Using court documents as data: opportunities and challenges for sociolegal scholarship

Abstract

Though the use of court documents as data is widespread within US sociolegal scholarship, their use remains surprisingly undertheorized as a methodological practice. This article, therefore, asks, what differentiates court materials from other forms of documentary data, and how do these attributes impact claimsmaking in law and society scholarship? Drawing on varied empirical examples from existing scholarship, we uncover five distinctive attributes: their multitemporality, their dialogic nature, the multiple truths they house, their multivocality, and their social productivity. Considering these attributes, we argue that court documents unite our diverse field of scholarship in two important ways. First, as an essential output of the legal system, they are arguably “our” data, shaping law and society as we know it today. Second, they both reify and obscure the power dynamics that make social inequality so durable, helping inequality appear “just.” Despite their underexploited promise for theory-building in sociolegal research, we also discuss the practical, epistemic, and ethical pitfalls to their use. Ultimately, ignoring these rich yet complex documents is to our field’s analytic peril.

Introduction

Court documents have a rich history in sociolegal research, figuring in many canonical studies using quantitative (e.g. Edelman et al. 1999, 2011), qualitative (e.g. Ewick and Silbey 1998; Rosenberg 2008; Haltom and McCann 2004), and humanistic (e.g. Haney López 2006; Penningroth 2003, 2008) approaches. Such documents include any official filing or record with a local, state, national, or international court, as well as semi-public arbitration forums, and can cover any substantive area of law. Given the social breadth of law, it is perhaps unsurprising how many sociolegal subfields engage with these materials as both primary and secondary data, including science and technology studies (STS) (Cole 1998, 2001; Jasanoff 1995, 2006; Lynch 1998; Mnookin 2001, 2014; Moore and Singh 2018; Singh 2017; Vogler 2019, 2021), policing and punishment (Hlavka and Mulla 2018, 2021; Lynch 2016, 2019; Myrick 2013; Norris and Grol-Prokopczyk 2015, 2019; Stitt et al. 2024), urban studies (Bartram 2022; Jang-Trettien 2021; Stuart 2011), racial and ethnic studies (Haney López 2006; Hartman 2008, 2022; Penningroth 2003, 2008; Welch 2018), and labor and discrimination (Edelman et al. 1999, 2011; Berrey et al. 2017; Nelson and Bridges 1999; Haltom and McCann 2004; McCann 1994).

Yet, despite their widespread use, court records remain surprisingly undertheorized in their methodological practice. The aim of this article is to initiate a conversation about the embedded assumptions and implications of using court documents as data. We ask, therefore, what differentiates court materials from other forms of documentary data, and how do these attributes impact claimsmaking in law and society scholarship? We focus here on the US tradition because it is the area of our shared expertise, although we hope to spur similar conversations for legal scholarship from across the world. We argue that US court documents boast five attributes that, taken together, distinguish them from other documentary data in empirical research and also reveal the essential relations of power at the heart of all law and society scholarship. First, court documents are multitemporal, speaking to the past, present, and future at once. Second, they are inherently dialogic. This is in part due to the nature of adversarial law and judicial review in the US but also reflects the importance of court documents as cultural artifacts. Third, court documents house multiple, often conflicting forms of truths, ultimately subordinated to “legal truth” as reflected in case outcomes. Fourth, court documents are multivocal, simultaneously speaking from and to a variety of constituents with different goals. And fifth, court documents are socially productive, engendering categories and forms of agency that organize social life.

Given these attributes, we contend that legal documents unite our diverse, interdisciplinary field in two important ways: first, and perhaps most obviously, court documents are our shared, institutional referent point, whose attributes, we show, have profoundly shaped claimsmaking in sociolegal research. As the output of the courts, legal documents are, in a loose sense, “our” data as law and society scholars. This does not mean we have a monopoly on them, nor that all sociolegal scholars must study court records. However, as the primary output of the courts, they are foundational to much of our scholarship, featuring prominently in both canonical and contemporary works. Our aim is to spark more explicit theoretical and methodological discussion around how best to exploit these data and manage their drawbacks in sociolegal research. Second, court documents reveal a social truth at the heart of all law and society scholarship: the courts are not, nor ever have been, the domain of “justice” in the philosophical sense. Beyond reflecting social inequities, legal documents define and reify the dynamics of power that make social inequality so durable. Here, we are indebted to critical theorists who have made similar observations over many decades. We draw on their insights below and urge empirical scholars to contend with them in future research. Crucially, because court documents are an official output of the state, they make structural inequities appear both normal and “just.” Yet rather than dismiss their scholarly value due to this unpleasant truth, we argue that court documents offer underexploited analytic promise alongside their potential pitfalls.

Our piece proceeds as follows. First, we define “court documents,” situating them within the broader categories of document and artifact, reviewing extant works that engage with them, and showing the kinds of questions court documents have been used to answer. Next, we lay out court documents’ five core attributes, as derived from existing research. How have these contributed to seminal findings in our diverse, interdisciplinary field, and how might researchers more deliberately use them? We describe how each of these attributes speaks to an overarching social truth: that court documents index the involved parties’ and other stakeholders’ power. Often, documents reflect the coercive might of the state or powerful private actors over lesser opponents. Sometimes, however, they show how courts may be used as “weapon of the weak,” to upend expected outcomes. Certainly, questions of power need not feature in every sociolegal study – yet such dynamics operate, whether we choose to center them. Through varied examples we illustrate how court documents have contributed to empirical findings, enduring themes, and theoretical insights foundational to our field. In the penultimate section, we discuss the challenges of using court documents as data in sociolegal research, highlighting both their power to shed light on social inequalities and their tendency to reify them. If we ignore court records’ theoretical implications – their promise and pitfalls – we do so to our own analytic detriment. We conclude our piece with some ideas for future research.

Court documents as data in extant research

We define “court documents” as any official filing with a court at any level. Such filings can be criminal, civil, or administrative, filed by a lawyer or layperson, and can operate at all scales of governance, including at the blurry public–private divide. Such documents are generated through an adversarial process in the US, the setting for our analysis, but every country produces its equivalent. In the US, court documents include everything from indictments and complaints to trial transcripts, motions, sentencing memos, the documentation of evidence, professional evaluations, amicus briefs, and, of course, formal judgments issued by judicial panels. Typically, as official documents, court filings are publicly accessible, if often procured at some hassle and expense. What unites these diverse documents is their filing, and subsequent record, within a state-sanctioned dispute forum. Thus, while related, an attorney’s personal notes, a District Attorney’s internal memo, or a legislative policy paper would not count as court documents under our definition.

As a document type, court filings are “paradigmatic artifacts of modern knowledge practices” (Riles 2006: 2; see also Burazin et al. 2018). In recent treatments, “documents” embrace a vast range of bureaucratic objects (e.g. official meeting minutes, school and medical records, tax filings, and much besides) (Riles 2006; Heimer 2006; Stark 2019; Vismann 2008). In her influential volume, Riles subsumes their study to broader questions of ethnographic method. By contrast, historians and comparative-historical methodologists have long treated them as primary data in themselves. Whether as focal or supplementary data, court records have figured in existing studies in three broad ways, which we detail here.

The courts as political actors

One key body of scholarship uses court documents to spotlight judges’ role as political actors and policymakers. Such works mostly mine federal decisions to reveal how courts have defined certain policy spheres through their interpretation of core constitutional rights. Feeley and Rubin (1998), for instance, produce five case studies of Eighth Amendment jurisprudence to illustrate how judges engage in policymaking by setting specific parameters for action in their decisions. Similarly, Reiter (2012, 2016) studies case law to trace how 20th-century courts helped create supermax facilities and legitimize solitary confinement, including how courts not only addressed their constitutionality but also such facilities’ physical design and administrative structure. Simon (2016) takes a similar tack, using Brown v Plata and its precursors to show how contemporary mass incarceration and crime control policy evolved, treating California as a microcosm of national penal trends.

At the opposite pole, eminent studies combine court documents with other data to reveal the courts’ sociopolitical impotence. Rosenberg (2008) uses seminal case law to show how seldom the Supreme Court has actually been an agent for progressive change (see also Scheingold 2010). Haltom and McCann (2004) draw on case law to puncture the popular myth that courts foment American litigiousness, tracing it to sensationalist news coverage of unrepresentative cases. They argue such misconceptions matter since they distract from pressing questions of institutional capacity and the shortcomings of the welfare state (Haltom and McCann 2004: 30; cf. Felstiner et al. 1981).

Law as site of struggle

A second tendency of studies using court records spotlights law’s role as a space for contention. Notably, scholars of organizations, social movements, and discrimination have shown how power steers law’s interpretation and implementation. Edelman et al. (1999, 2011) use decades of judicial opinions to show how organizations shaped the “law in action” of civil rights in their groundbreaking work on legal endogeneity. Wilson (2013) studies court documents to supplement interviews with abortion-rights and anti-choice organizers, illuminating how social movements, litigation, and politics intersect. Epp (2009) shows how activists and lawyers jointly forced bureaucracies to implement formal legal gains in the spheres of sexual harassment, playground safety, and police brutality. Berrey et al. (2017) also paired interviews with records from randomly sampled workplace suits to reveal the onerous burdens plaintiffs face, belying the promise of employment discrimination law. Similar studies have spotlighted workplace gender parity (Nelson and Bridges 1999), including how activists wield legal discourse in fights for pay equity (McCann 1994). Triangulating court records with other sources, Paris (2010: 2) examines reformers’ fight for equitable school funding, highlighting the “interplay of law and politics in litigation-based” advocacy.

Law’s role as a field for struggle permits us to study actors’ relative social agency, while inviting methodological reflection. Theorists of disputing and legal consciousness, for instance, have mapped how varied factors impede or enhance subjects’ prospects for claimsmaking (Engel 1984; Ewick and Silbey 1998; Felstiner et al. 1981). Historians, legal scholars, and critical theorists have shown how archives typically omit marginalized groups yet occasionally bear witness to resistance. Hartman (2008), for one, warns against treating court files as records of “fact,” given that cases involving violence toward slaves privileged the accounts of white slave-owners and traders. Dayan (2011: xi) similarly points up law’s power to both make and erase personhood, creating “victims of prejudice or inheritors or privilege” (see also Felstiner et al. 1981; Myrick 2013). Scholars as different as Walter Benjamin and Robert Cover have noted the imbrication of legal documents and their interpretation with past and potential state violence (Benjamin 1968; Cover 1986). In contrast, Penningroth (2003, 2008) and Welch (2018) reveal how both free and enslaved Black people in the Antebellum South used local courts to pursue claims. Refuting the notion that 19th-century African Americans were wholly excluded from civil society, their studies chronicle skillful claimsmaking over personal property and legal standing.

Sorting and knowledge

Perhaps most commonly, scholars use court documents to study how law sorts social phenomena and ratifies knowledge. This research stream is broad and best subdivided.

Some researchers use court records to study the making and maintenance of criminal categories, like the “sexually violent predator” (Vogler 2019, 2021) or even new types of wrongdoing that don’t fit existing offenses (Singh 2017). Occasionally, scholars partake in the labeling, as with quantitative efforts at locating “violent extremists” (Whittaker 2021; cf. Norris and Grol-Prokopczyk 2015, 2019). Naturally, the law’s sorting and epistemic function often disadvantages those it categorizes. As Myrick (2013) notes, criminal records may depersonalize the accused, creating a one-sided “textual proxy” without their input.

Such efforts reach beyond criminal matters. Haney López (2006: 3), for instance, studies early 20th-century disputes to trace the legal construction of race, cases that “fram[ed] fundamental questions about who could join the citizenry in terms of who was White.” He shows that courts’ treatment of race differed markedly from scientific understanding, even at the time, nevertheless defining not only who was white but why (2006: 2). Scouring records, other scholars have noted law’s power to sort our relationships to property by distinguishing legitimate from illegitimate home ownership (Jang-Trettien 2021) or by ascribing liability for repairs (Bartram 2022).

Other research programs reveal law’s power to ratify new forms of knowledge-making but also law’s own fuzzy standards of proof. STS has tracked the courts’ role in contests over science and expertise (Jasanoff 1995, 2006). It has chronicled the reputational rise and fall of fingerprinting (Cole 2001), debates over forensic science (Cole 1998; Hlavka and Mulla 2021), fights over DNA testing (Lynch 1998), the use of digital communications as evidence (Hlavka and Mulla 2018), and reliance on imagery to ascertain guilt (Mnookin 2001, 2014; Moore and Singh 2018; Stuart 2011). Punishment scholars have also mined court documents to question the use of actuarial tools to predict future risk and offending (Lynch 2016; Vogler 2019, 2021), highlighting where such tools clash with other forms of expertise (Degenshein 2025).

Scholars of law, literature, and linguistics scrutinize legal discourse to illumine the boundary of proof and persuasion. Bennett and Feldman (2014), for instance, use trial transcripts to reveal narrative’s role in legal judgment, including to buttress inadequate evidence. Solan and Tiersma (2005) use case studies from varied disputes to examine how specific word choice can influence outcomes. Conley et al. (2019) similarly probe the links between language and legal power, drawing on case law, transcripts, and other court documents. Multiple chapters in Heffer et al. (2013) use court documents to trace how texts “travel” through, ultimately shaping, the legal process. Similarly, contributions to Ehrlich et al. (2016) use court documents to examine the forging of legal consent.

The use of court records as sociolegal data is neither new nor obscure. Many works cited here are canonical and lauded for their empirics. But despite their widespread use, little has been written of these data themselves, overlooked in favor of other methods to which they are so often yoked. Yet, as we argue below, it is often court records’ specific attributes that enable scholars to produce empirical and theoretical insights. Specifying those attributes and linking them to past and ongoing theory-building will promote a better grasp of the inferences we can draw from these data as well as their limits.

What sets them apart: the promise of court documents

Court documents offer rich data to sociolegal scholars across disciplines. But their value flows from characteristics that distinguish them, in kind or degree, from other documents and artifacts. We identify five traits that, taken together, set court records apart from other documents: their multitemporality, their dialogic nature, the multiple truths they house, their multivocality, and their social productivity. Certainly, not all these characteristics are exclusive to court documents. And yet, we contend, court records are singular among document types for how they combine these traits and for the density and weightiness of their presentation, providing scholars invaluable means for empirical and theoretical claimsmaking.

In what follows, we use examples from a broad spectrum of sociolegal works, including our own, to illustrate these traits and their relevance. Individually and in combination, these characteristics offer scholars great analytic purchase, dramatizing the role of legal institutions as means for both exercising and resisting power.

Multitemporality

Saidiya Hartman (2008) argues that scrutinizing the archive can tell us about our pasts, who we are presently, and who we hope to become. Indeed, all archival documents and artifacts preserve pasts for an imagined future audience (Burazin et al. 2018; Riles 2006). But court records distinguish themselves by putting past, present, and future explicitly in play, often simultaneously and in the same document. Further, the nature of our adversarial system and judicial review means that these multitemporal records are constantly being revisited and reconstituted, both within and outside the courts, where pasts become presents, then futures, ad infinitum. Court documents’ multitemporal qualities inform some of the most consequential theory-building in sociolegal scholarship to date.

In the US legal system, trial participants use court filings to define past events for present adjudication, with the aim of shaping the parties’ future, and, sometimes, broader futures via precedent. What’s more, this interplay of past, present, and future is conflictual: opposing parties present their versions of the past, but only one can overwrite the near future as the “winner.” This adversarial frame sets court records apart from other documentary artifacts that also entangle pasts, presents, and futures (Polletta et al. 2011). Court filings are thus not only a record but a record of how status and authority are asserted and challenged, records that allow us to glimpse how things could have been had the other side won.

In real time, court documents chronicle a “present” dispute as it unfolds over weeks, months, or years. Each side proffers its version of past events, with the rules of procedure dictating what their narratives may include (Burns 2001). Like meeting minutes, legal filings describe an event as it unfolds. But the stakes are greater. Judge’s rulings, lawyers’ filings, and trial transcripts capture moments in a narrative tug-of-war whose government-sanctioned consequences may reverberate beyond the case. This feature is why “disputing” has been a thematic pillar of sociolegal scholarship for over 50 years (e.g. Abel 1973; Galanter 1974; Mather and Yngvesson 1980; Felstiner et al. 1981; Edelman et al. 1999, 2011; Nelson and Bridges 1999; Berrey et al. 2017). The disputing literature has ably highlighted how court participants create, maintain, and occasionally challenge structural power in society, as in Galanter’s (1974) famous contrast of “one-shotters” and “repeat players.”

If much of the canon focuses on judicial rulings, case aftermaths, or litigants’ “legal consciousness,” some recent scholarship exploits court filings to show how status and power figure throughout the dispute process, shaping who gets to define past events. Hlavka and Mulla (2018), for example, combine records and court ethnography to show how text messages are assigned social meaning in sexual assault cases. The text messages, artifacts from the past, require present interpretation by lawyers to acquire the necessary meanings to influence an adjudicated future. We would push their analysis one step further. Criminal court records can reveal not just how our “digital exhaust” (Zuboff 2019) attains social meaning and becomes legally useful (e.g. Lageson 2021; Brayne 2020), reifying dynamics of power and social standing that exist beyond the courtroom, but also how it could be interpreted differently. Beyond recording disputes over past events, court documents condense the past and render it useful. Moreover, their utility can extend, sometimes unpredictably, to social and organizational actors far beyond the courts. Researchers should thus take care not to treat court documents’ contents as static accounts with fixed meanings. Edelman et al.’s (1999, 2011) landmark studies on “legal endogeneity” illustrate the point. Using federal judicial opinions, the authors show how the implementation of organizational grievance procedures became a successful defense in employment discrimination suits, with organizations then using past disputes to future advantage. As they write, “the professions […] filter and disseminate court decisions, which reinforce and legitimate organizations’ initial structural responses to law. And the circle closes as organizations continue and elaborate their responses” (Edelman et al. 1999: 447). Works on legal endogeneity – the notion that social actors targeted by regulation shape the very meaning of laws intended to constrain them – dramatize not only organizational actors’ temporally dynamic use of court records but the care these data require from scholars. Judicial opinions speak to more than case outcomes. We can see in them how past opinions change present and future organizational behavior, which in turn impacts future judicial decisions. Certainly, we may profit from treating court filings as past artifacts, akin to newspaper archives, old census figures, or personal correspondence (Brown and Shannon 2019) in historical research. But because of common law’s use of precedent, and even nonbinding dicta, a settled “past” might, at any time, be excavated and made to influence the present and future (Kagan 2001).

This brings us to how “the future” is rendered in court filings. Law, Scott Shapiro (2011) observes, is a means for making plans, for encoding a vision of the future. Such futurity can be agonistic, even expressly political, the efforts of individuals or organizations to broaden or constrain theirs or others’ future range of motion. In the policy realm, court documents bespeak powerful actors’ efforts to set down durable social structures and harness the power of path dependence. Sometimes court documents are sites and means for various kinds of prognostication [e.g. of the risk of recidivism (Degenshein 2025; Lynch 2019)]. But generally, “the future” has two distinct meanings in court records. The first concerns case outcomes, the goal toward which each party strives. Sociolegal scholarship has demonstrated that this future is overdetermined in both criminal and civil realms. In the US, the rules of criminal law and procedure, e.g. plea bargaining, largely favor prosecutors despite the ideal of “innocent until proven guilty” (Kagan 2001; Gottschalk 2006; Miller 2008; Lynch 2016). Typically, prosecutors are also better resourced than defendants, with white-collar criminals a notable exception (Hagan 2012; Kagan 2001). In civil cases, large organizations with top-flight legal teams – Galanter’s “repeat players” – are similarly systematically advantaged (1974; Berrey et al. 2017; Felstiner et al. 1981; Kagan 2001; Nelson and Bridges 1999). But in civil law, repeat players use litigation to favorably shape a second horizon: the landscape for future suits (Galanter 1974; Edelman et al. 1999, 2011). Leslie’s (2015) research on the Stanford Financial Group fraud, a $7 billion Ponzi scheme, illustrates how powerful private actors use litigation to enable and constrain future behavior. Court records show how Stanford wielded lawsuits, and its growing reputation for legal pugilism, as a cudgel to thwart its enemies’ future speech, silencing critics at relatively low cost. Judges themselves may also attempt to limit the future use of their own rulings. In Degenshein’s research, for example, the Ninth Circuit of Appeals’ decision supported the government’s use of electronic surveillance in US v Mohamud (2013) but made clear that the matter of surreptitious mass data collection was far from settled, an issue to be raised in future litigation [Degenshein 2024; see also US v Mohamud (2016)].

In both criminal and civil matters, records may preserve surprising inversions of power. As Calavita and Jenness’s (2014) research on prisoner claims shows, the structurally disfavored party does sometimes win. Moreover, such case outcomes have sometimes changed the material circumstances of thousands of prisoners (Feeley and Rubin 1998). On a humbler scale, sometimes the weaker party can trounce a bully. Returning to the Stanford fraud, a wrongfully fired Stanford employee with intimate knowledge of its history took a scorched-earth approach, initiating his own lawsuit against his erstwhile employer (De Maria vs. Stanford Financial Group et al. 2006). His lawyers drafted a list of unsavory witnesses and sought to depose Allen Stanford himself. Served its own medicine, Stanford chose to settle. By showing himself able and willing to blacken Stanford’s reputation – in the eminently discoverable and creditable medium of US court filings and transcripts – the weaker party was able to face down the stronger (Leslie 2015, pp. 135–7). In sum, court documents witness tactical struggles to affect a present and strategically shape the future through debates over the past. Not every study of court documents need engage with each temporal mode, of course. But, as evidenced by past works, recognition of court records’ temporal dynamism enhances theory-building in sociolegal scholarship. Whether recent or remote, and to whatever intensity of conflict, court documents chronicle how differently situated actors manipulate the medium of time. With hindsight’s benefit, scholars can trace which disputant’s hopes, as encoded in the record, were more faithfully borne out. Among other questions sociolegal scholars should ask are: At what timescales do these records operate? Whose pasts, presents, or futures are foregrounded, or altogether omitted? And to what ends?

Dialogic nature

Court documents also exhibit a related characteristic: their dialogic character. In the US context, this trait mostly flows from the adversarial legal process. That is because within a given dispute, every filing is made with an expectation of rebuttal. To trace the arc of a given case is to witness an ordered volley of procedural gambits and substantive assertions, where each move on offense has its defensive countermove. The local meaning and effects of a given filing are thus only discernible within the sequential, adversarial dialog that calls forth the filing. Such dialog, moreover, is not strictly dyadic, as some filings (e.g. motions) call on a presiding judge to rule on fundamental or procedural questions of law, which we discuss below. In legal disputes, then, there is an expectation that documents written for one setting (the present trial) may ultimately be used to make judgments in a future context (e.g. an appeal or a public judgment of character). Accordingly, these documents don’t simply gesture to a future case outcome but offer potential scripts for guiding future conversations, inside the courts or beyond. Within this stylized call-and-response, court filings reveal the production of status, including how actors gain or lose power, credibility, and specific liberties.

To be sure, court filings share this dialogic quality with other document types. Congressional transcripts are infamous for their verbal sparring. The notice-and-comment procedure in federal rulemaking elicits dialog between agencies and stakeholders. All forms of recorded correspondence, official or personal, are arguably dialogic. Indeed, court filings share with many a straightforwardly dialogic structure: two parties convene, contesting one another through motions and replies, and eventually the dialog ends. Yet court filings’ dialogic element is also distinct. That is because they record a present dispute while simultaneously laying groundwork for potential, future disputes. This dynamic is clearest with appeals. In the documents from Degenshein’s research on counterterrorism stings, one can spot lawyers strategically dropping rhetorical breadcrumbs during the initial trial, with an eye toward a future appeal. In US v Mohamud (2013), for example, Mohamud’s counsel received notice late in the trial that some of the electronic evidence used against their client had been collected under a controversial policy enacted in 2008. This policy, FISA Section 702, allows for warrantless electronic surveillance of US citizens and legal residents (US v Mohamud 2013; 2016; Degenshein 2024). Upon receiving this notice, Mohamud’s lawyers argued that the case be dismissed because Section 702 “violates the First and Fourth Amendments, as well as the separation of powers doctrine” (US v Mohamud 2016: 22–23). In response, the trial judge ruled that no violation had occurred. The defense attorneys were likely unsurprised – it is rare that a case be dismissed after going to trial (see Gramlich 2023). However, simply by raising the matter, the defense sowed seeds for a future appeal. Indeed, after Mohamud was found guilty, his team filed an appeal on the grounds that his First and Fourth Amendment rights had been violated, which the Ninth Circuit heard 2 years later, eventually siding with the government (US v Mohamud 2016).

The expectation that a defense pursue appeal, and that the initial or even secondary case outcome may not be the final word (Kagan 2001), typifies the open-ended yet choreographed “dialog” of court documents in our adversarial system, as we see it. Parties file motions that require immediate response from the opposing party and judge. But a motion may be filed, or an objection may be raised, as much for an anticipated future audience as the present one. In Mohamud, the Ninth Circuit was no longer ruling on Mohamud’s guilt or innocence, but on whether his constitutional rights had been violated. By dropping legally relevant breadcrumbs in early filings, then, lawyers can attempt to spur, or forestall, dialog between a presiding judge and a future appellate panel.

Considered more broadly, judicial review in the US allows judges to affirm, modify, or negate prior judicial rulings, and even assess the legal soundness of legislative or executive output (Kagan 2001). Court filings’ dialogic structure, therefore, not only has a dependable linear quality but, we maintain, can more dynamically spark dialog with other branches of government and even the broader public. The upending of legal precedent, for example, often marks an inflection point in law’s development, something we can see clearly with Roe v Wade (1973). Even before Dobbs v. Jackson Women’s Health Organization (2022) overturned it, smaller challenges to Roe helped antiabortion activists advance their legal cause (Wilson 2013). These culminated in the Dobbs decision. But Dobbs did not extinguish Roe’s legal, political, or cultural relevance. As binding precedent, Roe demanded that the Court explicitly, repeatedly addresses it in their Dobbs reasoning, clarifying why they were breaking with its 50-year holding. Even beyond Roe’s resilient legal discourse, we argue that the case – a collection of legal documents and judgments – remains a political and cultural touchstone. In her capacity as Vice President, for example, Kamala Harris gave a speech marking Roe’s 50th anniversary after its overturning (Harris 2023). Later, as a 2024 presidential candidate, Harris vowed to codify Roe’s protections into federal policy (Hoffman 2024). Further, mainstream media coverage of the Dobbs decision sometimes wholly omitted explicit reference to Dobbs itself, foregrounding instead the “overturning of Roe v Wade” (e.g. Totenberg and McCammon 2022). In each example, Roe, and the judicial reasoning behind it, has remained in dialog with legal practitioners, politicians, activists, and the news media.

Although law and society research does not often name this “dialogic” trait as such, it has nonetheless greatly impacted theoretical claimsmaking in the field. This is particularly true for the contentious literature on “rights litigation.” Some scholars argue that the courts play an important, sometimes direct, role in social change through rights litigation (e.g. McCann 1994; Feeley 1992; Ashar 2007; Feeley and Rubin 1998; Wilson 2013). Others are skeptical, claiming that the courts’ role, at best, is symbolic and, at worst, actively impedes social change (e.g. Albiston 1999; Bell 2004; Rosenberg 2008; Scheingold 2010). Yet both camps acknowledge the courts’ sway. Legal rulings both dramatize and, sometimes, concretely aid or effect social change, codifying or challenging existing practices and beliefs. Rulings are not uttered into the void. Rather, once filed, these official records become resources.

Court documents’ dialogic qualities can be observed in more subtle ways. For example, through their citation practices, lawyers invoke extant law to support their arguments. They interpret statutes and previous court rulings narrowly or broadly, as benefits their clients, and judges assess the viability of those interpretations. Sometimes this strategy has massive cultural impact, as in the case of Dobbs and Roe. But even law’s rote application can be dialogic in a sense, with every echo of the prevailing view cementing its taken-for-grantedness. Simard (2020), for example, has shown how law’s formalism, as enacted through ritual case citation, can inadvertently dignify repugnant views. He finds that contemporary lawyers and judges still cite Antebellum slavery cases when bolstering banal propositions of present-day property law (Simard 2020). In the absence of active disavowal, he argues, such dialog with outdated cases can tacitly bolster retrograde politics or inflict dignitary harms on third parties. Court documents’ dialogic nature, therefore, allows us to glimpse what kinds of arguments – doctrinal, but also rhetorical, moral, and political – tend to carry the day, or founder, within given sorts of dispute.

As Wilson’s (2013) multimethod research illustrates, court records are artifacts of struggle across time. The dialog within their back-and-forth reveals much about what a given legal party hopes to accomplish, whether their ploys are primarily substantive or procedural, case-specific or aimed at shifting the legal landscape, as we saw in the previous section. Yet, while major court cases like Roe spotlight the profound and ongoing dialogic quality of court documents, it is sometimes the least conflictual parts of court records that are most instructive. For example, which sets of facts go jointly stipulated or tacitly uncontested among the parties? Or how do filings display or challenge taken-for-granted ways of describing social life? Precisely since disagreement is their default mode, court records furnish unwitting snapshots of banal consensus and hegemonic thought.

Multiple truths

In part because it emerges from an adversarial process, “legal truth” differs from other kinds, including personal, moral, cultural, institutional, and scientific truths. In US courts, the judgments of “factfinders” (i.e. juries or judges) can supersede those of actual experts (Haney López 2006; Jasanoff 1995; Lynch 1998). As such, the legal space is one in which scientifically unproven – sometimes disproven – forms of expertise get their cultural foothold, promulgating views that are merely defensible, but not necessarily “true” (Burns 2001; Conley and O’Barr 1990; Jasanoff 1995). Though a range of powerful social actors routinely stretch the truth, factual claims ratified by courts and preserved in their records can codify inequity by tying legal “truths” to material outcomes (Mnookin 2001).

That legal truths differ from scientific truths is well established in scholarship. STS scholars have observed the difference between “open ended” scientific processes and “closed” or decisive legal processes as essential features of their respective forms of truth (Jasanoff 1995; Lynch et al. 2008; Lynch 1998; Cole 2001). Nowhere is this difference clearer than during Daubert hearings, where a judge decides whether a potential expert witness’s evidence is both reliable and relevant to the present case. Among other factors, judges weigh whether an expert’s methods or claims are accepted within their discipline. Yet scientific claims rarely enjoy universal acceptance within expert communities, as evinced, for example, by opposing lawyers’ tussle in the OJ Simpson trial over the meaning of DNA test results (Lynch 1998). In their expert’s Daubert hearing, the prosecution filed an overview of both the “technical background” of DNA sequencing and case law supportive of its use at trial (Lynch 1998). In their response, the defense elided technical questions altogether, invoking instead DNA’s unsettled status among scientists. Looking “beyond forensic science” (1998: 836) and casting doubt on DNA testing’s “general acceptability,” the defense cited credible scientists who had “publicly opposed using DNA profiling in criminal investigations” (1998: 839). As in all Daubert hearings, though, the judge was the arbiter of scientific credibility. This episode, then, and Daubert hearings more generally illustrate how “scientific” truth must be established anew at each trial (Lynch 1998). An expert or body of science greenlighted for one trial may be excluded from another, pending the strength of lawyers’ arguments and the presiding judge’s inclinations.

Scientific truths are not the only kind to butt up against the law’s need for resolution. Trial attorneys routinely use cultural tropes, stereotypes, and truisms in their arguments, even if these culturally resonant frames are scientifically null or even disproven. For example, in trial transcripts from Degenshein’s research, attorneys invoke terms like “double agent” or “internet avatar” as though they boast scientific validity or agreed upon cultural meanings (Degenshein 2024). Because the adversarial process rewards what judges or juries deem believable rather than what is factually or scientifically “true,” lawyers can use a broad repertoire of truths to argue their case at trial, which in turn becomes the basis for future legal reasoning. Scholarship on legal discourse has placed particular emphasis on opening and closing arguments, during which lawyers provide the court a broad conceptual framework for synthesizing the myriad evidence presented by experts and witnesses (Burns 2001; Bennett and Feldman 2014). These bookends establish the moral stakes underpinning the evidence, but they also reveal lawyers’ efforts to implant culturally resonant frames in jurors’ minds that will influence their assessment of the case (Burns 2001).

Which ideas may win the day cannot be divined from how those ideas circulate in other contexts. Indeed, postmodern theorists have long emphasized the importance of context for assessing linguistic meaning, warning that context is never stable (Derrida 1988; see also Presser and Sandberg 2015). Sociolegal scholars in the New Legal Realism camp have generated complementary insights, laying bare the epistemologies, cultural contexts, and communicative norms that continue to frustrate attempts to translate between law and science (e.g. Mertz et al. 2016; Riles 2006; Talesh et al. 2021).

Yet, because courts comprise an arm of the state, we argue, they don’t just test the legitimacy of various truths and labels – they give these legal standing and thus lend them coercive force. Court data thus present us with invaluable tools for studying law’s power to validate, launder, or discredit other forms of knowledge. Haney López (2006) illustrates this in his work on the legal construction of whiteness in early 20th-century federal courts. He contrasts “common knowledge,” or “popular, widely held conceptions” about racial categories, to scientific understandings about race from the time. Using judicial opinions, he shows how the federal courts repeatedly sided with the “common knowledge,” giving racist, scientifically dubious ideas official standing and cultural force. His research thus illustrates how questionable, sometimes harmful “truths” have gained life, and even a veneer of scientificity, in the courts where they have had tangible consequences for claims about immigration, detention, citizenship, property ownership, criminal propensity, and mental acuity.

Owing to US law’s adversarial build, the primary collision of truths it effects is that between parties. Whether in jury or bench trials, each party must work to put before a factfinder the more persuasive story. Yet, though each asserts their version within a broader “dialogic” frame, their goal is hardly collaborative truth. The parties may obfuscate facts, mischaracterize the other’s positions, and even impugn opposing counsel’s and witnesses’ ethics and competence (Berrey et al. 2017; Degenshein 2024; Kagan 2001). Functionally, court disputes serve to proceduralize away social conflict. But epistemically, they rest on the striking premise that legal truth must be pulled from the wreckage of conflicting accounts.

Given our adversarial system, then, one might assume there are never agreed-upon truths that emerge in court documents. Not so. Across genres of suit, parties often “stipulate,” or jointly recognize, a common factual bedrock (see also Burazin et al. 2018). In cases where the accused asserts an entrapment defense, for instance, both parties stipulate that the defendant committed the criminal act (Degenshein 2024, 2025; see also Norris and Grol-Prokopczyk 2015; Said 2010; Frampton 2013). Where disagreement arises is whether law enforcement officials unlawfully induced the outcome through their interventions. Thus, while adversarial law typically heightens epistemic disagreement, points of convergence can be just as telling about the larger social context from which they emerge.

Legal filings represent bouts of social conflict as refracted through officialdom. And in conflicts, social actors often pull from a grab-bag of “truths” to gain advantage. Court records, then, reveal important social truths even as they might muddle the truth. To researchers, they offer a unique view on – and countless means to study – how power and knowledge practically interrelate. Moving forward, sociolegal scholars should, therefore, probe links between the “truths” parsed in court documents and those offered in other social arenas. Where and how does expertise translate? Do certain experts and forms of expertise betray biases toward specific parties (e.g. plaintiffs, government)? If so, why? And what is the relationship, if any, between expert witnesses and case outcomes? Answers to these questions are worthwhile in their own right but are also consequential for policy discussions around who has access to justice.

Multivocality

At our most ideological, citizens and scholars alike imagine “the law” to speak in a singular voice to “society.” One glimpses this vision in Scheingold’s (2010) sketch of the “myth of rights,” average Americans’ faith that high-court utterances can on their own recast social relations. Court filings, however, give the lie to this notion. In their messy actuality, such documents prove to be richly multiple, at both their points of emission and sites of reception, in ways that are a boon to researchers. Consider the front end, where such documents afford scholars countless ways to study both lawyering’s social effects and its internal dynamics. Court filings’ formal qualities – especially, their unified authorial voice – obscure the friction and negotiation involved in their making. Two boundaries merit mention.¹

First is the line between lawyers and clients. Though court filings purport to bear the named parties’ interests and choices, they index the share of power between clients and counsel. As classic works attest, lawyers first have to find it worth their while (Johnson 1981; cf. Barclay and Chomsky 2014), or consonant with their values (Engel 1984), to take on a client’s case. Conversely, some clients, aware that filings comprise both procedural and emotional thresholds, must be nudged by their lawyers to fire the next paper salvo (Sarat and Felstiner 1995, pp. 43–5).² Filed cases, and the documents they beget, are thus a subset, points where client and lawyer interests find some alignment. Even then, court documents often see client preferences subordinated to doctrine’s dictates and lawyerly strategy (Merry 1990; Sarat and Felstiner 1986, 1995). Where complementary data, such as ethnography or interviews, can be had, the space between what clients think, feel, and want, and what their lawyers finally commit to paper, is a rich vein to mine for comparative insights on how principals and their legal agents manage their relationship. In a looser but still meaningful sense, court filings are as much products of wealth distribution as of their named authors. Given, as we know, that most “legalizable” social conflict is simply endured or settled outside of court (Felstiner et al. 1981; Galanter 1974), cases with lengthy document dockets are typically those where one or more parties has ample resources. Thus, both the claimsmaking that fills out the aggregate legal-textual corpus and the forbearance and silences that mark its borders, we contend, are products of broader political-economy and institutional design.

A second boundary runs between lawyers on the same side. Long-term trends show a growing share of litigation is conducted by firms and in teams rather than solo practitioners (Heinz et al. 1998). If we include state actors, we might surmise that most court documents are filed by corporate bodies (e.g. law firms, agencies, prosecutors’ offices), which means that their production typically involves collaboration. The content of a particular filing – its emphases and omissions, how cautious or aggressive its approach – may result from considerable politicking and negotiation. Penned in one voice, however, court filings obscure the divisions of labor behind their production. Such divisions typically track differences in rank and specialty, foisting rote and dull work on paralegals and junior attorneys, with prestigious tasks (e.g. doctrinal and rhetorical strategy, trial lawyering, interfacing with powerful clients) reserved to senior attorneys (Flood 2013; Nelson 1988). Here also, if triangulated with other data, court filings furnish means for comparison between the polished, univocal speech that legal convention demands and the often messy, hierarchical labor that produces it.

Consider now the back end. Just as they may hide multiple authors, so too can court documents speak to multiple audiences. These audiences can be expert or lay, intended or unintended, comprised of participants in the suit (i.e. litigants, lawyers, judges, jurors, and a dispute’s broader stakeholders) but also those in future rounds of litigation (e.g. appellate judges); other courts; fellow branches of government; journalists; scholars; policy outfits; commercial interests; background check and due diligence firms; the polity writ large; and some hazily imagined “posterity.” The staggering range of actors who can lay claim to, benefit from, or suffer the effects of court documents is itself an argument for their sociological import and the need to theorize their use. Discerning who these actors are, and which streams of discourse are meant for whom, is crucial for understanding the knowledge production that court documents effect.

The most scrutinized court documents are likely judicial opinions, especially from high-court judges. Often, judges seem not only keenly aware of their broad, varied audiences but of the weight of their role. Beyond their substantive rulings, such judges are tacitly charged with safeguarding the courts’ reputation, particularly from criticism, whether about judicial politics and overreach or the courts’ role in perpetuating inequality. While Supreme Court justices are likely most attuned to how their writings will be quoted, covered, debated, and critiqued, lower court documents often display a rhetorical flair aimed beyond the trial parties. This is illustrated in sentencing hearings from Degenshein’s studies of counterterrorism stings. At the sentencing for US v Kasimov (2019), for instance, the judge justified the sentence he imposed and explained why he did not impose the harsher term the government sought. In so doing, he pointedly criticized the common US practice of the “trial penalty,”³ stating,

While this Court is aware some believe “the first one to the table gets lunch, the last one to the table is lunch,” our Constitution and laws wisely refute that concept: to impose the so called “trial penalty” is to invite the cannibalization of our right to trial. This Court declines to dine at that table. Guilty pleas and prosecutorial efficiency play important roles as servants in the administration of justice; but guilty pleas and prosecutorial efficiency must never be allowed to assume the dominant role of masters in the administration of justice. (2019: 13)

His statement can be read as an admonishment of that trial’s prosecutors and as rationale for imposing a lesser sentence than the government sought. But its rhetoric suggests that the judge hoped to address the topic more broadly, raising it as a legal and ethical concern for audiences beyond the case.

It is not just judges, though, who aim court filings beyond the parties to suit. The full scope of address may only become clear as broader contexts come into view. Leslie’s research on the Stanford fraud presents such an instance. When authorities shuttered Stanford in early 2009, the Securities and Exchange Commission (SEC) filed a complaint against multiple Stanford entities and natural defendants, detailing the scheme’s history and mechanics (SEC v. Stanford International Bank). Concurrently, the SEC assisted in the Justice Department’s criminal probe and indictment of Allen Stanford and co-conspirators (US v Stanford et al.) and aided the appointed receiver’s efforts to recover investor funds. The manifest audience for these filings were participants in these litigation streams, plus some collective stakeholders (“The People of the United States” and “the investing public”). But there were other unspecified addressees. Some of the relevant context was already palpable. The US was months into a grisly economic downturn, and the recent collapse of Bernard Madoff’s scheme had further bruised public confidence. Yet, the full social meaning of the SEC’s filings would only grow clear months later. In early 2010, the SEC’s Office of Inspector General released a blistering report about the SEC’s 12-year failure to investigate Stanford (SEC-OIG 526). The response of media, investors, and legislators was scathing: withering op-eds tumbled forth, a former SEC Enforcement head was censured for ethical lapses in the matter, and several Congressional hearings followed (Leslie 2015).

As this background came into focus, the retrospective meaning of the SEC’s 2009 filings shifted. It grew impossible not to read these filings as, in part, desperate efforts to head off the damaging news it knew was drawing near about the SEC’s ineptitude. An unacknowledged audience for these documents, we submit, were those people whose hands can tighten or slacken an agency’s purse-strings: the US Congress. Moreover, read together, the SEC’s 2009 filings and its 2010 mea culpa offer us a deeper lesson. They suggest that in striving to grasp the social meaning of this documentary genre, it behooves us to consider not only a given filing’s obvious addressees but also its plausible ones. This consideration is most relevant for cases with precedential or even constitutional implications but it probably ought to guide all our inquiries.

Whether audiences are expert or lay, intended or not, their variable and contingent uptake of court records provides us analytical grist that we have only begun to mill. Scholars of law and the economy, for instance, might better map how closely, how concertedly, and to what effects commercial actors monitor their legal environment via documents produced in court suits, administrative proceedings, and arbitration fora. Theorists of “legal consciousness” have shown that ordinary people show a loose grasp of formal law’s structure and language (e.g. Ewick and Silbey 1998; Merry 1990; Silbey 2005). Though we know that court “paper” signifies to laypeople law’s power and exclusivity, less understood is by what pathways its contents percolate into shared culture and consciousness. We have much to learn, for example, from scholars of law and media regarding not just how fictional takes on disputing shape popular perception, but also about the growing influence of televised “legal analysts” in translating court proceedings and constitutional events for public consumption (Haltom and McCann 2004), especially in contexts of waning trust, worsening polarization, and media consolidation. The use of tools like nondisclosure agreements, the sealing of records, and expungement to limit the reach of certain kinds of legal data (e.g. Myrick 2013; Otte 2020; Wegar 2008) could also be more fully mined.

Social productivity

Lastly, court records grant access to what we term law’s social productivity. Both natural persons and organizations are constituted as social beings in part through law. It is in this sense that law is socially productive: it engenders identities, relationships, behaviors, and structures that might not otherwise exist. Certainly, it is with varying grades of agency that social actors get mixed up with law. Thus, the “duty-imposing” rules of criminal law, the “power-conferring” rules of civil law, and administrative agency rules (Hart 1961; Durkheim 1933; Rubin 1989) yield records that, epistemically and ethically, require distinct approaches. Across law’s motley forms, however, court documents figure as both tools and artifacts of law’s social productivity.

Scholars have long chronicled law’s power to reify, or even conjure into being, social groups, often with repressive results. Epistemically, legal taxonomies rival those of science, sorting people into ethnoracial (Haney López 2006), sex and gender (Vogler 2021), immigration (Gowayed 2020; Jensen 2023), biomedical (Kirkland 2016; van Wichelen and de Leeuw 2022), and criminal categories (Pager 2007). Such labels often harden into social statuses due partly to the durability and accessibility of the documents that buttress them. However, in spotlighting the law’s repressive productivity, scholars have tended to undersell its productivity more generally.

Authors as different as Hart (1961) and Durkheim (1933) observe that law in modernity does not merely – or even principally – impose duties and disabilities but also reshapes social relations by conferring on people new forms of agency and social being. Sometimes via performative pronouncements (e.g. “I declare you married”) (Austin 1975) and nearly always attended by paperwork, the law enables changes in status and legal personality that equip us to do things in the world we otherwise could not. Such transformations include the establishment or dissolution of marriages, guardianships, contracts, business partnerships, and corporations. They encompass certification and licensure, permit us to assert states of “good standing” (e.g. being current on one’s taxes, restaurant health inspection, or child support payments), and allow for the disburdenment of obligations (e.g. bankruptcy). Given the breadth of law’s productivity, and how often such phenomena lay the bases for suits, court records grant unique access to the range of means available for both constraining and expanding social agency through the power of legal form.

Moreover, it is precisely this range in the forms of law/society contact that court records invite us to exploit. We suggest that law’s productivity benefits scholars in four ways: it reveals the classificatory heart of law’s coercive power; it furnishes access to otherwise inaccessible social situations; it creates archives of countless actors’ social being; and it affords invaluable means for data triangulation; we elaborate each in turn.

First, law both reifies prior, and produces new, categories of being. Through its prerogative to sort, law erects hierarchies of belonging and respect whose stakes can prove dire. Older works on citizenship (Bosniak 1999; Somers and Roberts 2008) and recent studies of asylum and immigration bureaucracy show how law creates those gradations of inclusion and shelter (Gowayed 2020; Shiff 2020) on which social flourishing hinges. Shiff’s study of the 1990s spike in gender asylum claims, for instance, examines Board of Immigration Appeals decisions, tracing changing standards of “deservingness” from those that weighed contextual factors (e.g. political persecution) to criteria based on “immutable” ascriptive traits (2020) – shifts that favored asylees with gender violence stories over those fleeing war or gang strife. Recent works on sexuality, punishment, and disease show how law deems certain identities, acts, and diagnoses permissible or deviant (Hoppe 2017; Vogler 2021). Scouring 78 trial transcripts, Hoppe (2017) demonstrates how diffuse anti-gay sentiment can harden over a trial (from voir dire to sentencing) into institutional policy that renders HIV status a criminal rather than medical matter. Similarly, from classic audit-based employment studies to research on criminal record expungement and nondisclosure agreements, scholars trace how prosecutions and the documents they secrete produce stigma that can ossify into lower-caste status (Myrick 2013; Pager 2007) as well as how parties attempt to forestall the socially productive power of court records (Myrick 2013; Otte 2020). Though countless document types reveal a taxonomic impulse, court records stand out for how tightly and consequentially they hitch classificatory authority to coercive state power, producing high-stakes identities and statuses that appear natural.

Second, law’s productivity provides lines to social conflicts that otherwise would be spatially or temporally inaccessible. Trial records, for example, can serve as surrogates, however flawed, for courtroom ethnography at times when physical attendance is impossible, whether for political, epidemiological, or more simply practical reasons.⁴ Certainly, litigation is not a neutral medium. Court filings are stylized renderings of conflicts that might not have transpired at all but for the institutional genre. Wealth and power frame the self-selection mechanisms that determine who gets, or is forced, to leave a trace. Nevertheless, US and similar legal systems are historically so generative of records that these constitute, albeit partially, a trove of preserved social encounters. The access records grant to remote spheres of life is most powerfully illustrated in historians’ reconstruction of bygone disputes, like their accounts of African Americans’ litigative strategies in the Antebellum South (Welch 2018) or, more extreme, studies of ordinary people’s legal petitions in antiquity (Bryen 2013). In such cases, court records may be the sole evidentiary link to revelatory events. In short, though law is often repressive, and though legal conflict is often foisted on weaker parties, court records bear witness to boggling arrays of social actors’ attempts to assert, preserve, or recover agency.

Third, law’s productivity affords us ready archives of countless actors’ social being. Whether in criminal, civil, or blandly administrative matters, court records reveal people’s lives as woven on law’s loom: their plans, conflicts, failures, misdeeds, their making and unmaking of familial or business bonds, their changes in individual or corporate form or status. In providing them novel forms of agency, and by chronicling their subsequent conflicts in written record, the law is doubly productive, laying a bounty at scholars’ feet. Simply, much of social life in modernity is innately sociolegal and court filings are among its richest annals.

Crucially, law’s productivity allows us to leverage powerful others’ efforts for scholarly gain. Well-staffed prosecutors’ offices and deep-pocketed law firms often put massive resources toward piecing together the doings and histories of relevant parties, producing accounts of great scope and detail. Through subpoena and discovery, they compel disclosures that consign to public record narratives that might not have seen daylight. If slanted, the results are often a boon to scholars interested in those parties, who could not dream of assembling such data themselves. In his fraud research, for instance, Leslie benefits from the combined knowledge-making efforts of the SEC, DOJ, sundry private law firms, as well as their subcontracted experts (e.g. forensic accountants, IT specialists) (Leslie 2015; 2022a, 2022b). In enabling new forms of agency, then recording what transpires, the law’s social productivity provides scholars with a bountiful corpus for study.

Fourth, law’s productivity presents useful means for triangulation. Court records can supplement quantitative, ethnographic, and interview data or serve as epistemic checks on these. Quantitative scholars, for example, have only started applying the tools of machine learning, topic modeling, and natural language processing to legal texts (e.g. Aletras et al. 2016; Medvedeva et al. 2020; Ruhl et al. 2018; see also Bernstein 2020). For qualitative scholars, recent methodological debates on interviews and ethnography (e.g. Jerolmack and Khan 2014; Lubet 2017; Pugh 2013; Tavory 2020) underscore the need to bolster both approaches with other data. Court records may permit us to check respondents’ accounts of what they say they do (interviews) and what we observe them doing (ethnography) against otherwise inaccessible evidence of past behavior, from which we can draw confirmatory or complicating inferences. Finally, court records may help prepare researchers, ahead of fieldwork, to ask the right questions (interview or theoretical), or even to weigh the personal risks inherent in fieldwork itself (Hanson and Richards 2019; Nilan 2002; Sluka 2012).

As detailed below, court filings should never be taken at face value. Their use, as with any documentary source, requires sensitivity to context and to the “fluidity” of the law (Derrida 1988). Nevertheless, they may well comprise the most underexploited data source for social scientists relative to ease of acquisition. Additionally, for sociolegal scholars, court documents often form the unspoken backbone of our objects of study. As the output of formal legal processes, they are arguably “our” data, and we are well positioned to use and interpret these data as social objects. What’s more, they transcend disciplinary and methodological boundaries within sociolegal studies, uniting our diverse fields with a common referent point. Even so, we hope that researchers who do not don the “law and society” label will jump into the fray, exploring the meanings and implications of these documents as well.

Practical, epistemic, and ethical pitfalls of using court documents as data

Despite their virtues we have just described, court records, like any data, entail specific challenges. Their use involves practical, epistemic, and ethical pitfalls. Our aim is not to ward readers off from using court records, but to encourage reflexivity about their limitations. Accordingly, we once more raise a set of questions intended to spur discussion rather than offer the final word.

Practical pitfalls

Though produced using public funds and ostensibly publicly available, US court records are often difficult to obtain. LexisNexis, Westlaw, Bloomberg Law, and lesser-known databases are typically available only to law school affiliates or by law firm subscription. The public alternative, PACER (Public Access to Court Electronic Records), is expensive enough on a per-page basis to discourage the open-ended explorations that often yield good data.⁵ Freedom of Information Act (FOIA) requests provide an alternative path to some records, but FOIA is also cumbersome, potentially expensive, and often a frustrating experience for scholars. Nonprofits like MuckRock and online tools like FOIA Machine automate parts of the process, but scholars may still need FOIA attorneys’ help for complex requests. Certain legal records, particularly state and municipal, necessitate trips to courthouses or storage sites, entailing not only document fees but travel expenses. In our experience, lawyers and journalists involved in cases are sometimes willing to provide trial transcripts to researchers, though in ongoing litigation, or where appeals are anticipated, lawyers may be reticent. Additionally, direct web searches via Google and like engines, using tailored terms and parameters (e.g. “filetype:pdf”), sometimes yield desired documents (especially indictments and complaints), but are unsatisfying alternatives to private databases, particularly for obscure filings. Cheaper subscription-based PDF repositories (e.g. Scribd) may also yield fruit. Finally, some records of interest may be sealed, requiring petitions and court orders to unseal. Others, such as grand jury hearings, are secret by design and may remain so.

Once obtained, documents may present material challenges. One may have to subject PDFs to optical character recognition to enable text searches or have to change documents to other textual formats for ease of coding. Quantitative scholars interested in natural language processing may need to consider the structure of documents when deciding not only what questions to ask, but also in which documents – or portions of documents – to look for answers. Historians often face court records that are deteriorating, handwritten (often in inscrutable cursive), as well as spotty in coverage across the arc of a suit (see, e.g., Welch 2018). To locate, organize, and render such documents legible requires grit and ingenuity, and, not infrequently, the kindness and insights of archivists.

Overall, the price of access to legal databases remains the biggest barrier to more widespread study of court records. We hope that as more researchers across the ranks take up these data, at least the better-resourced university libraries may make these services more widely available in response to demand.

Epistemic pitfalls

Like most documents, court records enjoy the cachet of officialdom and can seduce researchers by purporting to reflect events as they “really happened.” Thus, they require the skepticism and attention to contexts of production that all documents do (see, e.g., Martin 2017; Riles 2006).

But court documents feature more specific challenges. First, it is a truism of sociolegal scholarship that most potentially suit-worthy conflict never makes it near a courtroom. In a broad sense, court disputes are aberrant (Felstiner et al. 1981) and, thus, we are limited in what inferences we can draw from their records. Further, suits that go to trial – yielding the most “complete” documentary record – are fewer, and likely even less representative (Berrey et al. 2017; Felstiner et al. 1981). Beyond questions of representativeness, court records are written in forbidding language (“legalese”). To best situate them, researchers may need a minimal grasp of civil or criminal procedure, as well as a sense of the judiciary’s structure that some scholars, especially those outside the sociolegal tradition, may lack. Related, since they emanate from the contact point between “society” and “law,” court records demand that researchers be disciplined in how they frame that society/law interaction. To be sure, lawsuits are an expression of real conflict, but court records present such conflict not in raw form but as refracted through law’s prism and prioritizing law’s needs (e.g. foregrounding justiciability and proper procedure over colloquial notions of justice). Scholars may be tempted to impute the aims and positions expressed in court records to the named parties but need constantly remind themselves that, like the strategic choices that underlie them (Sarat and Felstiner 1986), court records are a co-production in which parties’ lawyers, and the law’s dictates, take the lead.

Perhaps the thorniest epistemic challenge stems from Anglo-American law’s adversarial structure. In court records, every assertion is partial, in both senses, and made within an ethical frame that elevates lawyers’ zealous advocacy over various forms of “truth.” The researcher, hoping to reconstruct events, must constantly be on guard to discern the merely plausible from the factual. Even at their most putatively transparent, court records are deceptive for all they leave out of frame. Humanist scholars are especially wary of this. Hartman, for example, has long noted how legal archives efface the voices and experiences of slaves (2008, 2022). This problem is hardly consigned to the past. Recent works alert us to how much trial transcripts elide interactions and data of sometimes vital social-scientific interest. These include the backstage, off-the-record talk, agreements, and squabbles that occur routinely in court, as well as details about body language, diction, tone of voice, or trial participants’ demographic traits, data often crucial for assessing the operation of power in the courtroom (Hlavka and Mulla 2021; Kaufman 2020).

Ethical pitfalls

The law is coercive. Virtually every ethical pitfall surrounding our study of court records has basis in this fact. Whatever form it takes in a given case, scholars must contend with law’s coercion and, where possible, mitigate its effects. This challenge is both epistemic and ethical. This article’s authors and its likeliest readers breathe the sedative air of legal liberalism, whose “majestic equality,” Anatole France once quipped acidly, “forbids rich and poor alike to sleep under bridges, to beg in the streets, and to steal their bread” (2011 [1894]). Scholars working in this milieu must actively resist the soothing illusion of parity between disputants that court records present.

Archives’ tendency to reproduce inequalities has long been a point of scholarly concern, particularly among critical theorists. Foucauldian historian Carolyn Steedman, for instance, warns that the archive is “an idea not a place” (p. 321), arguing that knowledge is produced after the archive, rather than in it (2011: 321). Building on Derrida (1995) and the Subaltern Studies Collective, anthropologist Ann Stoler (2002:91) argues that it’s not just a question of “trusting” state documents, but whether the “conditions that produced those documents” have “altered [our] sense of what trust and reliability” mean. Similarly, inspired by Guha and Spivak (1988), Subaltern historians observe that colonial archives have been central to the making and maintenance of empire. Dayan (2011:xiii) has noted the “ambiguous” power and “witchcraft-like” seduction of court documents, in which legal practices shade into “rituals and belief.” We urge sociolegal scholars to heed these warnings.

We have long known that most court cases in the United States convene parties of unequal stature. This is markedly so on the criminal side. In all but a few topic areas (e.g. white-collar, corporate, and antitrust), legal cases tend to feature huge power and resource disparities between parties (Hagan 2012; Kagan 2001; Lynch 2016; Pfaff 2017). Indeed, even in corporate crime cases, the inherited know-how, intellectual wattage, and resources of some prosecutor’s offices (the famed Southern District of New York being an exemplar) dwarf those of most criminal defense firms. For ordinary property and violent crime cases, the differences are starker. Moreover, like criminal suits, civil suits also tend to pit better-resourced “repeat players” against “one-shotters,” creditor/debtor suits being the classic form (Galanter 1974). Thus, court records are typically products of lopsided, often inequitable relations. What does this observation require of law and society scholars?

Crucially, that we take care around two matters: triangulation and anonymity. Beneath its epistemic benefits, triangulation hides an ethical hazard. Take criminal law. It is a truism that the socially disadvantaged (e.g. racial minorities, the mentally ill, the poor) are more likely to have records, both real and, sometimes, made in error (see Lageson 2021). They are also more likely to be wrongfully accused and convicted, and less likely to enjoy effective counsel than privileged defendants (Berdejó 2018; Gross et al. 2023; Hashimoto 2011; Kagan 2001; see also Hoag-Fourdjour 2023). Their disadvantage is only compounded by the production of records that distill their social being to the criminal act. Given their greater likelihood of ensnarement in the system, the upstream possibility of prosecutorial misconduct and police perjury, the hurdles to expunging their records (Myrick 2013), and the cachet of court documents, do we not reproduce systemic injury by using their records as data? Though we must answer yes, triangulation offers a measure of mitigation. Where possible, we suggest, research based on court records should be checked against other forms of data (e.g. ethnography, interviews, newspapers, or surveys) so we might capture broader swaths of subjects’ lives and avoid mirroring the state’s reductive, often epistemically violent, point of view (e.g. Hartman 2008, 2022). Although, as Stoler (2002) cautions, reading state documents only “against the grain” may neglect those tensions and disagreements that sometimes rupture the calm of official accounts and reveal much about the workings of power.

Triangulation also cuts the other way. In calling for court data’s broader use, we have hinted that researchers might, as a matter of course, trawl for court records linked to those respondents they study via other methods. Let us stipulate that, as an epistemic matter, where court records can be had, they should. Does this maximalist take on triangulation not feature its own ethical snags? Searching for one’s study subjects’ legal records might yield valuable data that confirms, disconfirms, or complicates field observations or claims made in interviews. Yet, we might rightly feel qualms about conducting undisclosed “background checks” on study participants, particularly where individual rather than corporate subjects are concerned. As a first cut, scholars should limit their use of such data to records that are responsive to their study’s themes and aims. Then, they might look to the concept of triangulation itself to lessen harm. Given the adversarial structure of Anglo-American disputing, the use of a complaint or indictment, for instance, ought to be checked (i.e. “triangulated”) against the opponent’s filing in response. Though an imperfect fix – in both quantity and quality, filings merely index the parties’ resources – this goes some way toward blunting partiality.

Finally, we can lessen harm to respondents by safeguarding their anonymity even when using court records. Certainly, this is not always required. Where a particular case has achieved publicity or notoriety (as in Leslie’s focal fraud and Degenshein’s counterterrorism stings), there is no ethical gain in anonymizing the principals. Sometimes, as with Degenshein’s cases, a critical approach may ethically weigh in favor of using real identities, to offset biases in existing news accounts. However, where court records are used to supplement anonymous data, they must be subsumed into the latter’s standards. Practically, this means that researchers may describe the general build, arc, and disposition of a relevant suit but cannot cite to the specific record and must omit sufficient detail as to ensure a cold paper trail (cf. Murphy et al. 2021).

Conclusion

Mining insights from canonical and contemporary sociolegal works, including our own, we have specified five attributes that, taken together, set court records apart from other documentary data: their multitemporality, their dialogic nature, the multiple truths they embrace, their multivocality, and their social productivity. These features offer scholars countless pathways for research, while also posing practical, epistemic, and ethical pitfalls. We argue that taken together, these features support two claims. First, that court documents are a natural meeting point for sociolegal scholars, a common source to mine across disciplines, methodological traditions, and theoretical orientations. While sociolegal scholars hardly hold a monopoly on their use, court documents have been central to the rich and diverse corpus of scholarship discussed here, shaping the field as we know it. Second, each of the five attributes on their own, but especially together, show how the courts reify the dynamics of power that make social inequalities so durable. Rather than reflect the ideals of “blind justice,” these documents lay bare how powerful actors use the courts to their advantage, using coercive state power not just to beat back opponents, but to make structural inequalities appear “just.”

Additionally, while court documents are a long-cherished source of sociolegal data, their use remains undertheorized. Perhaps for this reason, these data have often been relegated to secondary or tertiary status outside historical research. Our aim is that scholars might use the five traits we have identified as rough guides for both extracting meaning from court records and for assessing, and even confidently theorizing, such records’ social effects. We hope those traits will serve as signposts that both novice and practiced users of court records may employ to orient their efforts. Indeed, we contend that court records’ distinct features lend them to greater analytic use, including as the methodological main attraction – but such use demands care.

We also hope that the foregoing ideas spur conversation, as there is much more to be said. Notably, we are eager to see the discussion extended to other legal traditions, beyond the Anglo-American adversarial model. The oft-noted cleavage between common law and civil law systems (Apple and Delying 1995), for one, is a rich vein for contrastive insights, even if, as comparativists note, their differences are overdrawn (Spamann 2024; Spamann et al. 2021). Would, for instance, our claims about court records’ power to define the past, sway the present, and set down durable structures that restrict our futures hold differently in civil law countries, where judicial precedent weighs much more lightly on court actors? What forms of “dialog,” distinct from those we highlight, might court records enable where trials are comparatively bureaucratic affairs and appellate review might involve de novo consideration of the facts? Similarly, how might an inquisitorial system – where judges are factfinders, jurors are often absent, and without the adversarial dramaturgy of common law trials – alter how court records impose a hierarchy among legal, scientific, and lay truths? These are just some of the prompts we hope other scholars will take up. Beyond questions of formal design, we are eager to learn which of our theoretical claims as to court records’ properties would hold differently – or not at all – in authoritarian settings with less judicial independence, countries where religious law predominates, or postrevolutionary or new constitutional regimes.

Whatever our hopes, our aims here have been modest. This article is not a full-fledged primer on how best to use court records, nor a final word on their theoretical traits. It is instead a diagnostic effort meant to spell out the promise and challenges of using these data. Later work could – and should – take a more prescriptive approach. That said, we hope the insights offered here will prove fruitful to those engaging with court documents in research, not only for those in the sociolegal camp but, ultimately, in all precincts of social-scientific and humanistic inquiry.