 Hello. I'm Carol Christ. I'm the Chancellor. Good afternoon. Thank you for this opportunity to join and welcome all of you to the Berkeley campus this afternoon for this wonderful and meaningful event that tanner lectures on human values. We're honored and humbled to be one of only nine universities from around the world that were selected to host this Distinguished Lecture Series every year. If we were to be judged by the company we keep, we could do far worse than to be joined by Cambridge, Oxford, Harvard, Michigan, Princeton, Stanford, Yale, and Utah. This series was founded in 1978 by the American scholar, industrialist, and philanthropist, Obert Clark Tanner, who was also a member of the Faculty of Philosophy at the University of Utah and an honorary fellow of the British Academy. It's somehow both sobering and comforting to realize that today, nearly half a century later, the impetus behind the establishment of this lecture series could not be more germane, more relevant, or more important. Tanner's goal in establishing the lectures through the Tanner philanthropies was to promote the search for a better understanding of human behavior and human values. An objective that could not be more simple to describe or more challenging and complex to pursue. He hoped that the lectures would advance scholarly and scientific learning in the area of human values and contribute to the intellectual and moral life of humankind. Suffice it to say that one need only conduct a cursory review of the headlines in recent years to know that Tanner's objectives and concerns resonate today as they have in the past. I also deeply appreciate that Tanner and our university share a common interest in and dedication to using education, knowledge, and understanding to support and advance the greater good. So too do we share his capacious understanding that human values should be defined as broadly as possible. Making room for diversity of perspectives and cultural and historical traditions in this exploration of ourselves and our species. And so the Tanner lecturers may be chosen from any discipline and the lectureships can and do transcend national religious and ideological divides and distinctions. The Tanner lecturers are chosen not on the basis of the particular views that they hold, but for their uncommon achievement and outstanding abilities in the field of human values. And so the very ethos underlying the process of selection conforms beautifully with the underlying values that launch this lecture series. The lectures from all nine universities are published in an annual volume. In addition, Oxford University Press publishes a series of books based on the Berkeley Tanner Lectures. The 13th and 14th volume of the series were published in 2021. Here at Berkeley, the Tanner lecturer is appointed through a faculty committee of which the Chancellor is the Chair. This is the least honorous duty I have. I congratulate my colleagues, Professors Jay Wallace, Hannah Ginsburg, Christopher Goods, Kinch-Hostra, Niko Kaladney, Kevvis Goodman, Stefan Ludwig Hoffman, and Rebecca McClennan for their brilliant choice of this year's lecturer, Charles Bytes. Now let me call on my distinguished colleague, Professor Niko Kaladney, to introduce Charles Bytes and today's commentators. Professor Kaladney will also moderate the discussion that follows. So it's a privilege to introduce Chuck Bytes, who is Edwards S. Sanford Professor of Politics and Director of the Program and Political Philosophy at Princeton University. After his undergraduate degree in history from Colgate University, Bytes received a master's degree in philosophy from the University of Michigan, and then a doctorate in politics and the program in political philosophy at Princeton. Before returning to Princeton in 2001, he taught at Swarthmore and Bowden. Bytes was editor of philosophy and public affairs for over a decade, and now serves as an advisory editor. Since my own drafts have seen the business end of his red pen, I can testify personally to his gifts as an editor, and I would hazard that everyone in the field has benefited from his eagle eye and sound judgment, if not as a writer, then as a reader. Bytes has held fellowships from the Rockefeller, MacArthur and Guggenheim Foundations, the American Council of Learned Societies, and the American Council on Education, and he's a member of the American Academy of Arts and Sciences. Bytes's first book, Political Theory and International Relations from 1979, spawned a now massive literature on global justice by asking how theories of justice, such as that of John Rawls, which were designed for single nation states taken in isolation, might apply across borders. His most recent monograph, The Idea of Human Rights from 2009, suggests that political philosophers should conceive of the subject of human rights in terms of a practice with a distinctive origin and history. The question is whether and for what reasons this practice is to be supported and endorsed. The monograph in between, Political Equality, from 1989, distinguished several distinct normative criteria packed into slogans like one person, one vote, and then applied those criteria to real or alleged democratic ills, such as campaign finance and gerrymandering. His lectures this week, in a way, returned to the arguments of political equality from the vantage of an even less happy American democracy, albeit with the half comforting coda that things could still get worse. Philosophers in general and political philosophers in particular have long been criticized for having their heads in the clouds and their feet not firmly planted on the ground. That is to say that they construct ideal theories of how things should be with insufficient attention to how things actually are. There is even a somewhat comical tendency for political philosophers to roundly criticize the field for being insufficiently realistic and to insist that it really must start paying more attention to the facts without themselves actually cracking the binding of the empirical research of the social scientists on their own campuses. Byte stands alone in actually doing what political philosophers keep telling each other that they ought to be doing. If there are other political philosophers who engage with the work of social scientists and the self-understanding of actual political practitioners, there's no other political philosopher who does it better, more knowledgeably, more skillfully, and more fixedly on what in the end really matters. The payoff is in part that Bytes helps us to come to better terms with our own typically inquit reactions to the facts. If we're disquieted, say, by Professor Gillins' findings that when the affluent and the non-affluent disagree on policy, outcomes are sensitive only to the preferences of the affluent, what exactly is the source of the disquiet? And once we trace our disquiet to its source, should we be on reflection disquieted? But there's also a payoff in the other direction. Bytes' engagement with what people are saying and thinking closer to the ground sharpens the normative orientation that informs political philosophy of even the aerial castle construction type. When political philosophers construct theories of human rights, for example, what exactly are they doing and what should their aims be? Chuck, I'll now call him Chuck, Chuck has endeared himself to students and colleagues over the years with his easy sense of humor, modesty, and utter lack of pretension, which are rather out of keeping with your aforementioned academic achievement and intellectual stature. On his sense of humor, realizing that I had become rather a high maintenance commentator at Tanner Lectures at Princeton that Chuck was organizing, I said by way of backpedaling joke, well, there'd better be Evian in my dressing room. When I checked into the hotel, Chuck had actually seen to it that there was a bottle of Evian waiting for me in my room. On his modesty and lack of pretension, when first approached with the invitation to give the Tanner Lectures here at Berkeley, Chuck hesitated from genuine concern that he would not be able to come up with a series of lectures that would be up to the occasion. Having read the lectures, I can say with confidence that this modesty is as misplaced as it is disarming. But I'll leave the spoilers at that. Please join me in welcoming Chuck Bites to the podium. All right, I think you could probably all hear me, is that right? Yes, okay. So let me start, Nico, just by saying thank you for it. I was gonna say a two-generous introduction, but you now have me self-conscious about saying things like that. So let me just say a generous introduction indeed. Thank you to the Chancellor and to the Tanner Committee for your invitation. I'm honored by it. And thanks to what I can only describe as a dream team of commentators, I'm honored by the three of you as well. A decade ago, it was hard to find a political scientist hard to find a political scientist who feared for the resilience of democracy in America. Nancy Rosenblum, a political theorist and a friend of mine observed in 2014, accurately at the time, that although there was an, I'll quote her, an acute sense of democratic failing, few observers worried that it would lead to fascism or civil unrest. Four years later, two of her Harvard colleagues published a book called How Democracies Die, in which the American case was central, and it became a bestseller. I share the anxiety about our institution's resilience far more acutely, I think, now than I did in early 2019 when I chose the subject of these lectures. But that anxiety is not my subject today. I'm interested in the intimations of democratic failure that Rosenblum noted among professional observers of American politics well before Trump. So these lectures are an exercise in by a political theorist reading political science as democratic theory. And I'll just say as an aside, a colleague reading these said to me, what could possibly go wrong? So I'll ask two questions. The first is, what are the arguments that our representative institutions are failing? We'll review some evidence from political science, but this isn't mainly a question about facts. It's about their normative significance, what I'll call, what I'll say is diagnostic. Second, what might we learn about successful democratic representation from a critical look at the diagnoses of failure? Taking the phrase for the moment as a placeholder, what do the diagnoses suggest about the requirements of fair and effective representation? This question is prescriptive. We'll consider the diagnostic question today and the prescriptive question tomorrow. The intimations of failure that I'll describe are not products of the Trump years. If what political scientists say is right, representative democracy in the United States is enduring challenges. The urgent practical question is whether these challenges are amenable to any feasible repair. That question I'm afraid is beyond us or at least it's beyond my own professional competence as a political theorist. What we can do is to ask how we should judge whether representative institutions are failing and how we might recognize improvement. The ideas of good representation that we find in contemporary political theory often seem to me too abstract to inform practical criticism of existing institutions. I'm gambling that a critical reading of political science might be a stimulus to take seriously elements of democratic representation in the United States that can escape the attention of political theorists who are thinking more abstractly. So one final preliminary note. In the last few decades, as some of you will know, political theorists interested in political representation have elaborated conceptions of representation whose scope extends well beyond conventional understandings of the subject in political science. Since my aim here is to engage with political science, I'm gonna leave those conceptions aside. Omitting some analytical refinements, I'll understand democratic representation more conventionally as a process in which individuals are selected typically by popular election to occupy offices defined by duties to participate in making law and overseeing its execution. There are important borderline cases. So, you know, for example, administrative officers who have delegated rulemaking responsibilities and possibly elected judges. But the main idea is that political representation is a process for making authoritative public policy by individuals who were chosen by election to do so. Now, I know for some purposes that conception will be too narrow. I hope not for mine. The catalog of putative failures of representation in the literature of American politics is really quite long and I can't possibly rehearse all of them. I wanna begin with five heavily stylized facts drawn from that literature. It's an incomplete selection. I won't try to defend it other than to say that people who report these facts usually say that they bear on the quality of democratic representation in the US, their signs of failure. To start some three caveats. First, these facts seem to me to be fairly widely accepted in political science, but there's certainly no consensus. Our time's limited and so I'll have to leave most of the qualifications aside or as it turns out to footnotes that none of you have been able to see. Second, things might be changing. You know, we may be at an inflection point, in particular with trends in public opinion shifting with changes in the Republican electorate. Finally, a third caveat. There's a hard question about whether what seem to be failures of political representation are actually institutional dysfunctions or refractions of endogenous changes. And those can be hard to distinguish. Institutions that are unproblematic in some circumstances can become problematic in others. This is another question that I'm afraid is beyond me as a political theorist, though it isn't beyond any of my three commentators and I hope that some of them will take it up. So let me list the five facts and just gesture very briefly at some reasons why people diagnose them as signs of normative failure. Later, we can examine the diagnoses more carefully. First fact, the first fact is about gridlock and it concerns the productivity of the Congress. By some measures, it's been declining literally for decades. Let's say, although only tentatively, that gridlock is a failure to make or change policy on important issues. I'm gonna come back and ask whether this is adequate. Now, the poster child of gridlock that all of us are familiar with is fiscal brinkmanship that leads to government shutdown. But the more serious concern is that Congress isn't responding to problems that seem to need a response. Sarah Binder, who's one of the great scholars of this subject tells us that gridlock has been increasing since the 1970s. She writes, stalemate at times now reaches across three quarters of the salient issues on Washington's agenda. Well, gridlock looks bad for representation. It reinforces the status quo bias that's built into our constitutional system of checks and balances. It interferes with the updating of policy as circumstances change. It obstructs oversight of administrative agencies and complicates their work applying policy to changing circumstances. In an environment of increasing income and wealth inequality, it blocks changes in policy that are aimed at reducing it. Gridlock is not distributionally neutral. At the structural level, gridlock shifts influence to the executive branch and to the courts and reduces the effectiveness of congressional oversight. Those are all consequences of gridlock that might be seen as failures of political representation. That's the first fact. Second fact is really a set of facts traveling under a common, though possibly misleading heading of polarization. So I think there are at least three dimensions to this. First, ideological divergence among members of Congress and more generally among party elites and activists has increased dramatically over the last four or five decades. There's no longer any overlap of ideological positions between Republicans and Democrats in the Congress. Dramatic change from 50 years ago. Elite polarization has been asymmetric with Republicans moving farther to the right than Democrats have moved to the left. According to what seems to be the majority view among political scientists, although hardly the consensus view, increasing elite polarization has not been matched by ideological polarization in the mass electorate. And to the extent that the electorate has polarized, that has followed rather than preceded elites. Now, those who take this view, and as I say, that's not everyone, they believe it's a failure of representation. As Morris Fiarina influentially put it more than a decade ago, there is a disconnect between an unrepresentative political class and the citizenry it purports to represent. Now, some who disagree about this first dimension of polarization mostly agree about a second. And that's this. At both the elite and the mass levels, people have sorted themselves into ideologically more consistent parties. There are fewer moderate Republicans and fewer conservative Democrats. Now, I think the deep source of this development is to be found in the Southern realignment, a racially driven response to the Civil Rights Revolution, but it has a prehistory. Now, if you think about it, partisan sorting isn't itself a failure of representation. It's just a phenomenon, but its consequences might contribute to failures. Sorting can encourage gridlock by making legislation more difficult. It can create incentives for more extreme elite polarization as politicians compete for position within their own already sorted parties. When I wrote that sentence, I didn't imagine I'd be reading it in the middle of October, three weeks before an election, which is just bearing out its truth. Sorting might also be a source of a third dimension of polarization, sometimes called affective polarization or partisan animosity. Two political scientists reported last year, and I'll quote them, that Republicans and Democrats loathe each other more now than ever previously measured in surveys, more than ever. Negative partisan polarization has been accompanied by a decrease in cross-cutting social identities in a process that the political scientist, Liliana Mason, calls social sorting. The sources of affective polarization are obscure, but I'd note that two of its closest students argue that the single most important factor has been the growing racial divide between supporters of the two parties. Now, like sorting, affective polarization isn't necessarily a failure of representative institution, but like sorting, its results can be damaging. Some think, and there's some evidence for this, that affective polarization can cause voters to change their previous policy or ideological positions and destabilize the system's capacity to hold elected officials to account or that it can incentivize elite resistance to compromise contributing to gridlock. And I think maybe the most harmful result may be the motivated erosion of norms of forbearance and loyal opposition that are critical to the functioning of representative institutions. And again, I didn't imagine we'd be seeing so much evidence of it in the daily news when I wrote that sentence. Our third and fourth facts involve disparities and policy responses, responsiveness by wealth or class, that's the third fact, and by race, that's the fourth. So possibly the most remarked upon fact, the new stylized fact of American politics, according to some writers, is the tendency of policy to track the preferences of the wealthy, even when these diverged from those of the middle class or the median voter. Marty Gillins's pathbreaking book documents that fact. His findings aren't entirely controversial, but I think they represent a preponderant view. In a slightly later paper, Gillins and his co-author Benjamin Page conclude starkly, and I'll quote them, that in the United States, the majority does not rule, at least not in the causal sense of actually determining policy outcomes. And again, that certainly sounds like a democratic failure. And like gridlock, it continues to a further failure, the failure to stop the growth of inequality of wealth since the Reagan years. Conventional models of democratic politics, of course, would predict just the opposite. A more recent stream of research on the same subject might qualify these findings a bit, though I think not in their foundations. A study by Jeffrey Lacks and his colleagues suggests that affluent influence may reflect partisan differences. They find that democratic senators are more responsive to home state opinion in aggregate than Republicans, while Republicans are more responsive to their home state wealthiest partisans. And the study concludes, and I'll quote them, that it's Republican senators, not Democrats, who are primarily responsible for the overall pattern of affluent influence. But still the diagnosis of failure is similar. They write, partisan distortion and representation violates norms of equal voice. Well, the study of policy responsiveness to racial and ethnic groups, which is the subject of our fourth fact, is less advanced, although it's developing. The most, in the most ambitious recent study of race and responsiveness that I'm aware of, Zoltan Heinal, the political scientist, gives evidence that race, even more than wealth, is the most important demographic feature that in his words determines who wins and who loses in American democracy. And he says that this is true in three respects, of Black's probability of winning in races for office, of voting for candidates who succeed, and of getting policy outcomes that they favor. The measure of policy outcomes that he uses is limited to policies involving government spending, so it excludes a good deal. But still, Heinal can identify no other group that loses, which is to say whose preferences fail to correspond to outcomes more often than Black's. And he argues that the influence of race can't be explained by any of the other factors that we might have thought would be relevant, like class, education, political orientation, and partisanship. And here again, you'll notice a theme emerging, there's an asymmetry. During periods when Democrats control the Congress and the presidency, Heinal's data show that the correspondence gap virtually disappears. And he takes that to be a sign of fair representation under Democrats. He takes the racial gap and correspondence at other times to be a failure. So let me turn to my last, the fifth, and I think maybe the most complex intimation of failure. It's partisan bias and districting of the House of Representatives and many state legislature. We have a kind of different kind of partisan bias in districting for the Senate, but I'm not gonna address that here. Partisan bias is a function of the way that single-member territorial districts have been drawn within a jurisdiction. And as the first approximation, though just an approximation, it's a condition in which a party that wins fewer votes than its rival can win more seats. Gerrymandering is one source of partisan bias understood that way. But it's emerged in the last decade, though it's hardly a new discovery, that partisan gerrymandering is not the only source of bias in districting. In many parts of the US, Democrats tend to be more geographically concentrated than Republicans. So even without deliberate manipulation, the districting system, taken nationally, would display partisan bias in favor of Republicans. And it's getting worse. Now it seems clear that people who manipulate district lines for partisan advantage fail in their duties of public officials. There's more to be said about what the failure is, but I'll leave that aside. Because most political scientists think that the more basic failure is in the result of partisan gerrymandering. It distorts the representation of parties in legislatures. Well, if gerrymandering is objectionable for that reason because it distorts the representation of parties, then presumably bias produced by political geography, what's sometimes called unintentional or accidental gerrymandering would be objectionable for the same reason, an interesting and troubling thought. And that's why I describe the source of failure as partisan bias rather than intentional manipulation. The purported failure is the asymmetry in the terms of party competition. And that may not be the worst of it. Although it isn't necessarily so, gerrymandering also tends to increase the number of safe seats, making the system less responsive to changes in the partisan division of the vote. It can reduce electoral competitiveness. And if that's a failure, it's a distinct failure from partisan bias and it can, because it can happen without partisan bias. So let me take stock. I've listed five purported failures or sources of failure in representation in the U.S., increasing gridlock in Congress, three dimensions of polarization, ideological polarization among elites, the sorting of voters into more ideologically consistent parties and growing partisan animosity. Third and fourth, a greater responsiveness of policy to the preferences of wealthy and non-black voters than to the non-wealthy and blacks. And fifth, partisan bias in districting for Congress and state legislatures. Now, this is by no means a comprehensive inventory of the intimations of failure that one could find in recent political science. I haven't mentioned, for example, the transformation of the communications environment of the last 20 plus years, which among other things has stimulated what Nadia Urbanadi describes as a revolt against intermediary bodies whose consequences for political representation, she thinks, are still undetermined. And I haven't referred to the important subject of representation in the states where some of the pathologies I've discussed at the national level are reproduced even more alarmingly. The facts I have mentioned are in some ways disputed, but I think it's enough that many political scientists accept them. Our question is how these facts illustrate shortfalls from expectations of fair and effective representation that we might reasonably endorse. I mean, I don't doubt that we face failures, but we need to be clear about the standards of fair representation that justify the diagnoses because those standards will inform our view of what fair and effective representation would be like. So the problem is that the facts don't map neatly onto distinct diagnoses. So rather than examine the facts one by one, I'll organize the diagnoses under three headings. Failures of effectiveness, failures of correspondence and responsiveness and failures of fairness. And I'll talk about each of those things separately. So let me begin with failures of effectiveness. Congressional gridlock is its most prominent manifestation. Is that something we should worry about? Well, modern scholarship on what's usually called congressional productivity began when it was observed that a long period of unified party control of the executive and legislative branches came to an end in the 1950s. This caused anxiety about representation. In a seminal paper in 1988, James Sunquist, the political scientist noted that the reigning theory of legislation in American politics had to do with party government. It held that parties coordinate the process of legislation with the leadership produced by a president elected by the whole people to advance a program. And according to this theory, that's only possible when there's unified party control of both branches. So the anxiety at that time about divided government was that popular majorities would no longer be able to direct public policy. Sunquist worried that divided government would produce, as he put it, ineffective compromise for which neither party could be held accountable. A little bit later, the political scientist David Mayhew asked whether divided government really did render government less effective and he found counterintuitively that divided government wasn't less productive than unified government. Those findings in fact have been challenged. More recently than Mayhew's book. But that issue was sort of orthogonal to our main interest. What matters here is the claim that in the last several decades, there's been a trend toward more gridlock with moments of congressional productivity mostly occurring during periods of crisis. So after 9-11 in the 2008 financial crisis during COVID. Well, why is propensity to gridlock a failure? Well, as Jenny Mansbridge observes in one of several important papers on the subject, government's capacity to get something done is important. And more important in modern circumstances when we're confronted with challenging problems of collective action than in the world that our constitutional system was designed for. Gridlock obstructs solving collective action problems. It prevents things from getting done that ought to get done and that might be enough to explain why it's a failure. Well, the persuasiveness of that diagnosis depends on how we identify the things that ought to get done. Binder quotes Bob Dole, who was supposed to have said, if you're against something, you'd better hope there's a little gridlock. Dole reminds us that views about what ought to get done are not likely to be politically neutral. People differ about which collective action problems ought to be solved by government action and about how the cost of solving them ought to be distributed. So I think Dole's reminder maybe might suggest that to make a case that gridlock is an institutional failure, we need an idea of what ought to get done that's independent of political controversy. And that's what Binder sought in the study that I mentioned. She constructed a list of issues that constitute what she called the policy agenda and she measured gridlock as a ratio of failures of Congress to act on issues on that agenda to the total number of issues on the agenda. And what she meant by policy agenda, I'll quote her, was, the range of policy ideas plausibly on the radar screens of policymaking elite and active electorates. Well, there's some methodological issues here that I'm gonna leave aside. The real question is whether we ought to regard gridlock judged by that sort of measure as a normative failure. I mean, maybe the idea is that the agenda of issues on the radar screens of the policymaking elite represents something like a consensus view of issues facing the country. And I wonder if that idea is plausible. I mean, it's hard to resist the thought that a proponent of small government conservatism, Dole, for example, might disagree that a lack of legislative productivity is a regrettable failure of the institutions of representation. It could be a sign that the Madisonian elements of the institutional structure are working in the way that they were meant to work. Well, neither of the, I admittedly, somewhat contrived positions that I've distinguished can be the whole truth. I think there's obviously a legislative agenda that we might regard as relatively detached from reasonable partisan controversy. On any sensible view, matters like extending the debt limit to pay for commitments already made and funding routine operations of government that serve common purposes, those things at least belong on a common agenda. And when gridlock obstructs action on matters like that, I think we have reason for regret regardless of our policy commitments. However, many issues are controversial and the failure to act on them can't plausibly be presented as a failure to act on a common public agenda. They're issues like climate change, gun control, anti-poverty policy, and so on. Even updating the minimum wage is a matter of partisan dispute. So I think the politically neutral idea of gridlock stands in contrast to the view that was taken by Sunquist in the paper I mentioned earlier. His concern was that when partisan control of the government is divided, to quote him again, the normal tendency of the US system toward deadlock becomes irresistible and the president's program is frustrated. On that account, the failure of representation lies in gridlock's obstruction of majority rule by undermining the operation of the party system. And I think that gets us a normative diagnosis of at least part of the failure of gridlock and one that would survive Bob Dole's challenge. I'll return later to the idea of responsiveness to the policy preferences of the majority as a kind of failure. So I think that turns out to be a much more general diagnosis of some of the facts I've mentioned. But first I wanna just for a moment press the question whether gridlock is distinctively a failure of representation. I mean, I think that depends on how it's explained. The most common explanation is that gridlock arises from the combination of three factors, each a necessary condition. First, a constitutional system of checks and balances together with the Senate filibuster that create what some political scientists call a gridlock interval in which policy change can't occur even if it's favored by a majority. Second, elite polarization that makes compromise difficult. Even if, and third, a closely divided electorate in which elections are often competitive so that an out party can regard itself as having a reasonable chance to win in the next election. So the thought here is that in the absence of decisive control of both the executive and the legislature by one party, successful legislation requires the parties to compromise but polarization narrows the space for compromise and the prospect of winning the next election decreases the incentive to compromise. Well, elite polarization and close electoral divisions are of course not features of the system of representation. They're aspects of the political environment. So if gridlock is a failure of the system, the reason must involve the effects of the institutional rules on congressional productivity and that means that we can't avoid a judgment about the justification of those rules. Now as we know, there's a well-worn Madisonian story about the wisdom of the countermajoritarian features of the constitutional structure. The story doesn't apply to the filibuster but it isn't really hard to imagine a compatible justification. And the gist of it is that the structure should restrain temporary majorities from precipitous actions, what Madison called sudden or violent passions or deceptions by what he called fractious leaders. I think without laboring the details, the underlying view is that political representation should be structured so as to produce policy that tracks something like the common interest over time. What Madison called the permanent and aggregate interests of the community while at the same time respecting people's rights. As a general matter, I don't think it's far fetched to think that on a reasonable interpretation of the permanent and aggregate interests of the community, constitutional checks and balances might have been justified by those kinds of aims. However, the Madisonian view depends on some background assumptions about how the constitutional provisions are likely to operate under the prevailing political circumstances. I think the propensity for Riddlock as a failure of representation, the case made by some people today, rather than an indication of success, would have to be that under our contemporary circumstances, the Madisonian institutions are too obstructionist. They resist movements of policy that would track the majority's settled view of the common interest. Today that case seems to me reasonable, although I think the details are beyond our scope. For our purposes, the point is that we can't explain why the propensity for Riddlock is a failure of representation without a theory that connects the provisions of the political constitution that enable it with a normative view of the aims of democratic representation. It's not always a matter of falling below some neutral threshold of effectiveness or productivity. We need a judgment about what representation ought to accomplish. So let me turn to the second category of diagnoses that I wanna talk about. These are failures of congruence and responsiveness. So the idea that policy should reflect citizens' preferences, that idea is ubiquitous in the literature about political representation in the United States. And we see this in commentary, I think about most of the putative failures that I mentioned. As I've said, Riddlock might be reckoned a failure of responsiveness if the legislature consistently fails to act on issues on which majorities prefer action. Elite polarization without a corresponding polarization in the mass electorate might be a failure of responsiveness if the legislature produces more extreme policy than what electoral majorities want. Durable tendencies to satisfy the political preferences of the wealthy rather than of the majority or of whites rather than blacks when these conflict might also be seen as failures of responsiveness. So perhaps not of the same kind. These worries all involve claims about distortions in the relationship between the political preferences of some group and either the political positions of representatives or the policy outcomes that they produce. The trouble is that the content of that complaint of unresponsiveness can be equivocal and however it's interpreted, it's not obvious why it should trouble us. So let me say something about that. And I need to begin with a clarification. There's no agreed conception of responsiveness among political scientists or at least of that word responsiveness. What some mean by it is also called congruence or sometimes correspondence. Others mean responsiveness in a different and I would say a quasi-technical sense. So let me say something brief about each. The political scientist Bingham Powell describes congruence as a fit between the preferences of the citizens and the committed policy positions of their representatives. Others speak of congruence between citizens' preferences and policy outcomes. But either way, the idea is that when there's high congruence, citizens mostly get the outcomes they want. Or anyway, the median voter who's pivotal for a majority gets the outcome she wants. Understood that the important point here is that if we understand congruence that way, it's not a causal notion. It might come about for other reasons than that the legislature is actually responding to voters' preferences. I mean, some exogenous feature might explain both preferences and policy. Responsiveness in the quasi-technical sense I have in mind is different. The most natural way to describe it, though it's not always the way it's described, is with a counterfactual. A representative is responsive to a group. A constituency, say, or an income class. If the representative would change her position in response to a change in the policy preferences of that group. So you might say a responsive representative is one whose positions can be explained, at least in part, by the positions of those whom they represent. So the point I wanna make here is that these ideas are related, but they're not the same. High responsiveness is an indication of influence, whereas high congruence is not, or at least not necessarily. And moreover, though this would take a bit of explaining, a responsive system might produce, although a responsive system might produce high congruence, it's a contingent matter whether it will do so, and it doesn't always. So I stress this difference between congruence and responsiveness, mainly because political scientists disagree about which is the better standard of successful democratic representation. And to adjudicate that disagreement, we should ask how, if at all, we might understand either standard as a measure of democratic self-government. So let me begin with congruence. And what I wanna say here is that notwithstanding what might seem like it's prima facie plausibility, it's doubtful that high congruence of preferences and policy in itself is a plausible sign of democratic success. And that's partly because we have familiar reasons to resist giving preferences as such much weight in judging the democratic credentials of policy. Nico Collodny has explained those reasons persuasively in his forthcoming book, embarrassing by saying his terrific forthcoming book. It's also partly because the idea that policy should satisfy preferences can be embarrassed in some circumstances by the fact that representatives help to shape preferences. They're not formed exogenously to the process of representation. So those are important considerations, but for the moment I leave them aside because I think there are two more basic considerations to take account of. First, the concern for congruence is intention with the fundamental and I think straightforward idea of political representation, which is that the whole people delegate to a smaller number of the responsibility of making law and policy. Political representation is a division of labor. Considerations of efficiency and effectiveness and perhaps some other considerations argue in favor of that division of labor. I mean it reduces the cost of decision making and if it's properly institutionalized under favorable background circumstances, it improves the quality of legislation. But neither consideration suggests that the institutional structure should aim for outputs to track the first order policy preferences of the people. The second consideration has to do with the broader values of constitutional government. Institutions of democratic representation are elements of a larger constitutional structure. A democratic constitution seeks to enable the people to govern themselves while protecting against various pathologies that elective representative institutions might be expected to exhibit. So as Christopher Eisgruber, for example, observes, we might hope that constitutional self-government would serve values of stability and effectiveness as well as of popular control. And if something like that view was right, then we ought not necessarily expect representative institutions embodied in a larger constitutional democratic structure to give any independent priority to achieving congruence between preferences and policy. Now we might, of course, have reasons to care about congruence that are not inferences from values of democratic self-government. So we might, for example, believe that policy outputs should advance the common interest and that we might think is a contingent matter and at least in favorable deliberative circumstances, which is a demanding condition. High congruence with constituency preferences is an indicator that policy does that. Or we might think that persistently low congruence for some groups, the poor or ethno-racial minorities, is evidence of systematic disregard for their political interests. Now those are both important considerations, important democratic considerations, but in each case our basic interest in congruence derives from a further concern. It's not about satisfying preferences as such. Now the responsiveness of policy to preferences, I think, is a different matter and this is really the point I wanted to make. And that's because whereas congruence is a feature of political outputs, the outputs of the legislative process, responsiveness is a feature of the inputs. It's a fact about the extent of a constituency's influence over the policy process. It's a matter of realizing popular control of government. Responsiveness of policy to preferences matters for popular control and input value, as I say, in a way that correspondence, which is an output value, does not. And that's true even if we grant that preferences are partly endogenous to the process of representation. That fact isn't obviously relevant to the reasons we have to value our collective capacity to govern ourselves. And maybe the force of the point would be clearer if I put it negatively. Taken to the limit, if policy were not responsive to preferences at all, we could hardly say that the people rule. And here I'd observe I'm just channeling something that V.O. Key wrote decades ago. He said, unless mass views have some place in the shaping of policy, all the talk about democracy is nonsense. So the point here is that a failure of responsiveness as distinct from congruence, if it were persistent, would constitute a failure of representation to be democratic. Now unfortunately to say this much is not to say that we should expect policy to be perfectly or highly responsive to preferences or that all failures of responsiveness are equally to be regretted. Policy responsiveness can vary over time. It can vary across issues. And I think there's just no obvious way to say in general how the value of popular control bears on the optimal extent of responsiveness. That's a problem for those who think that responsiveness is the kind of the sole standard for evaluating when representation is fully democratic. Well, so far we've been considering or I've been considering policy responsiveness as an aggregate feature of democratic representation. I think that follows the central tendency in the empirical literature which tends to treat aggregate responsiveness as a democratic desideratum. However, as disparities in policy responsiveness by class and race suggest, that can be myopic. Those differences raise distinct questions not about aggregate responsiveness but about political fairness. And so I turn now to those failures, my third category of democratic failures. So I think it may seem that all of the putative failures in my inventory might seem to involve some kind of generic unfairness but distinctive issues about forms of unfairness arise for three of the facts in my inventory. Those are partisan fairness in legislative districting and disparities in responsiveness of policy by wealth and race. And the appearance of a kind of distinctive form of unfairness I think derives from what seems to be an inappropriate distribution of political influence over election outcomes in the first case and over policy in the others. And that in turn might suggest that we should understand those failures as violations of a democratic ideal of political equality. Now I've argued elsewhere that that's a complex ideal and it's not easily described in a single formula but for the sake of this discussion let's provisionally adopt a single formula even though we may have second thoughts. So following your colleague Joshua Cohen we might say that one component of political equality in his view there are also others that one component is that citizens should have equal I'm quoting him here equal opportunities for effective political influence. Now that's abstract and following Thomas Scanlon we might lower the level of abstraction a bit by saying that that idea requires that each citizen have equal access to the means for attaining office and more generally influencing policy through the electoral process. So I'm looking here for some vaguely operationalizable notion of what equal opportunities for political influence might involve. Now the question is whether partisan bias or disparities in responsiveness are unfair because they fail to afford each individual equal access to the means for attaining office or influencing policy and to telegraph the point that I hope will emerge in all three cases a diagnosis of unfairness depends on the mechanism that produces it but once we identify a plausible mechanism it's no longer clear that the problem is that individuals have unequal means of influences I think there's more going on. So let me begin with partisan bias in districting. People sometimes say that the votes of those who are disadvantaged in the bias system are diluted or that they receive less weight. Rawl says this for example in his discussion of gerrymandering in a theory of justice. Dilution or receiving less weight sounds like a violation of equality and we need to ask what exactly is the violation? Well whatever else it requires political equality requires that formal opportunities that is to say opportunities defined by law for influence over election outcomes should be equal. So importing here an idea from the technical literature on voting power we might say that voters have equal formal opportunities if they have equal as the theorist put it a priori chances to cast a decisive vote. Now a decisive vote is a vote that would change the outcome of an election. I suspect the only person in this room who was actually cast a decisive vote is Nico who's described it on his website. But most of us have almost no chance of casting a decisive vote. The point is that we want those chances those a priori chances to be equal. So what's an a priori chance? Chances are a priori when they're calculated in ignorance of voters actual political preferences. The precept one person one vote promises equal a priori decisiveness. And the problem is that bias systems can satisfy one person one vote. gerrymandered systems do satisfy one person one vote. And in that respect there's no violation of political equality. That makes the notion of vote dilution due to gerrymandering or partisan bias more puzzling. Now a more substantial impression of inequality emerges when we consider voters as partisans. And here I'll abbreviate a longer discussion and say that it's possible for voters to have equal a priori chances of being of influence. While having unequal chances of influence once we take account of the likely voting behavior of others in the district. I mean after all my chances of casting a successful vote are gonna depend on how everybody else in the district votes. The voting theorists call that and I'm skipping over some technical details here ex ante decisiveness to distinguish it from a priori. The problem is that the vote dilution objection to partisan bias can't be that it produces inequalities in ex ante decisiveness. And the reason for that was obvious once we reflect that partisan bias doesn't affect the voting power of every member of the party it disadvantages in the same way. A voter for the disadvantaged party who's as they say cracked or separated into a competitive district is more likely to be ex ante decisive than a voter for that party who's packed into a district with a large majority of co-partisans. And that's clearly not what people what the familiar objection to vote dilution intends. So the lesson I wanna draw and again jumping over some further discussion is that in spite of the usual framing it's not obvious that we should understand concerned about vote dilution as a matter of the value of votes taken one by one. It's a kind of systematic unfairness endemic to single member districts to SMD systems. And the significance of it emerges only at the level of the jurisdiction as a whole. A voidable partisan bias thwarts electoral competition between parties by setting one party at a disadvantage that can't be justified as necessary to obtain whatever are the benefits of electoral competition in this sort of a district system. I'm gonna leave that for now and say more tomorrow about how this is a kind of unfairness. But let me just say to follow up on something I said earlier partisan bias isn't the only representational harm that's associated with gerrymandering. It can also reduce the competitiveness of elections by creating what we call safe seats making the system as a whole less responsive. And I think an adverse impact on electoral responsiveness would be a distinct democratic harm because it can occur even in the absence of gerrymandering. And again, more about this tomorrow. Now let me turn to disparities by class and race because those disparities may seem like a simpler kind of problem. If policy is consistently more responsive to the preferences of the wealthy or to those of non-Hispanic whites than to those of others when their preferences conflict then it might seem obvious that the advantage groups enjoy greater access to the means for influencing policy may seem obvious and that may in the end be right but before we agree it seems to me we ought to understand more clearly the mechanisms that produce differences in policy responsiveness. We need to explore that question I think separately for issues of wealth and race. So let me start first with wealth. We know that participation in political activity increases with income. And the rich may get the policy they want because representatives know that they pay more attention to politics and are more likely to vote. We also know that members of Congress are typically wealthier than their constituents and we have some evidence that they favor the interests of wealthy constituents because they identify with them. Now that favoritism may be objectionable in both cases but neither mechanism is obviously a matter of inequality in the means of influence. Still money is a means of influence and at least some of the skewed responsiveness plausibly reflects the fact that the rich control so much of it that's why we call them rich and they can use it to influence policy. Most of the scholarly attention to the devoted to the political influence of wealth is focused on two mechanisms. The first is the system of campaign finance which as we know depends on voluntary private contributions and the second is the system of lobbying and interest representation. I wanna say something about the first of these today and I'll come back to the second tomorrow. Well election campaigns are financed from multiple sources but we know that the bulk of campaign contributions increasingly comes from wealthy individuals that is as opposed to organizations. Those contributions can influence policy outputs in various ways but mainly by influencing the priorities of the candidates that they support but that influence is mediated by the election process itself. The candidate has to win their campaign in order to be in a position to carry out their policy priorities so the unequal means diagnosis of unfairness holds that campaign finance, the campaign finance system allows the rich to exercise greater influence over election outcomes by devoting greater means to supporting campaigns. But I think that idea is complicated and I doubt that we can fully explain the appearance of unfairness as a violation of a principle of equal means of influence for individuals. The principle is hard to interpret for the scaled up context of electoral competition and it doesn't register all of the reasons for concern about fairness in that scaled up context. So let me say something about each problem. The first one about interpreting the principle emerges when we recognize that the value of money in political campaigns depends on who's spending it. Most studies show that on average contributions to incumbents do not make much difference to their chances of success whereas equal contributions to challengers make a larger difference and one reason, of course, is that incumbency is usually itself a kind of resource. It involves command of various non-monetary means to influence election outcomes. So this means that equalizing the capacity of individuals to contribute to election campaigns won't necessarily redress the unequal impact of contributions on election outcomes. And so we might wonder whether the point of the equal means of influence requirement is really to equalize equal means of influence or instead to equalize a more abstract capacity to affect election outcomes. The problem is that although the latter seems more basic it might not be feasible to achieve and it might not even be feasible to make conceptually clear. And there's also a deeper problem. If we think about a system of competitive elections as an institutional process with certain purposes then standards of fairness emerge at the level of the system that are not concerns about the distribution among individuals of the means to influence outcomes. So let me give an example. Campaigns are partly exercises in communication among candidates and potential voters. They try to mobilize and to educate. Now a central interest of citizens in that competitive context is deliberative. Roughly speaking, it's to have an information environment that's conducive to making epistemically responsible judgments about how to vote. One complaint about large inequalities in resources available to candidates is that they might skew the deliberative environment for citizens in ways that undermine the citizen's capacity for responsible judgment. Now that deliberative interest seems to me really quite critical but satisfying it may not be compatible with satisfying the interest in being equally able to influence the outcome of voting which that's the equal means requirement. I could say more about this but let me just say quickly that the theoretical point here is that there are multiple evaluative perspectives and the outcome oriented perspective generates pressure to equalize the means of influence over election outcomes for individual citizens. That's we might say the perspective of citizens as agents but there's another perspective that of citizens considered so to speak as the beneficiaries of political competition and they have interests in being enabled to make good judgments about how to vote. And looked at from that perspective we might wish the system to promote competition between candidates or parties in a way that enables people to deliberate responsibly and to make informed choices and that doesn't necessarily require that citizens have equal means. Indeed, it's not a matter of the distribution of influence at all. The perspectives of citizens considered as agents and as beneficiaries of political competition those perspectives just don't necessarily converge and yet each one of them seems to be relevant to judgments about the fairness of a system of campaign finance. So we've been concentrating on the influence of wealth on the election of representatives but of course there's also other ways that campaign contributions can influence policy. The most familiar I think is that the need to attract and retain donor support can be an incentive for representatives to adopt their donor's policy positions even if the representatives don't judge those positions to be best on their merits. It's a form of judgment corruption. As TM Scanlon observes there is a failure here but it's that representatives are induced to be responsive to the wrong reasons. That is to say reasons that would be excluded in the performance of their duties as representatives and that seems to me entirely right but we should see that the underlying objection to the democratic principle isn't except incidentally to a violation of equal means of influence. It's to the undermining of the representative's motivation to perform as the legislature's justifying purposes require them to perform and that's just a different problem. So let me turn more briefly to the problem of differential responsiveness by race. Zoltan Hainal's study which I mentioned earlier was argues that policy is consistently more responsive to the preferences of non-Hispanic whites than to those of blacks. And that seems unfair on its face. The difficulty lies in diagnosing the unfairness and that difficulty is that it's not clear what the mechanism is that produces it. So I mean this is obviously I think too large a topic for me to say very much about here so I'm just gonna offer a few speculative comments. The Hainal study wasn't designed to identify a mechanism but you could read it to rule out some mechanisms. So blacks of course are the paradigmatic, discreet and insular minority. So one might think that they lose disproportionately because their political interests are systematically opposed to those of the majority. But Hainal's data show that while blacks have some political interests in common, they also differ about some interests and they share some interests with others. But whether there's agreement or disagreement, black voters consistently lose more often. Or maybe the disparity is mediated by differences in wealth or education or political views or partisanship. But the pattern remains even after those factors are controlled for. Institutional rules are not enough to explain the pattern of outcomes though probably changes in districting, for example might improve the descriptive representation of blacks and maybe make policy marginally more responsive. And then it has to be said that because black voters tend to live in more populous states to the extent that they have shared political interests, the small state bias of the Senate may contribute to the disadvantage. Now Hainal's own data show that black voters succeed less often in electing the representatives that they want. And they're also represented by blacks less frequently than non-Hispanic whites are represented by whites. In those respects maybe one could say that blacks are less well represented in legislatures. But that only pushes the mechanism question back a step because now we need to know what skews rates of electoral success by race. And this is speculative but it's hard to resist Hainal's own suggestion that racial prejudice among non-Hispanic white voters sometimes expressed as we've seen in racialized stereotypes and increasingly aligned with partisanship is a significant part of the mechanism that explains the relative lack of electoral success of black voters particularly in biracial contests. And similarly it's hard to resist Hainal's suggestion that discrimination by public officials contributes to the lack of policy responsiveness to black preferences. And we have experimental evidence to support this speculation, some done by David Brookman, your colleague here. According to this analysis, the problem isn't inequality in the means available to blacks for taking advantage of political opportunities. The unfairness could remain even if political means could somehow be fully equalized. I think the objection is that institutions allow discriminatory treatment to defeat black citizens' efforts to use their institutional opportunities to protect and advance their political interests. So I think to the extent that there's an institutional failure here is not a failure of political equality understood in the way I was describing earlier. It's a failure to protect against substantive inequity. And that's yet another dimension I think of the kinds of unfairness that we can find in the system of representation. So let me just finish today with sort of two observations. First, maybe not surprisingly, what emerges from these reflections is that there's no single standard of success or failure of institutions of representative democracy. They can fail in multiple ways. And we've seen some of them, they can fail to be decisive when there's conflict involving the means of attaining widely shared ends. They can frustrate popular control of government by being unresponsive to changes in the majority will. They can produce consistently inequitable outcomes for politically vulnerable minorities or perhaps even politically vulnerable majorities. And they can fail to support or enable deliberative environments that make for fair contestation and epistemically responsible choice. And I think we might think of each kind of failure as a shortfall from achieving a purpose that we could reasonably expect representative, democratic representative institutions to satisfy. And I think part of the work of a theory of fair and effective representation is to identify those purposes and the rules that should govern the mechanisms that they seek to achieve them. The second observation follows from our discussion of failures of fairness. In the cases we discussed what appears at the outset to be a matter of unequal influence turns out to mask a more complicated set of failures. And those dimensions only become clear when we think about representation as a system rather than only as a set of dyadic relationships. I think the preoccupation with constituency representation can obscure the critical role in determining policy outcomes that's played by the structure of political competition throughout a jurisdiction, the functions of parties within political competition and of course the political environment. To me, the conclusion is that for a critical theory of political fairness, the challenge is to understand the principles that should regulate democratic competition in its various forms at the level of systems as well as at the level of individual constituencies. And that's our subject for tomorrow. Okay, so we'll now take a brief intermission, let's say 10 minutes. It's a pleasure to introduce Martin Gillins. Gillins is Professor of Public Policy, Political Science and Social Welfare at the Luskin School of Public Affairs at UCLA, where he was very recently, but no longer, I think, and with probably a sense of great relief, Chair of the Department of Public Policy. Before joining the faculty at UCLA, Gillins taught at Yale and Princeton. As envious as we here at Berkeley might be that he's teaching at our rival to the South, we can take pride in his being a Cal product. He earned his PhD in sociology at Berkeley after a BA in sociology and philosophy at UC Santa Cruz. His books have won numerous awards. He has held fellowships at the Institute for Advanced Study in Princeton, the Center for Advanced Study in the Behavioral Sciences at Stanford and the Russell Sage Foundation. He is a member of the American Academy of Arts and Sciences. Those laurels may pale, however, in comparison with perhaps the single highest honor to which a liberal academic can aspire, a segment on The Daily Show where John Stewart interviewed him about his work in 2014. Gillins' sobering first book, Why Americans Hate Welfare, answers in a nutshell because of race. Gillins demonstrates that Americans massively overestimate the proportion of the poor who are black and that among white Americans, this overestimation bears a strong relationship to opposition to welfare. Gillins also shows that the news media too exaggerate the degree to which poor people in America are black and consistently associate images of blacks with the least sympathetic stories about poverty. No less sobering is the book Affluence and Influence for which Gillins is perhaps best known for which he garnered the coveted John Stewart interview. I found it so jaw-dropping that despite its unusual accessibility for a work of cutting-edge political science, I had to read it twice before I could absorb what it was trying to tell me. Affluence and Influence, along with Democracy in America, what has gone wrong and what we can do about it, co-authored by Benjamin Page, examine the relationship between what Americans tell surveys they want federal policy to be and what federal policy actually is. Gillins finds a modest association between citizen preferences and policy outcomes, which is reassuring as far as it goes. However, he also finds that when poor and middle-class Americans disagree with top-decile Americans on federal policy, actual policy outcomes strongly reflect the preferences of the most affluent but bear virtually no relationship to the preferences of poor or middle-income Americans. As you have already heard, this is precisely the sort of seeming democratic failure that Bytes aims to come to terms with in his Tanner lectures, which makes Gillins an ideal commentator. So to hear Professor Gillins' commentary on Professor Bytes' commentary on Professor Gillins, please join me in inviting Martin Gillins to the podium. Thank you, Professor Kaladney. I'm thoroughly confused about what I'm going to be commenting on now, but maybe we'll see how it goes. Thank you also, Chancellor Christ and Professor Bytes, of course, and Jane Fink, who's done a terrific job with the logistics of this event and this three-day event. So as you can see my eyes, I'm not a political theorist. I'm an empirical political scientist. Most of my research looks at sort of quantitative analyses of American politics, but my attempts to understand the facts of American politics are motivated by some of the same concerns that Professor Bytes addresses, right? Is our political system providing fair and effective representation to American citizens? So in the next few minutes, I want to share one set of facts about representation in the United States that stems from my research that you've already heard a bit about, focuses on sort of what shapes policymaking in Washington and I'll explain what I take the sort of normative implications of those facts to be. So the facts relate to what Professor Bytes describes as fairness to individual citizens or constituency responsiveness. And it might be fair to say that that concern has been sort of my preoccupation, at least in this sort of aspect of my research. So the project that I'm gonna describe a bit of is focused on understanding who influences policymaking in Washington and how influence over government policy or what I'll call policy responsiveness differs for different kinds of Americans, especially for those with less and more income and also what role organized interest groups sort of lobby policy makers in Washington play. So to gauge the influence over policymaking, I examined about 2000 proposed policy changes that were, you know, took place or didn't take place, but were proposed between the mid 1960s and early 2000s. And these possible changes in federal policy sort of spanned the gamut of things that the federal government does or plausibly could address, things like proposals to raise or lower taxes, you know, healthcare, education, foreign military engagements, abortion, same sex marriage, broadest range of issues as I could find data on. And I looked at policies both that made it onto the congressional agenda and those that didn't. So policies that sort of could have been addressed conceivably by policy makers in Washington, but weren't. And the point of that is to encompass in my analysis both agenda setting, which is to say sort of influence over what policies actually get taken up by policy makers and then the decisions that are made about those policies, what policy to actually adopt or not adopt. So for each of those 2000 or so proposed policy changes, I recorded the strength of support or opposition to that proposed change among Americans at the bottom, the middle and the top of the income distribution using surveys of, you know, national surveys of the American public. And then I also included a set of measures of how a sort of set of interest groups that are among the most powerful in Washington sort of fall on those 2000 or so policies that needless to say took a lot of time and employed quite a few graduate students over a period of years. But eventually I had these measures not only of public opinion, but of interest groups like the Chamber of Commerce or the National Rifle Association, the AARP, AFL-CIO and so on. And what I found, as you've heard, is that when taking into account the preferences of middle income Americans and affluent Americans and interest groups all sort of considered at the same time that the proposed policy changes were much more likely to be adopted if they were supported by the affluent or by interest groups than if they were opposed by those, you know, sets of actors. But again, once taking into account the sort of influence of all of these different sort of sets of political actors, the preferences of poor and middle class Americans made no difference at all. So a policy that was strongly opposed by the middle class was just as likely to be adopted as a policy that was strongly favored by the middle class. Again, once taking into account other actors' preferences. Now, that doesn't mean, of course, that the affluent always got what they want or interest groups for that matter. I mean, for one thing, affluent Americans don't all agree on the same policies. Interest groups are often in competition with each other. And as Professor Bytes mentioned, there's a strong status quo bias, right, that's built into our structures of government and exacerbated by political polarization in recent decades. So many policies favored by interest groups were the affluent, in fact, were not adopted. And maybe even more importantly, many policies that were supported by middle class or poor Americans were adopted, right? So the middle class and the poor, in fact, do frequently get what they want from policymakers in this country. But it's not because policymakers are responsive to the preferences of those groups, okay? So middle class and poor Americans get what they want when they agree with the preferences of the well-to-do or interest groups or when what they want is the status quo and there isn't enough sort of push in the other direction to overcome the sort of biases built into our political system. So this distinction between getting what you want and influencing what you get, right? This corresponds to the distinction that Professor Bytes was making between congruence, right? Simply the alignment of policies and preferences that come about for whatever reason and influence or responsiveness, the ability to actually shape what policymakers do. Now congruence is important, right? The less congruence there is between sort of policy outputs or other government decisions and what a group of citizen votes wants, the less satisfied that group of citizens is likely to be with their government. But still, if the majority of Americans get what they want, only when they agree with elites, then we don't have a functioning democracy, right? Congruence can be high in many autocratic or theocratic polities, at least across many issues or time periods, but of course that doesn't make those societies democratic. Nor does congruence alone give ordinary citizens any assurance that policy will change if conditions change and lead citizens to sort of re-examine what they want from government. So let me turn now to what I take to be the normative importance of these facts. Now my work is focused on fairness to individual citizens, right? And it turns out in this work, and of course my work's not alone in this as other political scientists have addressed these same issues with other kinds of data and other sort of analytic techniques, but in my work and in theirs for the most part, it turns out that all citizens are not equal and in fact certain kinds of citizens like the poor and the middle class are treated quite unfairly, right? They in fact lack effective influence over policy making decisions in Washington. Now there's many reasons why we might not want majority preferences to translate into government policy in sort of an immediate one-to-one fashion. We wanna protect minority rights. We may value policy stability, although I will say that empirical studies of public opinion suggests that in fact the public in the aggregate is quite stable in its views and when sort of policy preferences change among the majority of Americans, it's almost always for some like perfectly logical reason, changing economic conditions or an attack on our country or some other change in the situation. And of course we also have to recognize that there's a limit to the sort of range of policies on which sort of the majority of the public are very many members at all, really have the ability or the interest to form meaningful preferences, right? And even policymakers don't have the information that they need on most issues and they have to rely on others and that's part of what representative government is all about. So there's a limit to the sort of range of issues on which we might sort of plausibly look for a connection between what the public wants and what the government does. So there are a set of limitations to this sort of one criterion that I focus on in this work of responsiveness and equality of responsiveness. Nevertheless, I would argue that despite those constraints and limitations, that this notion of equal influence in the sense of the responsiveness of policymakers to public preferences is in fact, if not they then, if not the then certainly a central benchmark of what fair representation and effective democracy means. And I'd say further that the vastly unequal responsiveness of the sort that my work reveals is clearly incompatible with any notion of fair and effective responsiveness. Indeed is incompatible with really any common understanding of democracy. And going a little further, I'd argue that other normative concerns that exist at the systemic level, things like the anti-majoritarian structure of the Senate, that advantages small and rural and currently more conservative states or gerrymandering, which advantages one party over another, interest competition in the form of lobbying and so on, the contestability or competitiveness of elections. A lot of these concerns are rooted, if not entirely, then largely, and I would argue broadly and deeply, in concerns over the equal and effective influence of all citizens. So just to briefly mention a few examples that I think sort of fall into this category, like gerrymandering advantages one party over another. Why is this problematic? Well, it's problematic because it leaves the adherence, the voters or the supporters of the disadvantaged party with less influence over government than the adherence or the supporters of the advantaged party. I follow Professor Bytes in suggesting that the key outcome here is not the sort of one-to-one correspondence between a voter and his or her representatives, but rather the preferences of voters or a certain set or kind of voters and the output of the political system. And it's that disjuncture that gerrymandering and partisan other sort of structural features that create partisan advantage violate. We could take something like the pro-business tilt of lobbying organizations, right? Which is like pretty dramatically tilted toward business and away from other interests. And I would say that that's problematic to the extent that it pulls policy away from the preferences of citizens. Gridlock that Professor Bytes had talked about a few minutes ago. It's problematic for I think precisely the reasons that he articulated or for one of the reasons he articulated was that it resists movements of policy that would track the majority's settled view of the common interest. I don't know if majority's views are often reflective of their understanding of the common interests, but they're certainly reflective of their understanding of their own interests. And in the aggregate, those represent at least the, in one sense, the public's view of what is best for it. And so to the extent then that Gridlock prevents government from responding to those preferences, it becomes anti-democratic. So in sum, I agree with Professor Bytes's conclusions that the preoccupation with fairness to individual voters can obscure the critical role in determining policy outcomes that's played by the structure of political competition, the functions of parties within it, and features of the exogenous political environment. But I would suggest that sort of the normative evaluations of those sort of structural features bring us back to maybe not completely, but nevertheless substantially, to concerns about fairness to individual citizens and the equal responsiveness of government to everyone. So I'm well aware that I've left out a lot in these brief comments today. I haven't talked about how preferences are formed or the role of social groups really in the forming and contestation of preferences over government actions. There's complex issues having to do with the strength of preferences across different individuals or groups on different kinds of issues. Citizens have desires for descriptive representation in addition to policy outputs that correspond with their preferences. I haven't discussed citizens' obligations in a political system. So there's a lot that I've left by the side. But my claim really is not that fairness to individual citizens and responsiveness to their preferences is the only normative criteria we should apply in sort of evaluating the quality of our democratic institutions, but really that it is nonetheless a central criterion and one that is going to sort of emerge as we try to understand the fairness or the desirability of these other sort of features that are more structural or systemic level. Thank you. So I'd now invite Professor Beitz to the podium for his response. Well, thanks again, Nico, and thank you, Marty. I don't think I've been told I have up to 15 minutes to comment and I think I won't probably use the whole 15 minutes and partly because I'm not sure I have 15 minutes worth of things to say, but also I think we'd like to probably save some time for to hear from people who are in the audience. So let me begin by thanking Marty for the comments, which among other things explain more clearly and in more detail than I did in the lecture earlier. The really path breaking research that's reported in his book. And on reflection may not have said enough to convey the enormous influence that that research has had in political science. Together with some work by Marty's and my former Princeton colleague, Larry Bartels, Marty's work has made it impossible to ignore the class bias of public policy in the United States. That's something that a good deal of empirical work in the previous few decades probably overlooked, although it's not always been ignored. Schachtner, for example, wouldn't have been surprised by Marty's findings. I'm not sure that I think the one substantive thing I wanna say is that it's not clear to me how much we really disagree. But there are some things that Marty said toward the end of his comments that suggest that he reads me as not taking policy responsiveness of the kind that his research studies seriously enough at the level of individuals. And just to remind you, his research documents the relationship between preferences for various kinds of policy changes that are expressed in the survey responses of individual citizens and policy outcomes. And he finds that policy only tends to change in response to the preferences of most citizens when the change is also favored by the wealthy. And again, the preferences at stake here are preferences that are documented in responses to surveys. Marty thinks that that violates, or maybe it's evidence of a violation, of a democratic principle, the principle that citizens should have roughly equal influence over policy outcomes. Now, one reason that I'm not sure how much we disagree is that Marty agrees that there are some reasons why we might not always want majority preferences to determine policy. So perhaps when that would offend minority rights or it would make for harmful instability. And I think it's important here just to note that he also recognizes that the public may not be able to form meaningful preferences on what he calls technical or complex issues. Now, I guess I think that those categories of exceptions to the majority preference principle seem to me pretty substantial, particularly the last. And I think we know from what's now decades of public opinion research that most of the public does lack meaningful preferences about many policy issues that we might describe as technical or complex. And unfortunately, we don't have a very reliable institutional means of finding out which policy issues belong in this category. But offhand, they probably constitute a fairly large part of the policy agenda. And if that's right, then it's just not clear that the right normative diagnosis of Marty's findings is that they demonstrate a failure of what we might call procedural equality or the principle of equal effective influence over policy outcomes. It seems to me more plausible to think that the objection is to the substance of policy. I mean, maybe that it devotes an unjust share of social resources to satisfying the interests of the wealthy than the interests of others. And I can imagine a fairly strong argument that that in itself is unfair. But it's obviously a different problem than the problem of unequal preference satisfaction. So as I've said, this may not be a disagreement. Marty might think that a continuing lack of responsiveness to the preferences of the non-rich when those conflict with the preferences of the rich is evidence of a failure to treat the interests of the non-rich fairly. And that might well be plausible. But in that case, we shouldn't say that responsiveness to preferences is the basic requirement of democratic decision-making, except in the indirect sense that responsiveness to preferences may be evidence that policy is responsive to interests. I'm not sure if this is actually a distinction that matters for the conduct of empirical research, but I do think it matters for kind of getting the diagnosis of the democratic significance of the results of empirical research, right? So Marty has said a lot that I could also comment on, but I think I'm gonna just leave it there and we're gonna hear from people in the audience. I hope so. Okay, good, thank you. Why don't you stay at the podium? All right. Yeah. First, we owe you a round of applause. Thank you so much. And if there are questions, then Jane Fink has the mic. Yeah, Professor Mansbridge. Sorry. I'll have some time to comment tomorrow, so I'm asking a question only because no one else had a hand up. So just to respond to your last point, Chuck, thinking about complexity as a reason for not responding to the preferences, the immediate preferences of the public, that wouldn't explain the bias to the rich, would it? Unless the rich somehow, the interests of the rich mapped onto complexity. In other words, yes, it's true that there are reasons to deviate from the preferences of the public, granted. And you could list one, two, three, four, five, six, seven reasons, but unless those reasons mapped on to the interests of the wealthy, they wouldn't explain Marty's finding. Marty's finding would still point to a normative failure in responding to preferences. In other words, it's sort of like controlling for class. You control for complexity and you then continue to look at an effect on the wealthy. Now I would say there was a normative effect after controlling for complexity. What would you say to that? I may have spoken too quickly, wouldn't be the first time, but I don't think I meant to suggest that complexity in itself is meant to play some explanatory role. I think my point just was that the, I mean, to me, the question is kind of how to understand the normative failure that we see when we see that policy tends to track the preferences of the rich when they conflict with the preferences of others. And the point I meant to make with it, I'm not sure that I think that the failure just is a failure of equal influence of other people in the electorate. I mean, there's a failure, but the failure seems to me at least on the face of it to be that decisions are reflect an unjust or unfair allocation of social resources to satisfy the interests of the rich. And then we could think about the mechanism that makes that possible. And we may wanna say that the mechanism has to do with the undue influence of the wealth of the rich in the political process. But then I think what makes that undue influence objectionable isn't that it produces outcomes that don't correspond with the survey, with the political preferences of people as they're reflected in surveys. I think it's that our system is enabling the rich to use their money in ways that produce outcomes that are substantively unjust in the sense that I was just trying to articulate. So I didn't mean to suggest that complexity is somehow, I mean, I think one of the lessons of thinking seriously about representation as kind of view of all people have explained to us is that it's a complex systemic process. And it can be a mistake in trying to model representation simply as a kind of a larger scale version of individuals trying to conduct themselves as democratic citizens in the way that they would if they were members of town meetings. But that's really just a reflection about how we articulate the reasons for the failures of political outputs to satisfy the substantive criteria that we think they should satisfy. And I think sometimes the answer is just that people are, that institutions are obstructing people and exerting influence over outcomes. But I think that often it's not that, I think it's often other features of institutions and the focus on individual attempts to exercise influence can just deflect us from these other ways that institutions can be systematically objectionable. So it's not to say that, sorry, sorry, sorry back there I know, these guys have headphones on. So I think there are times that the equal influence test fails. I just don't think that it gets us a whole picture of what kind of unfair democratic procedure, what it is that makes democratic procedures unfair. So Christopher Coots. Thanks very much for this wonderful lecture. I look forward to tomorrow. My question is a little bit connected to Professor Mansbridge's maybe it goes to a little bit to tomorrow. One thing I didn't hear you say much about is the kind of epistemic dimension of democracy. I think you made reference at a couple points to deliberation. But I mean, one of the findings as I understand it is that our electorate is increasingly sorted on it by educational dimensions with particularly college educated Democrats having very distinctive policy preferences. And if we think about education sorting and if I mean I one hopes the education results in more accurate perceptions of forms of fairness and unfairness as well as more effective and less effective responses to policy systems. I mean, that seems like a place where you might think that I mean, I guess it's an old million thought we want some inequality in influence if that inequality flows to more educated voters. At least that would be one argument for that. I mean, I don't want to endorse the kind of Jason Brennan view of restricting voting but you might think that there would be various, some patterns of unequal influence that actually might get us better outcomes including outcomes that are more responsive to real interests that are shared by the public. And so I was just asking really to comment on that. Yes, well thank you, Chris. I guess two things. First on the first part of your point which is that I didn't say very much about the epistemic environment in which people form their political preferences. I do think that's a chunk that's missing from these lectures. I felt there's a whole subject there to be addressed and I just couldn't find a way to do it in the time I had. But as I will say tomorrow, I think in sort of in multiple parts of what we might think of as the public sector or the public sphere, I think there are significant questions about the way that our formal and informal norms structure what we might call deliberative contestation. And I think those norms make a big difference in the quality of the environment in which people learn and form the preferences that they form. So I think that's a major topic. Just, I just haven't addressed it here. I hope to be able to do that in the future especially because I think that the changes in the structure of the public sphere in the last 20 or 30 years are just hugely consequential. Now on the other question about whether there's an argument that people who are better educated should have more influence. I don't know if you ever attended a faculty meeting you might wonder whether that's a good idea. I just think of the famous million argument that you mentioned that people who were, I think in principle the argument was that people who had a higher capacity for political judgment should have more votes. And operationally the way we identified the people who had a higher capacity for political judgements was either by looking at their university credentials or looking at whether they were members of learned professions. As you know, Mill himself came to abandon that view later in his life. Not because he thought it was in principle wrong but for I think the good reason that there was no way to figure out who actually had more political competence and probably measuring it by level of education was not the most reliable way. So I don't know, I mean this is an argument we often have with our students about, which is kind of meant to bring out questions about what the real foundation is of our convictions about political equality. If we granted Mills empirical assumptions what would we think wrong with his proposal? And I think there are some fairly powerful things that might be said about that. I think I'll say a little bit about some of this tomorrow. But I think it's not just about producing good outcomes. I think it's also about other features of a democratic political environment that influences the way that people think about each other in themselves. Jay Walts and then Martin Jay. Thanks very much. Thanks very much. I checked for the, and Marty for the, she moved to New York, remarks to Dan. I just wanted to pick up on an issue that I think touches on most. Chris was talking about, sorry. I wanted to pick up on an issue that dovetails with what Chris was talking about. There was one point in the lecture where you talked about the information environment in which preferences get shaped. And this is when you were suggesting that maybe one of the problems with the influence of money on our political culture is that the effects it has systematically on the information environment in which democratic citizens are making decisions about whom to vote for and how to exercise their right to vote. And I guess my question is just an invitation to you to say more about that. That seems intriguing to me that there certainly seems like there is an interest that we have in a kind of information environment that's gonna be conducive to making informed choices at the ballot box. But at the same time, that interest doesn't seem significant unless it connects up to our interest in individual kind of equal political influence, maybe the kind of procedural equality that you were trying to distinguish this systematic issue from. I mean, if we had an ideal information environment in which we made beautiful decisions and we cast our ballots, but no one counted them, you know, seems like that wouldn't do very much for us, just to put it crudely. So could you say a little bit more about the systematic conception of unfairness that you were gesturing toward here? Well, let me say something, it probably isn't gonna be enough, but so there's kind of two issues here. One is about the relationship between campaign funding and the epistemic environment of campaigns. And the other is about whether our interests in let's say an adequate epistemic environment can be analyzed in terms of interest and equal influence. So, you know, on the first point, I think, I mean, there's a lot of empirical literature and I don't know how well-developed it all is, but literature about the, so what does campaign spending do? Well, most of it pays for communicative activity, not all of it, but most of it pays for various kinds of communicative activity. And the kind of standard thing to say about that is that inequalities in financial resources available to candidates can produce a skewing in the epistemic environment. That is not, that may advantage one candidate over another, but may not serve the deliberative interests of citizens. And I think there's kind of some empirical questions about whether, how true that kind of picture is, but setting them aside, this is something political theorists do. They'll say, let's just set aside the empirical questions and go straight to the principles. I think that the question there is just sort of, is what's the nature of the interest that we take as citizens in the environment that's created in the context of political campaigns and how do we understand the harm that's done if communications actually are skewed in the way that this kind of common view suggests? And I just don't, this connects with the second point. I just don't think that the interest there is fundamentally in citizens having equal influence. As I was trying to say in the lecture, I just think we have more than one evaluative viewpoint when we think about the kind of structure of the deliberative environment. One, I think I don't mean to deny that we have an interest in having some kind of equal influence over outcomes. I myself think that that interest kind of under-determines a good deal of the institutional structure, but it determines some of it. But then I also think we have interests in the character of the environment that are not interests in having equal influence. They're interests in being able to form views we can accept as kind of responsible views about how we ought to vote. And I just think it's a different perspective. I think it's a perspective of the, and that's because the underlying concern isn't so much that my vote have as much influence as yours. It's that my vote contribute to the right outcome, and that is to say an outcome that's best for me and people like me. And there the evaluative perspective is not the perspective of the citizen's agent trying to exercise a share of control over government. It's the perspective of the citizen who is gonna be the beneficiary of the things that government does. So I say, and I say, I think it's in a footnote, although I may say it in text tomorrow, but this is a distinction that goes back to something Hobbes says when he remarks in Leviathan that citizens are both the makers in the matter of government. Of course, in his view, citizens are the makers of government only in one act, and after that they're just the matter. But the point is, there are these two perspectives, and there's no reason to believe that they point in the same direction when we're trying to make judgments about how our interests are best served in institutional design. Yeah, I would just add that I completely agree, and I think this particular sort of dimension of preference formation and the communicative environment is the one where it seems clearest to me there's a disjuncture that it can't be in some fundamental way reduced to a question of equality of responsiveness. I would also just put in a good word for the public and as much as Americans are widely understood to be poorly informed and probably also widely viewed as at least half of them being bamboozled into thinking something that goes against their interests. And while there's a fair bit of truth in that, there's also under appreciation for the sort of wisdom of the crowd and the sort of reasonableness of public opinion taken as a whole that we could maybe talk about more tomorrow, Wednesday. I was struck by your rhetoric of failure, which seemed to presuppose a normative notion of success perhaps projected backwards and agronistically to the founders. You mentioned the Madisonian element that produced a compromised constitution, which from the get-go created, I think, a tension between the kind of democracy that you and I think I also share as a normative goal, a tension between that and a restriction against the dangers of unchecked democracy, something that was there, of course, in the constitution in a variety of ways that I need not spell out from a restricted franchise to a free-fits rule, to the Senate, to federalism in general, and lots and lots of ways in which the constitution was not in any extreme way democratic. Now, one might say, and there are people today who have even gone so far as to argue for constitutional convention to create a new one, I think, of people like Sam Moyne, that the constitution itself is the cause of our dilemma, that we can't assume that democracy is our telos because the constitution is a weight against it. Now, one might argue that that's the case, but there's also in a somewhat optimistic countercase that could be made, which is that all of these obstacles also prevent what might be called the demagogic democratic populist abuse, which is latent in the Trump movement that we now see that has the potential to become a majority in a kind of 51% way, and which would have absolutely no use for those checks, and which would, in a sense, run roughshod over the Madisonian dimensions of the constitution. So in a weird way, it's a double-edged sword, an obstacle to what you want, an obstacle, one hopes, to what they also aspire to. Well, thank you. I think those are actually very good points, and I don't know what, how much I can say in response to them. I mean, I do agree with the sort of thrust of the comment that it's just too simple-minded to say, it may not be wrong, but it's too simple-minded just to say that the problem with the constitution that it's insufficiently majoritarian, and we should change it to make it more so. I think the Madisonian arguments are not, in principle, wrong, that a constitutional structure should, while on the one hand enabling durable majorities to govern, on the other hand, should obstruct some of the predictable pathologies of systems of majority rule. I think that's what the argument you described kind of comes down to, and there certainly are ways of understanding populism that see it as a predictable pathology of majority rule. So, at some level of abstraction, I wouldn't disagree with what you, with the sort of the basic point of principle you make, and I hope the optimistic projection is right, that our institutions are such that they might obstruct a kind of a rise of kind of Trumpist populism in the future, but I'd go back to something I said early on, there's a difficult question about how to understand the relationship between institutional structure and the kind of exogenous political environment. And it seems to me just naive to think that those things don't actually participate together in determining the kinds of political outcomes that we're likely to see. And that's to say that there may be ways the political environment can change that can undermine the institutional structure and cause it not to operate in the way that its designers thought it would operate. And I think we may well be getting to that now especially because I think that the designers didn't think about action by popular majorities that would undercut the kind of procedural norms of democratic rule, but that's kind of what we're seeing. And if the kind of populism that you refer to were to kind of develop, one of the reasons it would prevail in what looked like a majoritarian way would be by undermining the procedures that guarantee that when we count votes we actually know who the majority is. And so, you know, when I said earlier on that I thought that the persuasiveness of the Madisonian case depends on some typically unarticulated assumptions about the political background, that's what I meant. When the political background changes in ways that undermine the functioning of the institutions, the argument for the institutions itself has to change. And I think people who are kind of making the argument that you're suggesting now that we should think better of the Madisonian structure than we sometimes do need to come to terms with kind of just that phenomenon, that the political environment may be changing in ways that will just frustrate the Madisonian expectations. Well, thank you all for coming. And I hope we'll see you here tomorrow for part two. But before we leave, why don't we give just one more round of thanks to our Tanner Lecturer and to our first commentator, Martin Gibbs.