THE PENAL PENDULUM

Everyone from academics to journalists and politicians seems to know about the penal pendulum in the United States (e.g., Stuntz Reference Stuntz2011; Bernard and Kurlychek Reference Bernard and Kurlychek2010).Footnote ¹ According to the conventional wisdom, punishment in the United States shifts every few decades between lenient/punitive or rehabilitative/non-rehabilitative orientations in what seems to be an endless cycle—like a pendulum. The pendulum metaphor has been a mainstay of textbooks and lectures on penal history, and a general backdrop to discussions about penal change. Perhaps reflecting the metaphor’s built-in assumption that the swing from one extreme to the other is natural, automatic, even predictable, no one has offered a compelling explanation for what causes these recurrent shifts throughout time.Footnote ² However, each of these moments of profound change has been studied intensely; over the past several decades, for example, scholars have constructed macro-level accounts to explain the most recent pendulum swing. According to these accounts, (inter)national political, cultural, and economic changes shifted public and elite opinion about punishment, leading to “the punitive turn”—the decline of the rehabilitative ideal and the rise of mass incarceration, among other penal changes that began in the United States in the 1970s and extended to other Western countries (e.g., Beckett Reference Beckett1997; Garland Reference Garland2001; Simon Reference Simon2007).

In the course of this new work, however, scholars began questioning the assumptions that went into pendular-style accounts that emphasize sudden, consuming ruptures with the penal past. In the years around 2000, scholars increasingly argued that both “rehabilitative eras” and “punitive eras” are misleading designations; focusing mostly on the recent punitive turn, they illustrated that this period (and others like it) was internally more variegated than an all-encompassing support for punitive/non-rehabilitative—or lenient/rehabilitative—punishments (e.g., Brown Reference Brown2002; Maurutto and Hannah-Moffat Reference Maurutto and Kelly2006; Hutchinson Reference Hutchinson2006). Scholars also showed that while official rhetoric might change, on-the-ground practices frequently did not reflect these changes (e.g., Lynch Reference Lynch1998; Cheliotis Reference Cheliotis2006; Phelps Reference Phelps2011). In the years around 2010, scholars found that while a selective composite or “national-level” account gives the impression of punishment’s perennially shifting goals, the overall trend breaks down upon closer inspection, especially when examining individual states or locales (e.g., Campbell Reference Campbell2007; Miller Reference Miller2008; Barker Reference Barker2009; Perkinson Reference Perkinson2008; Lynch Reference Lynch2010; Schoenfeld Reference Schoenfeld2010). With these lessons in place, macro-level theories of punishment that attributed major shifts in punishment to major shifts in society no longer made sense. New theories were needed, and a new generation of scholars, including myself, has been working to that end (e.g., Campbell and Schoenfeld Reference Campbell and Schoenfeld2013; Rubin Reference Rubin2015, Reference Rubin2019).

In Breaking the Pendulum: The Long Struggle Over Criminal Justice, Philip Goodman, Joshua Page, and Michelle Phelps use these challenges to our understanding of penal change—the internal variegation of even staunchly punitive eras, the gaps between rhetoric and practice, and the local-level variation—as evidence supporting their call to leave the pendulum metaphor behind once and for all and as a provocation to offer a new perspective for understanding penal change.

AN AGONISTIC ACCOUNT OF PENAL CHANGE

Goodman, Page, and Phelps’ new perspective centers around the penal field—Pierre Bourdieu’s field concept as extended to the punishment realm by Page (Reference Page2011). The penal field consists of all of the actors (and groups) who seek to shape punishment policy and practice, including both “challengers” (those trying but failing to change the status quo) and “incumbents” (the reigning policymakers). The field is a spatially and temporally dynamic concept: which actors appear in the field, and their positions of relative influence, can change.

Against this backdrop, Goodman, Page, and Phelps argue for an “agonistic perspective”; they see struggle, or conflict across different actors and groups with different levels of power and influence, as the primary mechanism of penal change (Reference Goodman, Page and Phelps2017, 8). Traditional pendular accounts frequently discussed conflict, but only as it related to pendular shifts: inspired by a change of heart, a new group emerges to reform punishment and fights with those groups who prefer the traditional way. For example, the rise of prisons following the American Revolution is often attributed to white, male, middle-class reformers who, inspired by general Enlightenment-era humanitarian sentiment or various forms of socioeconomic insecurity, fought against supporters of corporal and capital punishment. By contrast, in Goodman, Page, and Phelps’s agonistic perspective, conflict is perpetual, if frequently invisible, only occasionally leading to large-scale change.Footnote ³

Goodman, Page, and Phelps argue that, since conflict is ongoing, change ultimately occurs when groups gain the cultural or economic resources to influence policy. For example, in general, correctional officers have had little direct impact on official policy since the prison’s origins;Footnote ⁴ but where they became organized, well-funded, and strategically aligned with other groups (such as in 1980s California), they were well positioned to influence state-level penal policy (Page Reference Page2011; Goodman, Page, and Phelps Reference Goodman, Page and Phelps2017, 126). It is not that correctional officers changed their views about punishment, but rather their position in the field and thus their ability to change policy had improved. Penal change is thus less a product of shifting views than of different players’ shifting status in the penal field. And what appears to be a shift in the pendulum is actually the visible product of frequently invisible but ongoing contestation among shifting coalitions of actors.

While they focus on groups’ changing positions within the penal field, Goodman, Page, and Phelps also recognize that the penal field is not autonomous from other forces in society. In a compelling metaphor, they argue that macro-level changes —“trends in the economy, politics, social sentiments, intergroup relations, demographics, and crime”—act like tectonic plates shifting under the field—shifting the players’ positions on the field to have more or less influence (Reference Goodman, Page and Phelps2017, 13). While these macro-level social changes do not automatically cause penal change, they present opportunities that shape the contours of competing groups’ struggles. But like tectonic plates, these shifts can occur suddenly, causing earthquakes, or gradually, causing a slow buildup of tension. Importantly, it is this ongoing conflict that, they argue, lays the basis for future change, essentially creating the fault lines that can be activated by the next seismic shift in society, creating the opportunity for groups who were previously challengers to become the new incumbents.

IN DEFENSE OF THE PENDULUM

Although Breaking the Pendulum’s central contribution is a useful and flexible agonistic model of penal change that marries the micro and macro levels, the book’s biggest takeaway is that we should once and for all abandon the pendulum idea that has long been used to characterize the macro level. In introducing this Review Symposium on Breaking the Pendulum, however, I wish to muddy the waters by defending the pendulum approach. I begin by highlighting two apparently irreconcilable characteristics of penal history.

Goodman, Page, and Phelps find that apparently national trends disintegrate if we break these trends down by state. Some states have their own oscillation, but they may be misaligned from the national trend, occurring earlier or later or for shorter or longer periods. Some states go all in, investing in a way that is consistent with the overall trend, while others skim the surface. It gets even messier at the county or institutional level. In addition to this variation across time and place, local trends are internally contradictory, with examples of policies mixing punitive and lenient motivations or rehabilitative and custodial goals, while policy and rhetoric are simply inaccurate representations of practice. Because national trends disappear upon closer inspection, they argue, we cannot take the pendulum account seriously.

Despite this empirically accurate statement, there is an apparent oscillation in penal history at the national level. The national level—the long-time favorite of the most influential accounts of penal change (e.g., Beckett Reference Beckett1997; Garland Reference Garland2001; Simon Reference Simon2007)—is the level at which pendular accounts are told. At this level, one can identify fairly distinct periods in US penal history; each period begins with a comparatively lenient or more rehabilitative approach to punishment that is followed within a few decades by a reactionary move away from those initial goals.Footnote ⁵ Of course, focusing on all states (including outliers) or examining every instance of innovation (including failed innovations) would not yield the familiar pendular pattern. To see this pattern, one must follow what has been the dominant strategy of focusing on bellwether states or the birthplaces of those innovations that catch on. If we continue to tell penal histories from this bird’s eye view—and if we acknowledge time lags, variations in the extent to which states sign on to penal trends, gaps between theory and practice, and staunch outliers—the pattern remains. It would not be the original, simplistic version that implies sudden, homogenous change following various moments of rupture, but it would still yield a pendular pattern.

Why does penal history, told at the national level, but not necessarily the local level, seem to follow this pattern? What are we to do with the fact that, despite the subdermal contestation and the variation across locales, there is an apparent oscillation in penal history? Ironically, we can get at least a partial answer using Goodman, Page, and Phelps’s agonistic perspective.

As Goodman, Page, and Phelps argue, actors matter and we need to bring them into our accounts. In fact, many people—voters, journalists, politicians, and penal actors—believe in the pendulum. Whatever its empirical inaccuracy, the pendulum idea still shapes many people’s social realities. A simple Google search for references to “pendulum swing prison rehabilitation,” which yields more than 800,000 results, illustrates the deeply embedded role of the penal pendulum in the public, administrative, and scholarly imagination. We need to examine how belief in this pendulum itself becomes a force for change, such as when politicians or penal administrators say that it is time for the pendulum to swing in the other direction. Whether they believe their own rhetoric or not, their words have power for their audience and help naturalize, and thereby legitimize, the idea that it is time for change—this is, they imply, just the way it goes. Such arguments, however, require some amount of time to pass between reforms, suggesting one element of the pendulum (length of the period) that is powerful.

Additionally, apparently national-level trends, with their pendular appearance, do impact what happens at the state and local level, as is evident in Goodman, Page, and Phelps’ account (see also Campbell and Schoenfeld Reference Campbell and Schoenfeld2013). While their main point is to illustrate the problems with national trends as clean depictions of consensus—widespread debate underlying national trends, the significant exceptions to national trends, and these trends’ frequently limited impacts on practice, even when adopted at the local level—the power of national trends still shows through.

For example, drawing on my research (Rubin Reference Rubin2013; Reference Rubin2015), Goodman, Page, and Phelps describe and problematize the role of a national consensus over how prisons should be designed and managed in the nineteenth century. On the one hand, almost every US prison formally followed the same model: the Auburn System (prisoners silently congregated to labor in large factory-like rooms during the day, retreating at night to solitary cells). On the other hand, this consensus falls apart upon closer inspection. For example, if we look at how these prisons actually functioned, we would find substantial variation in how they interpreted the Auburn System or the many gaps that emerged because of unenforced rules or purposeful deviations from the model (see McLennan Reference McLennan2008). The consensus manifested by states’ formal models also fits uncomfortably with penal reformers’ raging debate over how prisons should be designed—one coalition of reformers advocated the Auburn System, but another group advocated the Pennsylvania System (a model of long-term solitary confinement in which prisoners worked alone within their cells). These closer examinations complicate the standard pendular narrative of this period as one of profound consensus and homogeneity (Goodman, Page, and Phelps, Chapter 2).

Despite these important elements of heterogeneity, the growing consensus manifested by the Auburn System’s dominance on paper was persuasive to states still deciding how to plan their prisons. State actors visited the more numerous Auburn-style prisons as models for their own new prisons and were encouraged by the normative power of Auburn’s popularity (Rubin Reference Rubin2015). Illustrating this normative dimension, a state prison in Philadelphia, which remained the lone holdout to this trend by following the Pennsylvania System of long-term solitary confinement, was criticized mercilessly for deviating from the growing norms about prison design (Rubin Reference Rubin2013; Reference Rubin2020). Thus, even where there are gaps in practice or extensive debates, national trends that mimic cyclic change become phenomena we need to explain as well as phenomena that help us to explain other developments (Rubin Reference Rubin2019).

To take another characteristic of pendular thinking, penal trends are thought to have a certain internal momentum to them. Like a real pendulum, punishment’s swings back and forth are seen as mechanistic—natural, automatic, and self-propelled on their own reinforcing momentum. Goodman, Page, and Phelps spend a fair bit of time arguing against this idea specifically to emphasize the importance of actors (Reference Goodman, Page and Phelps2017, 7–13). Penal change does not just happen, they explain; instead, people make the change happen—they fight (hard) to make it happen. This point is an important corrective to many macro-level approaches that sometimes avoid tracing the process by which macro-level variables play out at the level that involves actual actors (a flaw in my own work sometimes, e.g., Rubin Reference Rubin2016). But by introducing the importance of agency and specific actors, Goodman, Page, and Phelps may have overstated the case against the mechanical nature of change.

I would argue that, consistent with a pendular approach that describes penal change mechanistically, penal change frequently does create its own momentum, a kind of path dependence, which itself becomes an important force or variable. Throughout penal history, there are numerous examples of penal change having significant consequences for future punishment (Schoenfeld Reference Schoenfeld2018). This is particularly true for experiments in penal reform that fail spectacularly—it is news of their failure, not an exogenous change in public opinion, that causes a reaction, prompting new reforms and experiments or reviving older experiments. In my own research, I find that fatal experiments with solitary confinement were an important reason why state legislators and prison administrators avoided the Pennsylvania System and its long-term solitary confinement, despite substantial differences between the Pennsylvania System and earlier, more extreme experiments with solitary (Rubin Reference Rubin2020). In this and other examples (e.g., Rubin and Reiter Reference Rubin and Reiter2018), penal reforms created certain paths and placed others off limits. Penal actors still exercised agency by generating strategic narratives about past policy failures, and by advocating or defending particular practices. But they did so in a larger context in which past penal reforms cast a long shadow, shaping the contours of what was conceivable—and defensible (Rubin Reference Rubin2019). In this sense, these reforms have downstream consequences that push developments toward reaction, always building on, responding to, or reversing what came before. While recognizing that actors can creatively use these raw materials, it is important that we do not overlook the mechanistic elements of penal change.

****

This Symposium issue contains three distinct perspectives on Breaking the Pendulum and a forward-looking response from the book’s authors. First, Johann Koehler, drawing on political and historiographical theory, confronts the notion that both penal change and contestation are axiomatic features of penal history. He argues that by focusing on visible disputes between agonists, we overlook those “antagonists” whose views are so marginal that they do not make it to the policy table. Exclusion of these antagonists, Koehler argues, represents an important source of consensus even while agonists continue to struggle over competing visions of criminal justice.

Next, Geoff Ward argues that the pendulum narrative that Goodman, Page, and Phelps dispute was never an accurate depiction of racialized criminal justice. Using juvenile justice as an example, he shows how pendular narratives of juvenile justice erase blackness and systematic racism; in addition to offering a colorblind account of penal history, those accounts overlook a large and important counterexample to the notion of pendular shifts in penal policy. Rather than oscillating between retribution and rehabilitation, Ward argues, penal policy aimed at black youth has always been shaped by retribution.

Fergus McNeill then asks to what extent the agonistic perspective applies beyond the US context and its emphasis on incarceration. He considers early twentieth-century probation policies in Scotland—a policy domain known in the literature (and McNeill’s own earlier work) as a location of stability and consensus, characterized by ongoing progress rather than pendular shifts. Drawing on oral histories, McNeill finds disputes at the level of policy debates and on-the-ground practice, below the surface of apparent consensus, as Goodman, Page, and Phelps have argued we should expect to find. His essay suggests the agonistic perspective travels well, even to an unlikely context in Europe.

Finally, Philip Goodman, Joshua Page, and Michelle Phelps respond, refining their theory in light of these comments. In particular, they offer the concept of conflictual consensus to embrace Koehler’s point that, while conflict may characterize the penal field, some measure of consensus is provided when even the staunchest opponents will agree that some positions are off the table. Understanding how decisions are made about what constitutes legitimate and illegitimate punishment, and which groups are deemed legitimate enough to weigh in on this decision, Goodman, Page, and Phelps argue, will be an important area for future research. They also call for research like McNeill’s, that extends their framework into other countries and that considers other sources of penal stability, such as the role of racial inequality that Ward has pointed out. Finally, reevaluating penal reform’s apparent cyclic nature that I discussed, Goodman, Page, and Phelps introduce the concept of “penal zombies,” referring to the way in which policies and practices can be revived over time by competing groups, ultimately outliving their creators (and their creators’ intentions).

As these several commentaries illustrate, Breaking the Pendulum is a generative work tackling a number of issues, including one of the most central questions for punishment and legal scholars: how do we describe the past and what does it tell us about the present and future?