrived, they arrested Crawford after Mirandizing both Crawford and Sylvia and interviewing them both twice. Because of Washington’s marital privilege which states, “a spouse or domestic partner shall not be examined for or against his or her spouse or domestic partner, without the consent of the spouse or domestic partner” (Washington Revised Code Annotated § 5.60.060(1)), Sylvia did not testify at Crawford’s trial. Instead, her tape-recorded police interviews were admitted (over objection by the defense) as (hearsay) evidence at trial. Such tape-recorded statements are considered hearsay, because there is no physical person – the person (Sylvia) who witnessed the event – to take the stand, swear an oath, testify in court, and be subjected to cross-examination.
Because the original witness, Sylvia, was not present and because Sylvia’s prior statements “asserted the truth of the matter asserted” (Federal Rules of Evidence, 2023), they are considered hearsay and are therefore objectionable. Further, because the tape-recording was played rather than having Sylvia take the stand and swear an oath to tell the truth, the defense had no opportunity to question the truth of her testimonial evidence in what Justice Scalia has called “the crucible of cross-examination” (Crawford v. Washington, 2003, p. 61). The Confrontation Clause of the Sixth Amendment (U.S. Const. amend. VI) promises the accused the right to confrontation. Because all those measures were not in place, Sylvia’s recorded interviews, though admitted through an exception in this case, were hearsay.Footnote 1
Hearsay is defined extensively throughout this chapter. For now, suffice it to say that hearsay is the process of repeating during in-court testimony a story that somebody else (not in court) told the witness on the stand and in which the repeated account is presented as substantive evidence of truth regarding the legal matter at hand. Hearsay is an account of an account, a story about an event that the witness on the stand did not themselves witness. In the case of Crawford v. Washington (here meaning the trial), Sylvia’s recorded testimony given to police in a prior context was hearsay because it was played outside of the context of sworn in-court testimony, rendering it impossible to cross-examine the truthfulness of the evidentiary statements in that recording. Though hearsay is typically inadmissible, there are many exceptions – occasions when hearsay can be admitted during a trial – such as was used in the Crawford trial. When these exceptions are applied, legal discourse and debate such as the ones analyzed here arise regarding when and how any hearsay may be admissible.
At trial, Crawford was found guilty. On appeal, the intermediate court upheld that verdict. The Washington Supreme Court upheld Crawford’s conviction after determining the hearsay evidence was both properly admitted and reliable. Ohio v. Roberts (1979) (hereafter Roberts), which was the US Supreme Court precedent when Crawford was tried, opined that hearsay may be admitted in line with the rights promised by the Sixth Amendment as long as the statements in question “bear adequate indicia of reliability” (p. 5) and “particularized guarantees of trustworthiness” (p. 66). Using a nine-factor test it had developed in line with this reasoning, the Washington Supreme Court upheld Crawford’s conviction, agreeing with the trial court that Crawford and Sylvia’s statements were “virtually identical” and “interlocking” (Crawford v. Washington, 2003, p. 66) and thus reliable and trustworthy enough to admit as hearsay. Crawford then appealed to the US Supreme Court. The US Supreme Court agreed to hear the case, citing the so-called Confrontation Clause of the Sixth Amendment when writing that “the question presented is whether this procedure complied with the Sixth Amendment’s guarantee that, ‘[i]n all criminal prosecutions, the accused shall enjoy the right … to be confronted with the witnesses against him’” (Crawford v. Washington, 2003, p. 38).
Ultimately, hearsay is about language – when and how language can be/is trustworthy. The law is constituted in and of language, whether it be documents, statements, transcripts, testimony, precedent, or statute. Hearsay is a moment in the law where we hear the law talk about, discuss, and explain language in its own words. In legal precedents and treatises about hearsay, we hear the law asking many of the same questions about language that rhetoric does: How do we recognize an accurate account? What contextual parameters create a (perfectly accurate) linguistic account of an event? What role does the witness/speaker play in the creation of an account? And, ultimately, can we ever really trust language? Because law and rhetoric come up with wildly different answers to these questions, it’s important to dig into legal rhetoric and to understand how law’s view of language impacts all of us. In what follows, I argue that one of the things that happens when a statement is evaluated as potential hearsay is that the rhetorical consequences to the ethos of the speaker are ignored and scrubbed from the statement. Instead, the truth of the statement is linked to the circumstances under which it was spoken and the contexts to which the statement refers, positioning logos as the key, and, in the end, nearly only thing on which truth rests. In his Essay Concerning Humane Understanding, (hereafter, Essay) John Locke (Reference Locke1690) asserts that truth is objective, resting on the empirical, that which can be confirmed by material proof, or what I am referring to as logos. In hearsay legal discourse, the law sees truth in the same way. In this chapter, I put Locke, Aristotle, and the law into conversation to better understand the consequences that such a fetishization of logos has for ethos and ultimately the speaker in legal contexts where the stakes are high.
In the traditional, Aristotelian rhetorical structure, three rhetorical components operate together in the process of making nearly any argument: ethos, pathos, and logos (Aristotle & Kennedy, Reference Aristotle and Kennedy2006). For Aristotle (and this is by no means a definitive or comprehensive description), the appeals to ethos, pathos, and logos have to do with the relationship between an audience and a speaker discoursing about a particular topic. These so-called appeals are used by the speaker to persuade the audience of a particular truth and win the argument. For Aristotle, ethos is a feature of the text itself; ethos functions within the speech or document. Ethos is formed and performed through the discursive structure of the argument. Over time, the notion of ethos has been developed to encompass a sense of the credibility of the speaker; the reasons why they are believable; their standing in the community; their potential to speak the truth; and their trustworthiness in word and deed (Carlo, Reference Carlo2020; Hyde, Reference Hyde2004; Sullivan, Reference Sullivan1993). In other words, over time, ethos has been linked instead to the speaker’s credibility and character. Logos has to do with logic and proof, the argument’s evidence and logical structure. Pathos is the emotional thrust of the argument, its ability to resonate with and move the interlocutor.
For Aristotle, in order for an argument to be persuasive, it must leverage these three appeals in a more-or-less balanced way depending on the needs, opportunities, and constraints of the situation, and therefore perhaps focusing on one aspect more than the others, but still operationalizing the three. This typical relationship between ethos, pathos, and logos, or at least typical in rhetorical theory, is disrupted in evaluations of hearsay evidence. Pathos is disregarded nearly completely, and ethos and logos are sent into a dance in which logos is assumed to be objective and far superior to and more trustworthy than anything so closely related to subjectivity, which is situated in (and linked to) the ethos of the speaker.
In hearsay legal debates, the disruption of ethos/pathos/logos appears in discussions about the relationship between a statement and a speaker and how to identify an objectively true account, which the law believes exists, and which would be admissible hearsay. A key argument made in this chapter is that the legal reasoning circulated in hearsay statutory and case law works to disrupt the typical rhetorical relationship between a statement and a speaker, in which a speaker makes a statement for an audience with particular rhetorical goals, embedded in a particular context, stocked with constraints, actors, and actions. This disruption happens whether the hearsay statement is deemed inadmissible or admissible. The other argument in this chapter shows that this legal line of reasoning, which overemphasizes logos, is affiliated with Locke’s notions about truth and empiricism. Indeed, Locke had great influence on Anglo-American law in its early stages of development.
I will focus my discussion of Locke on the ideas and concepts that he presents in his Essay (1690), where he places high value on sensory experiential learning and the physical world that can be experienced with the senses. Language, for Locke, comes after sensing and is thus essentially less reliable. According to Locke, language is mere representation, only ever pointing, always secondarily, at the empirical. In his description, sensory experience and the empirical are truth’s only sources, which language is always lacking. Language is wholly removed from the empirical world and merely indexes that which is empirically grounded in knowledge, rooted in experience. The way to control and manage the unruliness of language, in Locke’s philosophy, is empiricism, a turning to the real world of experience and circumstances to locate truth, which puts pressure on ethos.
In this chapter, I trace and describe the ever-evolving (but remarkably consistent) historical, legal positioning of ethos and logos by analyzing hearsay statute, historical treatise, and precedential structure. I focus a significant part of this chapter on the US Supreme Court’s decision in Crawford v. Washington (2003), which provides us with a close look into the language ideology and rhetorical legal structures that are at the center of questions surrounding the trustworthiness of language. I also analyze the Lockean language ideology circulating in Ohio v. Roberts (1980) (hereafter, Roberts) and Davis v. Washington (2006), which the Court consolidated with Hammon v. Indiana (hereafter, Davis/Hammon). What it means to be a witness; what it means for an utterance to be reliable and trustworthy; what it means to give testimony – all of these concerns about language have deep roots in the crucial period of high modern thought during which Locke wrote.
In what follows, I demonstrate that Lockean thinking about language, empiricism, and truth is present and circulating in hearsay law. I use the discussion of high modern language ideology to argue that in hearsay legal discourse, the rhetorical structure of ethos/pathos/logos is altered to apply and accommodate the empiricism that is linked to objective truth for both Locke and law. This accommodation delinks the speaker from their statement, reducing ethos to nearly nothing and extending the role of logos to an extreme degree. The process of evaluating hearsay in judicial opinions renders the circumstances surrounding a statement into proof (logos). This process has the effect of pushing aside ethos, because it is always potentially untrustworthy and would need to be tested via cross-examination. By delinking the statement from the speaker through a process that exalts the circumstances surrounding the production of a statement, the statement is transformed into something that is indelibly linked to circumstances and therefore infinitely repeatable without (it is assumed) altering the statement or its relationship to empirical truth. In the law there is a “translation of people and events into legal categories so that they can be used strategically in a struggle for the dominant interpretation” (Mertz, Reference Mertz2007, p. 159). In the case of hearsay, the statement and speaker are hypostatized in isolation from each other, flattened and translated into legal artifacts. This odd diminished and diminishing account of the speaker and their statement relies on Lockean ideas about language, empiricism, and context that have been woven into hearsay law for over 400 years.
6.2 Creating a High Modern Rule against Hearsay
The story of hearsay is long, beginning in England in the late 1600s (Thompson v. Trevanion, 1694). Our modern version of hearsay is closely tied to its historical developments through precedent and the continued reference of legal texts that are used as though there are direct, clean lines between historical precedent and modern-day applications of law. Crawford v. Washington (2003) and Davis/Hammon (Davis v. Washington, 2006) are riddled with historical legal arguments from the early 1700s. Hearsay is a product of case law, organized systematically, indexed, and handed down through the ages in judicial opinion and historical legal texts such as the treatises, abridgements of philosophical texts, and abstracts of legal and philosophical texts, which present legal process, procedure, and expectations. Recent US Supreme Court opinions reference legal rulings and treatises from across the seventeenth, eighteenth, and nineteenth centuries.
Hearsay erupts in a time when the relationship between law and community was changing. In the early seventeenth century, most members of a jury were likely to already know each other and have knowledge of (and even discussed) what had happened in their communities. Communities were small enough for there to be more intimate knowledge of the goings on of its members (Landsman, Reference Landsman1992; Langbein, Reference Langbein1996). Hearsay rules were unnecessary, even in jury trials, because the jurors likely already had prior knowledge of and were connected more closely to all aspects of the legal situation and actors – the accused, witnesses, events, and so forth (Landsman, Reference Landsman1992; Langbein, Reference Langbein1996). As communities grew and diversified, the intimate aspects of community knowledge began to break down. Changes in the language theories and philosophies were also evolving. Out of these changes grew a need to control statements in legal settings. Witnesses were expected to give accounts based on what they knew (had seen) from first-hand interaction with the empirically grounded world (Landsman, Reference Landsman1992; Langbein, Reference Langbein1996). They were not allowed to give accounts of what they had heard second-hand.
At the same time that laws and communities were changing, so were views about language. During the eighteenth century, language became a “mere” medium, only a conduit for information and experience, functioning through mimicry and representation (cf. Foucault, Reference Foucault2001). Locke’s Essay (Reference Locke1690) takes up questions of human knowledge, experience, and reasoning, considering the ways humans learn from sensory experience and learn to reason and use language both individually and socially. Locke argues that the knowledge of things and words was not innate but rather developed through sensory experience. He writes:
The Idea’s [sic] themselves, about which the Proposition is, are not born with [individuals], no more than their Names, but got afterwards. So, that in all propositions that are assented to, at first hearing the Terms of the Proposition, their standing for such Idea’s [sic], and the Idea’s themselves that they stand for, being neither of them innate.
According to Locke, neither ideas nor words are innate. Ideas and words are learned, laid down through experience, in an indexical, rhetorical relationship in which ideas and words point at experience, which provides the foundation of the true. Words are an impression of the empirical world on the so-called tabula rasa of the human mind. As Locke put it, “the Mind” is a “white Paper, void of all Characters, without any Ideas” (p. 37). According to Duschinsky (Reference Duschinsky2012, p. 513), “in the immediate context in which Locke was writing, the term tabula rasa was a familiar image,” likely due to the Aristotelian revival of the early seventeenth century. In On the Soul, Aristotle writes, “what [the mind] thinks must be in it just as characters may be said to be on a writing tablet on which as yet nothing actually stands written” (Aristotle, n.d., III.4). That is, the mind only has in it that which was written down in it via experience with the world. Nothing is innate.
Circulating in the late eighteenth century were ideas about the relationship between word and world. Some argue that before this time period, people considered the world and the world to be seamlessly connected, even entirely integrated (Foucault, Reference Foucault2001). In the modern period, when Locke was writing, language became separated from the world such that the word merely indexed the world, which is more real and credible than its representation in language. At the end of Book IV of the Essay, Locke asserts a division between language and empirical experience explicitly, categorizing the objects of understanding as: (1) “things as they are in themselves knowable” (physica); (2) “Actions as they depend on us in order to [achieve] Happiness” (practica); and (3) methods for interpreting the signs of what is, and of what ought to be, that are presented in our ideas and words (logica) (Locke, Reference Locke1690, p. 362). This short extract divides nature, language, and human behavior into three distinct categories: “things,” “actions,” and “methods.” These elements are sharp, operating separately and rationally. “Things,” or “physica,” are knowable through sensory experience. “Actions” are what humans do to manage a practical and productive life, operationalizing both “things” and “methods.” “Methods” are directly related to language and the interpretation of signs and words. In this construction, methods and language are removed from things. They are not overlapping. Their boundaries are clear. “Things” are knowable. “Methods” are merely interpretational.
One place where we see tight linkages between eighteenth-century philosophy and eighteenth-century law is in Sir Geoffrey Gilbert’s Reference Gilbert and Locke1752 abridgement of Locke’s Essay. Gilbert is adamantly against the admission of hearsay, and he structures his arguments in ways that reference Locke both directly and indirectly. Relying on Locke, for Gilbert, hearsay is language and not empirical experience; language is distant from the empirically grounded world; language is not truth. Truth resides in empirical experience. Hearsay is repetition, too far removed from the empirical world of experience to be tolerated by the burgeoning young law with its links to Locke.
In Gilbert’s abridgement of Locke’s Essay, the two scholars collaborate in a claim that what is knowable is the “being and existence of things not language” (Gilbert & Locke, Reference Gilbert and Locke1752, p. 264). Sounding very much like Locke, Gilbert asserts that “language is nothing else but the connection of sounds to ideas” (p. 264). In eighteenth-century empiricist theory (readily translatable to law), knowledge lives in things, the material world. Language is disconnected from the material, from “things,” from knowledge. According to this theory, language functions through representation, doing little more than indexing what is true and knowable in the material world.
The rule prohibiting hearsay is related to a set of ideas that cross paths with, reproduce, and reimagine those presented by Locke in the Essay, namely the high value placed on first-hand empirical knowledge – the eyewitness. The heavy emphasis on the empirical creates space for logos to be elevated and ethos to be diminished. Logos is where truth resides. Ethos is irrelevant because of its links to subjectivity, which can never be trusted on its own. The earliest legal documentation of hearsay, the English case Thompson v. Travanion, dates to 1694, only four years after Locke’s Essay was published in Reference Locke1690. Arguments about hearsay going back to at least 1785 conceive of hearsay as nothing more than speaking, which is inferior, not to be trusted. John Pitt Taylor, quoting English Justice Francis Buller, called hearsay “a mere speaking” (Taylor, Reference Taylor1872, p. 521). Put quite succinctly in another English case, Chettle v. Chettle (1821): “What is the evidence here? Mere hearsay. Nothing seen”; evidence that is visually witnessed prevails. In this construction, statements are clearly connected to the speaker through sensory experience, but they come after sensory experience. Statements are lesser. The tight focus on sensory experience puts the thing that happened, logos, at center stage of the “true.” That is, truth can only be considered truth when it is directly related to empirically derived experience and knowledge. The speaker, their ethos, and their linguistic account are a deterioration of sensory experience. Hearsay further degrades the empirical because it is doubly removed from Locke’s notion of “things.” Hearsay rests only on the precarious and unsteady foundation of language rather than the empirical steadiness of “things.”
Locke makes a claim about language, empiricism, and truth directly when he writes: “Any Testimony, the farther of it is from the original Truth, the less force and proof it has. The Being and Existence of the thing itself, is what I call the original Truth. … [I]n traditional Truths, each remove weakens the force of the proof” (Locke, Reference Locke1690, p. 338). That is, language is a weak proof. Indeed, language mutes the material proofs because they are distant from “the original truth”:
Hearsay is no Evidence … if a Man had been in Court and said the same Thing and had not sworn it, he had not been believed in a Court of Justice; for all Credit being derived from Attestation and Evidence, … such a Speech makes it no more than a bare speaking.
Hearsay is not evidence because it is “bare speaking” – nothing more than language divided from those aspects of the material world that might make it trustworthy, namely the speaker (to cross-examine) and the material world that it references (circumstances). In Crawford v. Washington (2003), Justice Scalia connects to this eighteenth-century concern with language, referencing Gilbert directly.
Language is only ever “a bare speaking” in the law unless the gap between the empirical and the linguistic is tightened and controlled. In The Law of Evidence, Gilbert writes:
The Attestation of the Witness must be to what he knows, and not to that only which he has heard; for a mere hearsay is no Evidence: for it is his Knowledge that must direct the Court and Jury in the Judgement of the Fact. … If the first speech was without Oath, an Oath that there was such a Speech makes it no more than a bare speaking.
By repeating full phrases from his earlier works, Gilbert reasserts a ban on hearsay. The goal of the rule against hearsay is to determine what can count as evidence. As Gilbert puts it elsewhere, “nothing can be more ‘indeterminate’ than loose and wandering ‘Testimonies’ taken up on the uncertain Report of the Talk and Discourse of others” (Gilbert, Reference Gilbert1791, p. 890). Here, wandering testimonies, reports, talk, and discourse are presented as synonyms for bad evidence. Hearsay rests on the unsteady ground of language. Thus, hearsay does not meet the requirements of logos – it has no substance – which is demanded by the law. Because of this emphasis on empirical evidence, logos, in debates about whether a hearsay statement may be admissible, the law misunderstands the full rhetorical structure of a statement, rendering ethos moot at best and dangerous (read subjective) at worst.
6.3 A Contemporary Hearsay Rule
Essentially a product and representation of legal language ideology, the rule against hearsay (and all of its exceptions) is developed in the US Federal Rules of Evidence (2023), officially adopted in 1972 and most recently reviewed, revised, and re-ratified in 2023. Rule 801 (Federal Rules of Evidence, 2023) in the US Federal Rules of Evidence defines the rule this way:
The following definitions apply under this article:
(a) Statement. “Statement” means a person’s oral assertion, written assertion, or nonverbal conduct, if the person intended it as an assertion.
(b) Declarant. “Declarant” means the person who made the statement.
(c) Hearsay. “Hearsay” means a statement that:
(1) the declarant does not make while testifying at the current trial or hearing; and
(2) a party offers in evidence to prove the truth of the matter asserted in the statement.
A statement, in this configuration, is an assertion, whether written, verbal, or gestural, given as evidence toward and about the truth of the statement inasmuch as it indexes the legal matter at hand – the evidence asserted as a true account of what happened. A statement is made by a declarant. A declarant would typically be the person who witnessed the crime; in hearsay law the declarant is the person who made the original statement, a witness to the crime, the person who witnessed the event in question. The declarant is not the person on the stand who merely parrots the words of some other person. Hearsay is a statement of evidence that speaks directly to the alleged crime under legal evaluation, asserting the truth of what happened, and offering evidence about the crime. It is made in court by somebody who did not themselves witness the events about which the claims are made. Hearsay, then, always involves a speaker on the stand who narrates a prior statement about a prior event that they only heard about secondarily but did not witness themselves. Typically, such an evidentiary statement that “asserts the truth of the matter” would be given only in the context of testimony about the declarant’s first-hand experience, which would then be subjected to direct and cross-examination. In court, such a declarant bears witness to that which they have seen with their own eyes. When hearsay is involved, the person who takes the stand is not the original speaker who saw the event and who can be cross-examined, as per the Sixth Amendment (U.S. Const. amend. VI), but rather it is another person, the person to whom the declarant spoke after the fact (or was recorded and replayed after the fact), and their testimony is typically not admissible (Federal Rules of Evidence, 2023, Rule 8.2). Therefore, the admission of hearsay (potentially) runs afoul of the US Constitution, and it is that exact relationship – between hearsay and the Sixth Amendment – that is debated in criminal courts and especially in this argument, by the US Supreme Court.
There are some twenty-three exceptions to the hearsay rule (Federal Rules of Evidence, 2023, Rule 803), and therefore, there can be long and extensive analyses of which types of out-of-court, truth-bearing utterances may be admitted. I am arguing that the hearsay rule is structured and worded as though it is about the utterance, the hearsay statement presented in court, when in fact it assesses the original speaker in many different ways, more, in fact, than it assesses the utterance. This legal deconstruction of the hearsay rule, which displaces the utterance in favor of a discussion of the speaker, envisions a relationship between ethos (the credibility and trustworthiness of the speaker) and logos (empirical proofs that work against the speaker).
6.3.1 Ohio v. Roberts (1980) and Its Demise
In the mid-to-late twentieth century, a number of cases take up the issue of hearsay. Our more recent story about hearsay begins with Roberts (Ohio v. Roberts, 1979). Roberts’ rhetorical structure helps shift the analytical focus away from ethos – the credibility of the speaker – and toward logos – empirical evidence presented as truth. In so doing, Roberts surreptitiously dismisses ethos from the equation, just as Locke and Gilbert would have it. The analyses of hearsay in the Supreme Court evident in reading the precedent in the law itself, including Roberts, regularly push off accounts that can be linked to individual subjectivity as fallible and thus inadmissible. Only those linked to the empirical world that somehow circumvent the now dismissed ethos of the speaker are found to be admissible.
There are a number of heuristics circulating in case law that operationalize the rulings of the US Supreme Court. They are used in lower courts to evaluate and measure a proposed hearsay statement. According to Crawford v. Washington (2003), the well-used heuristics established in Roberts (Ohio v. Roberts, 1979), written by Justice Blackmun, run afoul of the Sixth Amendment. In Roberts, Roberts was accused of forging a check and possessing stolen credit cards belonging to Bernard Isaacs and his wife. Roberts claimed that Issacs’ daughter, Anita, had given him the check and credit cards to use. During long and extensive questioning at a preliminary hearing, Anita denied Roberts’ claim that she had given him the stolen items. Though she was subpoenaed several times, Anita did not appear at Roberts’ trial. Because Anita could not be found and against the hearsay objections of the defense, Anita’s prior testimony from the preliminary hearing was admitted as hearsay during the trial. Roberts was found guilty based on the admission of Anita’s hearsay evidence. The case was nearly immediately appealed, with the appellate court finding that the trial court had not tried hard enough to find Anita and that they erroneously admitted her prior testimony. The Supreme Court agreed to hear Roberts (Ohio v. Roberts, 1979) and take up, again, the issue of hearsay.
The test developed by the Supreme Court in Roberts relies heavily on two sociolegal constructions. First, hearsay can be admitted if it fits within a “firmly rooted hearsay exception” (Ohio v. Roberts, 1979, p. 66) This component of the hearsay test points directly back to the framing of the US Constitution – any hearsay exception in place when the Sixth Amendment was conceived is typically deemed valid. If the exception was not a part of longstanding tradition, according to Roberts, admissible hearsay must be shown to bear “particularized guarantees of trustworthiness” (Ohio v. Roberts, 1979, p. 66). Roberts’ formulation rests heavily on a largely under-defined conception of “trustworthy” and the related concept “reliable.”
Relying heavily on the 1895 opinion Mattox v. United States (1895, hereafter Mattox), Roberts (Ohio v. Roberts, 1979) affirms the requirements of confrontation. According to Mattox,
a personal examination and cross-examination of the witness in which the accused has an opportunity, not only of testing the recollection and sifting the conscience of the witness, but of compelling him to stand face to face with the jury in order that they may look at him, and judge by his demeanor upon the stand and the manner in which he gives his testimony whether he is worthy of belief.
In this late nineteenth-century precedent, Mattox focuses attention on the declarant. For our purposes, there are two important concepts in Mattox that interlock: “conscience of the witness” and “worthy of belief.” In 1895, when watching a person give testimony, the jury was looking for imperfections and inconsistencies in the testimony, but they were also required to assess the conscience of the witness: their trustworthiness and believability. Their ethos. By analyzing the conscience of the witness, the jury can determine whether that person was “worthy of belief,” that is, trustworthy. Hearsay in Mattox is inadmissible because it disallows the jury from evaluating the witness’ ethos.
The requirement for hearsay to match the requirements of in-court, on-the-stand testimony as explained in Mattox v. United States (1895) positions ethos as an important feature of law at that time. Mattox asserts that only hearsay that meets the standards of an in-court performance of ethos tested in cross-examination is admissible. In at least this one iteration of hearsay (Mattox), the relationship between the speaker and the audience somewhat follows typical rhetorical lines of logic: The speaker makes assertions of truth to which legal rules and procedures are applied and, in the process, the speaker’s worthiness, conscience, and trustworthiness (ethos) are assessed by the audience, in this case a judge and jury (pathos), in order for the evidence (logos) to be presented and accepted as true.
Because Mattox v. United States (1895) places so much emphasis on the witness on the stand, when hearsay is admitted, the speaker’s ethos is positioned in such a way as to be open to peril and ultimately to erasure. In demanding an original witness, Mattox asserts extra weight on the importance of empirically grounded, experiential truth in relationship to arguments about the speaker. In discussions about hearsay and their heightening attention to the ethos of the speaker on the stand as they are related to the truths found in the material world, Mattox opens room for the ethos of the speaker who is not on the stand to be suppressed when the speaker is unavailable. This is because the person with empirical knowledge is not testifying. This assertion ultimately lays the groundwork for ethos to be suppressed altogether. Let me explain. When the spotlight is put on ethos and logos as they are in Mattox, logos is isolated from the speaker, seeming to work independently from ethos. Logos refers to the world of lived experience, while ethos refers back to the speaker of the utterance in question, alone. Here as well as elsewhere in the law, then, ethos is disparaged precisely because it is unique to the individual and must be assessed as such. The relationship of logos to objectivity is wildly preferred.
In its application of Mattox v. United States (1895) and other precedents, Roberts (Ohio v. Roberts, 1979) works to isolate logos in order to render it open to evaluation on its own. For example, Roberts applies Mancusi v. Stubbs (1972, hereafter, Mancusi) (which itself quotes from precedents, which also quote from precedents, and so on ad infinitum). According to Roberts: “The focus of the Court’s concern has been to ensure that there ‘are indicia of reliability which have been widely viewed as determinative of whether a statement may be placed before the jury though there is no confrontation of the declarant…’” (Ohio v. Roberts, 1979, pp. 65–66, quoting Mancusi v. Stubbs, 1972).
Using Mancusi v. Stubbs (1972), Roberts (Ohio v. Roberts, 1979) ultimately shifts from an evaluation of the ethos of the speaker/witness (Mattox) to an evaluation of the statement, isolating the statement and assessing it only in terms of logos – empirical proof. In Mancusi, the statement is “placed,” passive voice, before the jury with the speaker of the statement grammatically removed. This grammar makes it seem as though there was no speaker behind potentially admissible hearsay at all, which, in fact, resonates throughout this case law. Only delinked from a speaker and their ethos can a statement bear the so-called indicia of reliability and be admitted as hearsay. This procedure is possible because the speaker (ethos) has been first isolated and then excluded in legal discussion and debate.
The shift away from the speaker and toward the statement makes one further shift in Roberts (Ohio v. Roberts, 1979) away from an evaluation of ethos. The notion “indicia of reliability” that Roberts relies on uses precedential chains that reach backward to eighteenth-century ideas about reliability. Dutton v. Evans (hereafter Evans) also cited in Roberts, argues: “circumstances under which [the out-of-court speaker] made the statement were such as to give reason to suppose that [the out-of-court speaker] did not misrepresent [the accused’s] involvement in the crime” (Dutton v. Evans, 1970, p. 10). This reliance on circumstances to indicate the trustworthiness or reliability of the statement is one place where we see Locke’s language ideology peek through. Indeed, when the speaker is divided from the statement, circumstances – material conditions – become even more important and ethos is rendered immaterial, in every way that word can mean. The reliance on circumstances to provide assurance of truth when the speaker is unavailable to take the stand and be assessed is operational precisely because circumstances are empirical rather than linguistic, therefore placing such a statement closer to the truth. According to the logic of the law, “circumstances” (we can call them events and contexts) can render some statements inherently true. “Circumstances” are truths from the empirical, material world, and, as such, the law proposes, they control, even counter, the natural subjectivity of the speaker, creating space to push the speaker out of the equation completely.
6.3.2 Enter Crawford (2004) and the Redefinition of Hearsay
I turn attention now back to Crawford v. Washington (2003) and to Davis/Hammon (Davis v. Washington, 2006), both of which were written by Justice Scalia. For Justice Scalia, when analyzing a case and writing a Supreme Court opinion, “historical inquiry” is entirely relevant, if not required, in order to accurately understand the original meaning of any document, but especially a legal document (see Hannah & Mootz, Chapter 2 in this volume). In Crawford and Davis/Hammon, Justice Scalia uses historical research and performance when he insistently relies on (his interpretation of) the lexical and legal knowledge of the Framers. When Justice Scalia writes the following in Crawford, he is very clearly presenting a seventeenth-century-language ideology: “Where testimonial statements are involved, we do not think the Framers meant to leave the Sixth Amendment’s protection to the vagaries of the rules of evidence, much less to amorphous notions of ‘reliability’” (Crawford v. Washington, 2003, p. 61). Justice Scalia very obviously and purposefully inserts and applies his linguistic and legal language ideology about original intent in his opinions about hearsay. As he said in an interview, “words have meaning. And their meaning doesn’t change” (Senior, Reference Senior2013).
Crawford v. Washington (2003) overturned Ohio v. Roberts (1979) by operationalizing new terms, definitions, and rhetorics that have tighter restrictions and more stringent methods for evaluating hearsay, but that, nevertheless, share its language ideology, relying heavily on empiricism and further eradicating the desirableness of speaker ethos. As Justice Scalia presents them, the terms and concepts in hearsay discourse are defined in ways that explicitly and directly link present case law to historical texts and definitions through the copious use of eighteenth-century precedents and legal treatises.
Recall that Crawford v. Washington (2003) was a case in which a man, Lee, had been stabbed by Crawford, and in which Sylvia Crawford’s police interrogation was admitted as hearsay when Crawford invoked his marital privilege (explained further above). In Crawford, Scalia begins his arguments with a reference to a dictionary definition of the term “witness” published in 1828. Crawford states,
The text of the Confrontation Clause reflects this focus. It applies to “witnesses” against the accused – in other words, those who “bear testimony.” 2 N. Webster, An American Dictionary of the English Language (1828). “Testimony,” in turn, is typically “[a] solemn declaration or affirmation made for the purpose of establishing or proving some fact.”
First, note that the dictionary referenced in Crawford is as close to the time of the Framers of the Constitution as Justice Scalia could find – the 1828 edition. In that 1828 dictionary, a “witness” is a person “who bear[s] testimony.” This is not controversial. In the context of the law, it gestures quite obviously to in-court testimony. A witness, according to this dictionary from the early nineteenth century, “bears testimony.” According to the same edition of Webster’s dictionary, “testimony” is a “declaration or affirmation made for the purposes of establishing or proving some fact.” The passive construction of this definition excises the speaker of the declaration. Locke would be proud.
The concept of witness is the central concern of the Confrontation Clause of the Sixth Amendment (U.S. Const. amend. VI) (even though most of the discussion in hearsay case law in the twenty-first century has been about the statement), and so the Supreme Court must begin with the “witness.” However, the real work in this reframing of hearsay is with “testimony.” This case law presumes that testimony comes prior to the act of being a witness in the process of giving what the court deems inherently testimonial; in this process, the declarant is automatically transformed into a witness, who must always have been acting as a witness, regardless of whether they were on the stand, under oath, or are even aware of their witnessing. The involvement of a witness animates the requirements of the Confrontation Clause. (In the Sixth Amendment, the accused is promised the right to confront their accusers, so the witness must be present in the equation of hearsay.) The process of evaluating hearsay within this definition’s structure – the one that puts testimony in front of witnessing – removes the statement from the real, in-the-world speaker (Sylvia and others) and relinks it instead to a legal abstraction, witness, and ultimately to the circumstances in which and about which the statement is made. It is these circumstances that indicate whether a statement is testimonial, and it is testimony which proves the presence of a witness. With these legal acrobatics, the statement is transformed into evidence (or hearsay) that has nothing to do with the speaker or their ethos; it has to do only with the legal formation, witness. These statements are constructed out of logos – the empirical. Testimony relies only on circumstances for its relationship to real knowledge and truth. The formation of witness follows the identification of testimony. It is the empirical world, the circumstances, that can verify the truthfulness of testimony. It is no longer the “conscience” of the witness (Mattox v. United States, 1895). Moreover, it is testimony that produces a witness, which functions as sort of an abstract for anybody, not only the person whose utterance is in question.
To restate my argument, legal discourses transform a statement made in the world into legally recognizable testimony, by overemphasizing logos and censoring ethos. This takes the statement functionally away from the speaker by transforming the speaker into the legal abstraction, witness, which is equated with a speaker on the stand but should not be confused with an embodied person. In this way, Crawford v. Washington (2003) delinks the speaker from the statement, because testimony is always spoken by a witness, but testimony is proved not through reference to the witness but rather through reference to circumstances. However, in evoking the abstraction, witness, the Sixth Amendment to the Constitution is called into play. If the speaker is a witness (in or out of the courtroom and whether or not they know they are a witness), then the requirements of the Confrontation Clause must be satisfied. Any utterance deemed testimonial must be cross-examined.
It is the circumstance, the logos, that proves the testimonialness of a statement, and, in the process, an abstract witness is produced who has no ethos. According to the Supreme Court, Sylvia Crawford’s recorded police interrogation was testimony and thus Sylvia was acting as a witness even though she was not in court and had not sworn an oath, and even though she likely had no idea that she was acting as a witness in the legal sense. These two paired concepts have the power to reach through time, pulling people into their abstracting capabilities – turning real people into abstractions – and altering their relationship with their own speech.
After giving the dictionary definitions discussed above, Crawford goes on to say:
An accuser who makes a formal statement to government officers bears testimony in a sense that a person who makes a casual remark to an acquaintance does not. The constitutional text, like the history underlying the common-law right of confrontation, thus reflects an especially acute concern with a specific type of out-of-court statement.
In other words, some declarations are testimonial in the legal sense, even if they are produced out of court, while others are objectively not testimonial. And importantly, statements are testimonial not because the speaker was an in-court witness, but because of the circumstances surrounding the production of the statement, even those outside of the courtroom, indicated that the speech was testimonial. The trustworthiness of testimony is a feature of logos alone.
6.3.3 Enter Davis/Hammon (2006)
Davis/Hammon (Davis v. Washington, 2006), also written by Justice Scalia, supplies a single opinion for two cases as a way of rectifying perceived misapplications of law in lower courts. In one case (Davis), the Supreme Court admits hearsay as constitutionally sound, and in the other (Hammon), the Supreme Court determines that the hearsay is not admissible. Davis and Hammon are both domestic violence cases. In Davis, Michelle McCottry called the police when she was being assaulted by her former boyfriend, Davis. On the phone with 911, McCottry exclaimed, “he’s here, jumping on me again” (Davis v. Washington, 2006, p. 877). Police arrived four minutes after she called 911 and observed that McCottry was “frantic.” The excited utterance exception to hearsay, which was used in Davis, hinges on the emotional state of the speaker; the speaker must be shown to be “frantic,” “hysterical” (cf. Andrus, Reference Andrus2015). McCottry did not appear at Davis’ trial, and so her statements to 911 were admitted using the excited utterance exception to hearsay. These statements are hearsay because the person who said them – McCottry – was not available in court to be cross-examined. McCottry’s “excited utterance” is nevertheless determined admissible by the Supreme Court because it is inherently “nontestimonial,” that is, not spoken by a witness.
Hammon (Davis v. Washington, 2006) involves an assault by Hershel Hammon on his wife, Amy Hammon, resulting in a call to police, who responded to a “domestic disturbance” (p. 819). When they arrived, police found Amy sitting outside on the front porch. She appeared “somewhat frightened” but told them that “nothing was the matter” (p. 819). She gave police permission to enter the home, where they found a broken heater and other broken household objects. As is usual, officers separated Amy and Hershel and questioned them individually. After hearing Amy’s account of what happened, the police officer questioning her had her fill out a battery affidavit. Amy did not appear when she was subpoenaed to Hershel’s bench trial. Her statements to police when they arrived at her house were not admitted at trial as excited utterances because Amy appeared too calm when officers spoke with her. The battery affidavit, however, was admitted at trial. The Supreme Court disagreed with this decision. Amy’s statements were determined “testimonial” and were thus excluded because Amy had been acting as a witness when she spoke with police.
The arguments in Davis/Hammon (Davis v. Washington, 2006) compare and contrast the two cases, using the newly minted Crawford v. Washington (2003). At the center of the arguments in Davis/Hammon is the renewed p/re/conception of what it means to give testimony and act as a witness. Ultimately, Davis/Hammon reinforces the arguments in Crawford that some hearsay statements can be admitted without cross-examination, using the reasoning that they are obviously legally and semantically stable, reliable, and trustworthy because of their close relationship to the empirical world and the absence of speaker subjectivity/ethos. Some statements do not give testimony, they are not spoken by a witness, and therefore they don’t need to be cross-examined. Others are testimonial and as spoken by a witness.
In the recapitulation of the legal reasoning established in Crawford v. Washington (2003), Davis/Hammon makes the following argument:
A critical portion of this holding, and the portion central to resolution of the two cases now before us, is the phrase “testimonial statements.” Only statements of this sort cause the declarant to be a “witness” within the meaning of the Confrontation Clause. … It is the testimonial character of the statement that separates it from other hearsay that, while subject to traditional limitations upon hearsay evidence, is not subject to the Confrontation Clause.
The Supreme Court therefore determines that McCottry’s statements to 911 were nontestimonial (Davis), while Amy Hammon’s affidavit were testimonial. Davis/Hammon explicitly states that a testimonial statement “cause[s]” a person to become a witness. This is some kind of legal magic. According to Davis/Hammon, legal procedure and evaluation transform a person into a witness and in that process, the speaker is divested of their subjective, rhetorical link to their own language, cutting off ethos in perpetuity. Indeed, it is presumed that such a link never existed. The admissible statement is the product of the circumstances in which the statement is uttered. The speaker in this rhetorical construction is positioned as a sort of legal, vestigial tail.
Davis/Hammon goes on to further clarify the distinction between testimonial and nontestimonial:
Statements are nontestimonial when made in the course of police interrogation under circumstances objectively indicating that the primary purpose of the interrogation is to enable police assistance to meet an ongoing emergency [Davis]. They are testimonial when the circumstances objectively indicate that there is no such ongoing emergency, and that the primary purpose of the interrogation is to establish or prove past events potentially relevant to later criminal prosecution [Hammon].
Notice in this quotation that the use of the word “objectively” (again) relates us directly to Locke and his theory of truth, rooted in high modern notions of the relationship between language and the world of experience. In Crawford, “objective speakers” such as Amy Hammon, who was sitting on her front porch having just been assaulted, should know, according to legal logic, that her words were likely to be used in court. In that moment, as a reasonable person, she must have known that in talking to police, she was bearing witness. Or so the legal reasoning goes. Davis/Hammon further develops the distinction between testimonial and nontestimonial by introducing the concept “primary purpose” (Davis v. Washington, 2006, p. 822). In the “primary purpose” assessment, it is the “circumstances” that behave objectively. These objective circumstances – not the speaker – indicate what kind of statement it is: testimonial or nontestimonial. To be clear, hearsay is not a statement existing in some accessible past event, waiting to be located in the world and simply plucked out of the circumstances. What I am showing in this analysis is that these statements – testimonial or nontestimonial – are framed, constrained, and produced in the law; their supposed objectivity is a legal construct that the law itself refuses to see.
The rhetorical force of the statement in this legal context is established by analyzing the circumstances and the “primary purpose” (Davis v. Washington, 2006, p. 822) for which the statement was elicited. Circumstances such as an ongoing emergency produce a statement that can be considered nontestimonial, while a discussion with police would be considered (always and already) testimonial. Amy Hammon’s statements are excluded, labeled hearsay, because their primary purpose was to answer police questions, which an “objective speaker” would have known could be used later in court. Michelle McCottry’s statements, on the other hand, were a cry for help, and were therefore admissible as objective accounts of what really happened. McCottry’s speech was nontestimonial because she was asking for help. The “primary purpose” in both cases is determined by the context, the circumstances (i.e., the logos), surrounding the production of the statement, not by assessing the speaker, who is of course structurally absent.
The Davis/Hammon court therefore asserts that Amy’s statements given “under official interrogation are an obvious substitute for live testimony, because they do precisely what a witness does on direct examination; they are inherently testimonial” (Davis v. Washington, 2006, p. 830). Because police ask questions in similar ways as lawyers in court do, the resultant statements are assumed to function identically to in-court testimony. Not to get into the virtually innumerable ways that police and lawyer questioning are different. Circumstances, not the speaker, have discursive agency in both testimonial and nontestimonial configurations; they are two sides of a coin, proving each other correct. The speaker of a testimonial statement is translated into an always witness, a linguistic, rhetorical construction that places empirical circumstances at the helm of identifying the truth. And with that, the delinking of the statement from ethos is complete.
6.4 Conclusions
The legal language ideology apparent in the hearsay rule, along with its exceptions, uses a Lockean language ideology to re/organize and re/structure the relationship among ethos/pathos/logos, subjectivity/void/objectivity, and speaker/void/empirical evidence. In Crawford v. Washington (2003) and Davis/Hammon (Davis v. Washington, 2006), testimony is moved out of the courtroom to any place that a person could be speaking in circumstances that indicate that the statement is empirically true and that the person would necessarily know that their speech may be used later in court as testimony. When testimony is placed out in the world like this, the speaker is forced to bear witness without a connection to their own ethos, regardless of their discursive, rhetorical desires, and without the trappings of the courtroom context. The heavy reliance on logos transforms ethos into a forced, empty position. Witnessing becomes a product of legal reasoning, rather than an in-court, visible, literal procedure, and, as such, it can be applied to any speaker, speaking in circumstances that empirically proclaim that the statements produced therein are always already testimony. The orphaned statements that result from and are produced in hearsay discourse are hypostatized. They are diminished rhetorically, reduced to a subjective/objective dichotomy.
When hypostatized, statements lose their rhetorical flux and semantic flexibility. Speakers, too, suffer rhetorically and semantically through legal sedimentation. They are either witnesses or not, objectively. In either case, the speaker’s relationship to their own speech is weakened if not dissolved. To be clear, I don’t necessarily think a statement belongs to a speaker in a totalizing way. From a postmodern rhetorical approach, statements are produced by a broad set of actors, events, social discourses, and the like, all engaged in interaction. The ways in which the rule against hearsay dismisses human agency and instead positions it on the text may seem aligned with postmodernity. However, this dismissal of the speaker and their agency doesn’t disbelieve in sovereign agency; it merely rearranges the position of agency in the rhetorical situation. What interests me in this discourse is the fact that the law does not take into account the effects of social discourses, interaction, the productive effects of cultural values, the broader relationship between events, or the like, on things such as testimony (logos/language) and witnessing (ethos/embodied person).
I have argued that legal reasoning, similar to Lockean theories of language, identifies reality in events which are considered to be finite, real, and identifiable. Witnesses are instrumentalized such that they are legally useful. And this is the problem. Not only is semantic and rhetorical access to the statement undone in the evaluation of hearsay, but it is also undone in such a way as to instrumentalize the speaker and erase their ethos. Such a structuralist language ideology idealizes and obsesses over the empirical. Buried within the legal contest between “testimonial” and “nontestimonial” is the idea that signification and meaning are only the product of sensory experience that is translated into words and that are only true when they mean the same thing for everyone forever because of their relationship to the “real” world.
I have shown that the dismissal of ethos in favor of all logos all the time is more closely related to high modernity, structuralism, and a Lockean language ideology. Language is always problematic in a Lockean rhetoric, because, in his representational model of language, language is merely indexical and lesser than the empirical world of experience. Language is itself not real. Language is merely a channel for the empirical. Perhaps unexpectedly, a byproduct of this view of language is the dismissal of not only the important work that language does for and in the law (in a set of laws about language, no less), but also the dismissal of the pivotal role that speakers play in the rhetorical production of their speech. According to the legal reasoning circulating in hearsay legal discourse, if language is built on the unknowns that speakers bring to the rhetorical equation (motive, intention, emotion, etc.) rather than on a foundation of sensory experience, if language is not moored to something empirical, then it will float away into semantic chaos where it will not be legally useful.
Open access