KNOWING BETTER: IMPROVING COLLECTIVE DECISION MAKING IN HIGHER EDUCATION SHARED GOVERNANCE By CHRISTOPHER J. D. MANICK A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2016
2016 Christopher J. D. Manick
To my family
4 ACKNOWLEDGMENTS This project is not just about the power of collective intelligence. In many ways, it is also the result I sincerely thank the members of my committee: Les Thiele, Leslie due to their helpful feedback and support. Special thanks are due to Ian McMullen and Matt Chick at Washington University for inviting me to present a chapter at the Workshop in Politics, Ethics, and Society, and to its ever generous attendees, who provided invaluable feedback. The Friday afternoons spent at WPES have been formative, illuminating, and ha ve helped determine the broad strokes of this project. comments on a chapter I presented there. I o we a special debt to all the members of the 2015 Summer Institute for Civic Greene for the invaluable insights they shared and feedback they gave about this project. A chapter on academic civic renewal was ultimately cut, but the spirit of such an idea still underlies the entire project. Of the many graduate colleagues who have helped me along the way Jennifer Forshee and Dustin Fridkin deserve special mention. This pro ject has benefitted from their hospitality, friendship, wisdom, institutional insights, and intellectual engagement. Most of all, I would like to thank my family for always supporting my educational pursuits: Dad, Debbie, Cindy, Grandma, Yia yia, Megan, an d Traci. They have provided for me in countless ways throughout my education and I could not have finished this
5 project without their help. A special thank you is in order to Sarah Bisceglia, for being there for me every day and enduring the inhuman demand s a semi structured graduate program places its students and their loved ones. I also thank the Hunks Ryan, Matthew, Ch ris, Christopher, and Kiren Their constant support and distractions have been crucial for finding a sustainable work/life balance. Fina lly, I am most indebted to my mother, who did not live to see this project to completion, but whose unwavering goodness, empathy, and concern for helping others are woven into its pages. I am grateful for the motivation and purpose that both her life and d eath have given me.
6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ ...... 4 LIST OF FIGURES ................................ ................................ ................................ .............. 8 ABSTRACT ................................ ................................ ................................ .......................... 9 CHAPTER 1 FINDING REASON TO DEMOCRATIZE: AN INTRODUCTION .............................. 11 Introduction ................................ ................................ ................................ ................. 11 Building a Smarter Camel: The Epistemic Promise of Deliberation .......................... 12 Epistemic Deliberation, Defined ................................ ................................ ................. 16 Why Deliberation? Whose Deliberation? ................................ ................................ ... 20 Limits to the Argument ................................ ................................ ................................ 20 Epistemological Assumptions and Premises ................................ ............................. 23 Background and Context ................................ ................................ ............................ 27 Civic Renewal and Education Reform ................................ ................................ 31 Finding Reason to Democratize ................................ ................................ .......... 33 Chap ter Summaries ................................ ................................ ................................ .... 37 2 FROM PROCESS TO OUTCOME: THE DISCONTENTS OF SHARED ACADEMIC GOVERNANCE ................................ ................................ ...................... 40 Introduction ................................ ................................ ................................ ................. 40 Shared Governanc e: Its Meanings, Complexities, and Failures ............................... 41 From Inclusion to Division ................................ ................................ .......................... 44 Shifting Landscapes and Wicked Problems: Explaining the Divided Governance Trend ................................ ................................ ................................ ........................ 47 Reactions to Shifting Landscapes ................................ ................................ .............. 51 Divided Governance Through the Optic of Democratic Theory ................................ 55 Fairness as Rightness: The Procedural Conception of Democratic Legitimacy ................................ ................................ ................................ ......... 59 Collective Wisdom: The Epistemic Conception of Democratic Legitimacy ........ 65 Expressing Interests Versus Adjusting Beliefs ................................ .................... 72 Shared Governance: Wicked Problems, Procedural Sol utions ................................ 75 Conclusion ................................ ................................ ................................ ................... 80 3 THE POWER OF PERSPECTIVE: COGNITIVE DIVERSITY, EXPERTISE, AND DIFFICULT PROBLEMS ................................ ................................ .................... 82 Introduction ................................ ................................ ................................ ................. 82 ................................ ......... 83
7 The Expertocratic Case for MOOCs ................................ ................................ .......... 85 Cognitive Diversity: An Inclusive Approach to Wicked Problems .............................. 90 Diversity Theory ................................ ................................ ................................ ... 91 Uncertainty, Risk, Democratic Inclusion ................................ .............................. 93 Difficulty, Complexity, Cognitive Diversity ................................ ........................... 97 .................... 101 Conclusion ................................ ................................ ................................ ................. 104 4 WHEN TO TALK IN SCHOOL: AN EPISTEMIC THEORY OF DEMOCRA TIC DELIBERATION IN ACADEMIC SHARED GOVERNANCE ................................ ... 107 The Problem of Group Decision Making ................................ ................................ .. 107 When Deliberation Works ................................ ................................ ......................... 110 Epistemic Requirements For Deliberation ................................ ............................... 112 Deliberation as Decision oriented Debate ................................ ......................... 114 Deliberation as Reason giving Debate ................................ .............................. 119 Relevant Diversity, Democratic Participation ................................ .................... 128 Limiting Inclusion and Diversity ................................ ................................ ................ 132 Conc lusion ................................ ................................ ................................ ................. 135 5 THE SILENT MAJORITY RULE: RETHINKING THE EPISTEMIC VALUE OF NON UNANIMOUS DECISION RULES IN COLLECTIVE DECISION MAKI NG ... 137 Introduction ................................ ................................ ................................ ............... 137 The Epistemic Defense of Unanimity in Broader Context ................................ ....... 139 Necessary Conditions of the Epistemic Defense: Meta consensus, Shared Interests, and O racle Solutions ................................ ................................ ............. 143 Problems with Unanimity ................................ ................................ .......................... 149 Non epistemic Objections: Agreement Through Conformity ............................ 150 Epistemic Objections ................................ ................................ .......................... 154 Indivi dual Judgments of Collective Solutions: The Problem with the Diversity Argument ................................ ................................ ................................ ............... 156 Epistemic Acceptance and the Rule of Non oppos ition ................................ .......... 160 Non opposition as Realistic Stopping Rule ................................ .............................. 167 Conclusion ................................ ................................ ................................ ................. 170 6 CONCLUSION ................................ ................................ ................................ .......... 173 From Talk to Work ................................ ................................ ................................ .... 173 Institutional Design: Solving Problems, Implementing Solutions ............................ 175 Assessing the Advantages of Deliberative Problem Solving ................................ ... 180 LIST OF REFERENCES ................................ ................................ ................................ 186 BIOGRAPHICAL SKETCH ................................ ................................ .............................. 207
8 LIST OF FIGURES Figure page 5 1 Chart of the deliberative acceptance decision making process. ......................... 163
9 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Phi losophy KNOWING BETTER: IMPROVING COLLECTIVE DECISION MAKING IN HIGHER EDUCATION SHARED GOVERNANCE By Christopher J. D. Manick December 2016 Chair: Leslie Paul Thiele Major: Political Science This dissertation addresses the question: Should higher education governance and decision making be an elite or collective responsibility? It brings into conversation (i) the debate over the future of shared (i.e. participatory, faculty) governance in higher education, and (ii) research in democratic theory, specifically the epistemic defense of deliberative democratic decision making It argues on epistemic (or knowledge based) grounds that the answer to improved decision making in academic governance is better, n ot less, deliberation. C olleges and universities should use deliberative decision making mechanisms to tap the collective knowledge of its academic units to identify problems and likely solutions. To this end, I propose an epistemic theory of deliberation and inte nd it as an improvement and replacement to the governance practices that fail to meet the deliberation does not simply mean more; there are also limits to the usefulness of deliberative approaches. My argument supposes that deliberative and democratic practices should be used to the extent that they provide epistemic benefits. When
10 problems and solutions are collectively identified i.e. once a group knows what to do expert s often prove more effective at getting things done. I conclude that effective shared governance consists of both a deliberative democratic body that can figure out what to do and an expertocratic body that can then get things do ne with the nimbleness and adap nt shared governance practices.
11 CHAPTER 1 FINDING REASON TO DEMOCRATIZE: AN INTRODUCTION Introduction This dissertation addresses the question: Should higher education governance an d decision making be an elite or collective responsibility? It brings into conversation (i) the debate over the future of shared (i.e. participatory, faculty) governance in higher education, and (ii) research in democratic theory, specifically the epistemi c defense of deliberative democratic decision likely to be correct, whether in terms of facts or of values, if decision makers have is widely considered a failed or failing decision mechanism (Birnbaum 2004, Kezar and Eckel 2004) From the observation that it often does not work, some conclude that it quite simply cannot work As one university president puts it, (Rosenberg 2014) While it still enjoys paradigmatic status, critics see shared governance as an idea that has outgrown its usefulness. claim decision maki ng, and that shared governance protocols are simply unequipped to handle these conditions (Bowen and Tobin 2015, Duderstadt 2004, 2009, Fish 2007, Keller 2004, Pierce 2014, Roth 2015) It is a school bus when schools need a Ferrari. Thus, there are inc reasing calls to replace these inclusive, collaborative, or deliberative models with elite highly skilled oligarchic decision making bodies which I will refer to On the other hand, theories of democracy and democratic decision making are producing increasingly more evidence that large, diverse groups of people tend to
12 produce better solutions to problems and p redict outcomes more accurately than highly skilled problem solvers (Hong and Page 2001, 2004, 2012, Landemore 2012, 2013 Landemore and Elster 2012, Page 2007) This emergent property, referred to as follow democratic participation rules instances of real world participatory governan ce are increasing in popularity as experiments in participatory budgeting (Fung 2006, Fung and Wright 2003) crowdsourced policy making (Warren and Pearse 2008) and deliberative polling (Ackerman and Fishkin 2004, Fishkin 1991, 1995, 2009) support the cla im of collective intelligence. 1 If there is any truth to this research, then we can refute the claim that shared governance cannot work, even while acknowledging that it often does not. This suggests the failures of shared governance lie not in their inclusive or participatory nature a s many expertocrats charge but in how these practice s are used in specific settings. I believe colleges and universities neglect this collective wisdom at their peril. Building a Smarter Camel: The Epistemic Promise of Deliberation A camel, it is said, i s a horse built by committee. This sentiment is well illustrated on the hit television show Parks and Recreation which centers on the employees of the p arks department of a fictitious Indiana city g overnment. On an episode Camel a com petition is being held among departments to design a mural to be hung in city hall But rather than solving the problem of how to visually capture the spirit of their : a collage of images of all the things that the y individually prefer. Like the aphorism implies, the mlange that results is the product of 1 For a list of org anizations, see Fung (2015: 2).
13 compromise that betrays a lack of consistency, central leadership, or vision. While they enjoy making it, the committee members have produced a guaranteed loser, f eaturing such things as an image of a burning building, neon shapes, and rats made from garbage. In desperation, they call on a friend of the department Mark, who draws a simple, non offensive park scene. Though boring, Mark argues that the image will win because it gives the judges everything they are looking for in a mural (mostly because it whether to present their fun to make bu t unsightly mural or the bland but guaranteed winner. This episode reveals a popular and commonly held intuition: that while democratic decisions may be fair, include concerns and interests, and build relationships, they are not always good decisions viewers as obviously in an effort to save the department from its own myopia. It presents, in other words, a dichotomy: the democratic or the best. Such a dichotomy, I argue here, is a false one. We can see this b y considering literally a collage of the images that each person wants to see on the mural. This artwork by committee is an analogy for government work more generally: the product of negotiation and deal brokering with an eye to pleasing the decision making parties. In camel, he is applying a standard of correctness what the judges proposals will approve of.
14 Thus, the difference between the two images is not just that one is made collectively while an expert makes the other. It is also that the camel is a mass of prefere nces while the other is the product of judgments about what is the best choice. The democratic best dichotomy relies on this difference and is thus possible only if 2 we assume that democracy is a tool for aggregating irreconcilable preferences or interests such as interest group politics. But this assumption need not define all democratic decision making. When groups share interests or goals, such as how to promote a common good, democratic decision mechanisms like deliberation and consensus building tend to produce the best outcomes, not just the fairest (Mansbridge 1980). emergent property that pools knowledge, amplifying correct information while canceling out incorrect informatio n in such a way that the group collectively produces smarter decisions and choices than its members are individually capable of producing (Hong and Page 2004, Page 2007). This dissertation accepts the assumption of collective intelligence and, in doing so, necessarily sidesteps the intuition that democracy is only useful for mediating or aggregating disparate interests. Rather, groups produce collective intelligence when they share common interests, hold those interests as a standard of correctness for achi eving a mutually preferred outcome, and deliberate over how best to achieve that outcome. In simple terms, this means just that if we can agree on a preferred outcome, then democratic deliberation is likely the best mechanism for finding the solution(s) th at will get us there. To this end, I show how diversity and 2 Many words, phrases, and concepts in this dissertation should be italicized for emphasis. However, the required format disallows the use of italics for plain English phrases. In an effort to meet this requirement, I have u sed the regular font and indicated in footnotes which phrases would otherwise be italicized. Here,
15 deliberation can be used to improve the quality of collective decision making. Committees can design winning horses, not just camels. Doing so requires that we think about collective decision maki ng bodies as more than just fairness forums. Instead, we need to seriously consider how an institution like shared academic governance can use collective intelligence and smart decisions to some right (or at least best) answer concerning how it ought to be resolved that is 2014: 282). 3 This project thus attempts to improve the procedure s and practices of shared governance by answering questions about what Anderson (2006: 8) calls [D]o institutions of a particular type have the ability to gather and make effective use of the information they need to solve a particular problem? Given the epistemic powers of such institutions, what problems ought to be assigned to them? How can they be designed so as to improve their epistemic powers? Accordingly, t his project argues on epistemic grounds that the answer to impr oved decision making in academic governance is better, not less, deliberation. Epistemic democrats assert that in the face of future uncertainty in situations where knowledge of past events does not provide any insight into future events inclusive and demo cratic decision mechanisms are likely to produce better decisions than non democratic mechanisms because it maximizes the perspectives contributing to a solution Relying on these arguments, I suggest colleges and universities should use group deliberative mechanisms to tap into the collective knowledge of its academic units to identify 3 See Chapter 2 for a detailed discussion of epistemic harms.
16 problems and likely solutions. This project complements many arguments in favor of democratizing the academy on procedural or moral grounds. To this end, I develop an episte mic theory of deliberation that pays special attention to shared governance applications. Epistemic Deliberation, Defined Deliberation is here defined as the decision oriented exchange of reasons for and against a proposition or set of propositions (explained in detail below). It may be the case that epistemic value is definitional of deliberation, since the point of weighing reasons is to improve the quality of information in order to make a good decision (Miller 2013) If this is the case, then deliberation is not about fairness at all but rather about accuracy. This would be a very good reason to replace all aggregative dec ision mechanisms, like voting, with sessions of reasoned debate. This would also make for a very short dissertation However, neither history, common sense, nor the literature on deliberative democracy afford us reasons to think that everyday or democratic cases of deliberation meet this standard. Rather, we assume here that in order to produce epistemic value, deliberation must meet certain conditions. Chapter 4 explores the conditions under which deliberation meets this epistemic standard. For now, it wil l suffice to define our terms and avoid any unnecessary ways. As a form of practical reason, deliberation is traditionally understood as weighing and reflecting on re (Wallace 2014) However, the decision oriented quality has largely dropped out in everyday usage such that it now means to carefully consider a topic. For instance, Manin (2005: 1 5) reports that the Oxford English Dictionary used to define the term as:
17 1. The action of deliberating, or weighing a thing in the mind; careful consideration with a view to decision. 2. The consideration and discussion of the reasons for and against a measure by a number of councilors (e.g. in a legislative assembly). (Oxford 2015, Cf. Mansbridge 2015) Gone are both the reason giving and decision oriented components. One c ould perhaps argue that reasons are the things being considered and discussed, but that is by no means obvious from the definition itself. Similarly, Merriam Webster (2015) a discussion and consideration by a group of persons (as a jury or legislature) of the reasons for and there is no indication of a decision orientation. Deliberation is comprised only of whether or not deliberation is even a matter of practical reason. In sum, it is by no means obvious that deliberation can be said to definitionally possess either of these components. The more recent definition mi ght be said to describe the deliberation critics requires, among other things, reinstalling a decision orientation and reason giving. This debate has carried into the study of deliberative d emocracy. In a recent work, for instance and interests regarding matters of common three core attributes: first, communication must take place between people. This collective understanding stands in contrast to (or subsumes) what Goodin (2003) calls weighs reasons at the personal rather than
18 interpersonal level. The second attribute requires that members are willing and able to reflect on the discussion and change their minds when a better argument is presented. It thus relies significantly on honesty and fairness (ostensibly unproblematic because the issues discussed are of common concern.) Third, the information discussed must This definition is admirable precisely be cause deliberation in democratic theory has historically been weighed down by other normative requirements. For instance, early theorists (Cohen 1989, Habermas 1975, 1996) framed deliberation as reasoned 4 of full rational in general, not exclusive to the democratic contex t. In crafting this definition, she purposely omits other qualities e.g. public reason, absence of power, inclusiveness that are rather common to deliberative democratic theory. 5 She reduces these to standards for judging the level or quality of deliberati on. Thus, the giving of public reasons, the absence of power, or the right to inclusion are, inter alia inputs into the Epistemic value is likewise considered a standard. However, it is a unique list item i n that it is produced when deliberating improves the quality of information and 4 Cohen takes this term from Rawl s (1971) who himself took it from Kant (1788/1997) goal or standard itself. 5 Her full list includes: respect, absence of power, reasons, aim at consensus, common good orientation, equal participation/power, inclusion of all with legal rights, accountability to constituents, publicity/transparency, sincerity, epistemic value, and substantive balance. See Mansbridge (2015: 36). For an earlier version of this criteria see Mansbridge (1999)
19 leads to smart decisions; that is, it is an outcome of, rather than a factor contributing to, good deliberative process. It is not a quality of the deliberation itself but of t he outcomes deliberation yields Other standards can occur and be judged independently of each other. A deliberating group, e.g., can obey the rule of publicity or public transparency without aiming at consensus, being respectful of one another in debate, or giving reasons for their propositions. But collectively intelligent problem solving is only achievable when deliberation is done well. purposes, it may be too light. Neith er reasons nor decisions are mentioned, which leaves the reader unsure as to the practical uses of this definition (which has ostensibly o as to create meaningful discussion without devolving into Crossfire esque yelling? Whether you and I disagree on marginal tax rates, the morality of abortion, the stringency of an Iranian nuclear agreement, or what should constitute an intensive writing rubric, there is little promise of agreement in merely signaling our preferences to one another. We must be able to explain to others the reasons for our preferences. Thus, we cannot do away with the reason giving requirement. Furthermore, there is little to be gained by weighing and evaluating any information if the discussion is not oriented toward finally arriving at a decision. As I argue later, this requirement acts as a current that moves deliberation along, and not just for epistemic reasons. More im portant ly ( and as I will illustrate ) without giving reasons with a decision orientation, there is little chance that deliberation will produce epistemic benefits Therefore, I argue
20 here that if deliberati on consists of reason giving and decision oriented debate among a relevant number of diverse thinkers, then it will tend to produce the best or right answers. Why Deliberation? Whose Deliberation? making (Chambers 2003) makes for wiser or more effective shared governance practices. In fact, many critics complain that shared governance is already too talk centric. I argue that deliberatio n does in fact produce smart outcomes but only when it is done well. The field of deliberative democracy examines how group deliberation can ( and does) yield collectively justifiable political decisions. Deliberative democratic theory came about largely in response to aggregative conceptions of democracy that rely on majority rule (Bohman 1998, Cohen 1989, Gutmann and Thompson 1996, Knight and Johnson 1994) The aggregative model is built on the democratic ideals of equality and fairness : in a world of irre the best ( and perhaps only) way to extend political power to millions of politically equal person one vot which Elster (1997) which political legitimacy is based on reasoned public debate over the common good that follo ws the regulative ideal of consensus (Cohen 1989; Cf. Habermas 1996). Interests may still be irreconcilable. However, deliberation legitimates decisions by allowing everyone a chance to contribute to the process that produces them. This is considered an im provement over the aggregative model. Thus, if we think it is the case
21 (Estlund 1997) One possible answer is that deliberation adds epistemic value to democratic decision making; that is, rather than producing fairer decisions, it produces better ones (Bohman 2006, 2009, Cohen 1989, Estlund 1997, 2008, Landemore 2013, Mart 2006, Nelson 2008, Ober 2010) The emerging field of epistemic democracy, which I review extensively in Chapter 2, studies the ways in which democratic procedures produce desirable outcomes by pooling together the knowledge, insights, wisdom, and justified beliefs of deliberators. On this instrumental account, democratic decision making processes are valued at least in part for their knowledge producing potential and defended in relation to this. Epistemic democracy but shi fts their focus toward an outcome oriented consideration (Landemore 2013: 44). legitimacy, epistemic democrats are thus primarily concerned not with inclusion or participation as the legitimating factor, but instead with the ways in which group deliberation can produce collective wisdom. Rather than being considered intrinsically valuable, deliberation is defended to the extent that it produces epistemic benefits. In this projec t, I similarly focus on defending the effects and outcomes produced by deliberative participation. Thus, as we will see in Chapter 6, I call for extensive, inclusive, and participatory deliberation, but only as one element of a broader overall decision mak a largely theoretical angle. I argue that cognitive diversity is the deciding factor in deciding how large, diverse, and participatory the pool of problem solving deliberators shoul d be. The number of participants should be determined by the ability for the group
22 to find the right or best solution to the problem it is trying to solve. This means that we should increase the number of participants until doing so no longer improves the cognitive resources available and involved My proposal for epistemic deliberation is primarily targeted to situations where the academy is confronting wicked problems that pose an existential threat. While Chapter 2 takes up the topic more fully, it is sufficient for now to define a wicked problem as one that is incredibly difficult, if not impossible, to solve. This can be the case for many reasons: perhaps the solution is unknowable or impossible to achieve, or perhaps the problem itself changes or is so multi faceted that efforts to resolve one aspect undo efforts to resolve another. An example in academe could be the confluence of diminishing budgets, increasing global competition, crisis level student debt, and the near ubiquity of free information ( Barber et al 2013). I argue that when addressing problems of this magnitude, deliberation among cognitively diverse interlocutors is the best way not just to find solutions, but also to accurately identify problems and their likely consequences in the firs t place. Past attempts to solve wicked problems like these have been insufficiently inclusive and thus operated on an inaccurate understanding of the problem itself. For e in the 1980s, 6 a common response to these increasing pressures has been the tenure track, adjunct, temporary, or contract based) teaching faculty. Because this type of faculty position typically comes without healt h 6 See Chapter 6.
23 care benefits or retirement plans and often pays subsistence level wages, it was seen as a cost saving solution. However, hiring contingent faculty does serious epistemic harm to the university if the position is not likewise incorporated into academic policymaking. For example, participation in shared governance is often reserved for tenured or tenure track faculty members. However, they are responsible, on average, for less than 30% of all teaching hours in the US. Thus, there is a growing disconnect b etween the people who spend the most time in the classroom and the elite faculty who create academic and educational policy. Additionally, relying on subsistence level employment introduces problems outside of the classroom, which nonetheless affect the qu ality of teaching. Cases like this show that solving wicked problems requires taking full stock of both their causes and of the effects that proposed solutions will have on the institution. To do this, schools must be willing and able to tap the cognitive diversity and resources of all its stakeholders. 7 Limits to the Argument more ; there are also limits to the usefulness of deliberative approaches. My argument supposes that deliberative practices should be used to the extent that they provide epistemic benefits. When problems and solutions a re collectively identified i.e. once a group knows what to do experts often prove more effective at getting things done. So, effective shared governance must consist of both a deliberative democratic body that can figure out what 7 See Chapter 4 for a case study from Macalester College in which a faculty senate comprised of all types of faculty reached out to the student body and worked with them as equal partners in identifying and n Studies program.
24 to do and an expertocratic body that can then get things do ne with the nimbleness and adaptivity that current critics find The second co ad hoc bodies formed temporarily to execute specific policies. Chapter 6 of my project concludes by fusing the theoretical defense of epistemic democracy with the real world observations of inefficient shared governance practices. It will offer both structural and cultural suggestions. Structurally, my solution is a sequential, mixed decision making regime: problem solving should be done by a diverse and deliberative body, but putting those decisions and policies into action should be left to elites and campus leaders, who can act quickly and adaptively. It allows for both campus wide inclusion and quick expert action. Epistemological Assumptions and Premises Though he was no democrat, Aristotle offers the philosophical groundwork for the idea of collective intelligence. In the Politics he says: For the many, of whom each individual is not a good man, when they meet together may be better than the few good, if regarded not individually but collectively, just as a feast to which many contribute is better than a dinner provided out of a single purse For each individual among the many has a share of excellence and practical wisdom, and when they meet together, just as they become in a manner one man, who has many feet, and hands, and senses, so too with regard to their character and thought (1996: 76 ). There are two epistemological assumptions embedded in this passage. This project shares in these assumptions. First, knowledge is widely distributed across the population. The coordination of this knowledge is a necessary component of a democratic miss ion. But technological innovation brings with it the hyper specialization of occupations, which makes us all more interconnected and interdependent on others.
25 Integrated problems require integrated solutions. Thus, when a problem arises, it requires the co ordinated effort and information of many diverse people who can address Fuerstein 200 8, Goldman 1999, Kitcher 1990). Because knowledge, especially on campus, is hyper specific, this means that problems are often best solved by diverse groups of thinkers and problem solvers. In Chapter 3, I offer the example of massive open online courses ( MOOCs) and both the promises and perils that they offer to schools facing budget crises. Second, talking together is the best way to locate and coordinate this knowledge. This means we cannot ignore the role deliberation plays in the study of institutiona l epistemology. This project relies on diversity theory to explain how and why collective intelligence is an emergent property of deliberation and collective problem solving. For reasons I develop more fully in Chapter 3, diverse groups of problem solvers yield smart decisions when they approach one problem from many different perspectives and apply many different cognitive tools to find one solution. By talking together, we pool our beliefs about the best possible solution, starting with something and buil ding on it continually until something better is achieved. Collective intelligence is thus an knowledge. My goal in this project is to challenge the expertocratic claims t hat shared governance cannot work because groups are by nature 8 not good decision making units 8
26 or inferior to skilled experts Therefore, to prove successful, I must only show that it is possible for groups to collectively locate the optimal solutions to e pistemic problems. Furthermore, the epistemic deliberative model I present here would only have to outperform the expertocratic model. It would not need to prove right every time. Indeed to hold any decision making procedure to this standard would be unrea listic. We do not require that citizens always choose the best candidate, that doctors always make the right diagnosis, or that parents always provide the best example for their children, even though we hold these as ideals or standards of judgment. s more, this epistemic deliberative model only has to prove better in problem solving venues. I follow Page in suggesting the requirement be the presence of irreversible action 9 (2007: 5 6). An irreversible action could be an everyday activity like mowing a lawn, giving a lecture, or grilling a steak. It could also involve path dependent actions like drafting and enacting policy or enforcing a rule or law. Irreversible actions are best performed by skilled individuals. However, when planning that action (th e most efficient mowing route, the most concise word choice, the best steak preparation and grill temperature, or the most effective and appealing policy) the evaluation of reasons for and against a proposition (i.e. deliberation) finds a necessary and eff ective home. In other words, I do not claim every decision should be subject to group deliberation or that every task should be a deliberative one. This means large groups are necessary only some of the time. In doing so, I purposely exclude situations in which expert action is required or where one or a few individuals are more likely than a large group to make a good decision. We can assume, e.g., that the deans and 9
27 without asking ten biologists for their input. Regarding expertise, we should not assume that those same biologists would perform say, departmental audits with the same expertise as ten accountants (or one, for that matter). Conversely, those deans, depar tment heads, and auditors really have no place writing syllabi for introductory biology courses. In these cases, no problem is so difficult that one or a few experts cannot solve it Rather, I am concerned here with questions that are so difficult and comp lex that they transcend one realm of expertise, such as whether a school can simultaneously lower the cost of attendance, compete at a global level, satisfy stakeholder expectations, adapt to changing technological landscapes, continuously produce valuable knowledge, and maintain the tenets of academic freedom. I contend that in these cases a larger rather than smaller group is the wiser choice Background and Context This project is meant to show how the study of democratic theory can be applied to variou s real world problems. In doing so, this project is itself part of a larger, ongoing (Bohman 2007, Goodin 2005) about the meaning and value of democratic decision making. It differs from, and yet maps onto, much o f the prominent work done thus far on epistemic democracy. The following analysis is necessary because there are few explicit discussions of the standards and conditions that must be met in everyday practice to achieve epistemic value. 10 Here I 10 To be clear, this is not intended as a shortcoming in the literature, but rather as a sign of its nascency. A goal for this project is to suggest the philosophical literature on epistemic democracy has been developed into a solid enough foundation that deliberative democrats should start thinking practically about how to bring it into everyday use.
28 will briefly outline some of them. 11 In perhaps the most significant work on epistemic democracy, Estlund (2008) include some m inimal epistemic level; that is, democracies are legitimate to the extent that they produce, at the very least, better decisions than a coin flip. The quality of our decisions should therefore be better than if we were to make decisions at random. While ot her regimes might make even better decisions than democracies, we have procedural reasons for rejecting them. He thus argues for the most epistemically valuable democracy that the public is willing to accept on procedural grounds, theoretically striking an optimal balance between good and fair decisions. 12 While epistemic room in real world democracy. It provides a basis for building an understanding of the epistemic potential of the deliberative process. In perhaps the most recent significant work on epistemic democracy, Landemore deliberative and aggregative decision procedures. She fin ds that large scale, inclusive deliberation makes for smart crowds that make smart decisions, while aggregative procedures like majority rule voting can prove an effective way to capture these decisions. She argues convincingly that the search for collecti ve intelligence finds both procedures complementary. From her work we learn that deliberation is a venue for 11 F or an extended overview, see Chapter 2. 12 Alternatively, Peter (2013) argues that both procedural and epistemic considerations are important and deserve their place in our decisions, but that we should be willing to make epistemic compromises in the name o (See also Peter 2009)
29 producing epistemic value and that aggregation can efficiently collect and harness smart decisions. Both deliberative and aggregative readers thus have good reason to put epistemic value at the forefront of democratic theory. However, her thesis is largely formal/theoretical. It sets out to show how both procedures can function together and in harmony and necessarily stops short of describing what de liberation should look like in actual practice. Similarly, Ober (2009, Cf. 2010, 2013) aggregation and deliberation, in which groups collect and process relevant information before ultimately deciding how to act through a voting procedure. He investigates, e.g., how a making groups, each of which was a diverse network of tribes. These smaller groups produced epistemic value during their deliberation. Relev ant experts would present information and everyone would get the chance to weigh it, accepting what seemed right until the group was able to make a collective decision. By breaking the massive council into small diverse groups, each was able to present one voice during deliberation, thus reducing what could be a cacophony of voices to a public conversation between ten inclusion, and democratic norms to produce what Ober finds to be epistemically optimal decisions. This dissertation considers this philosophical framework, formal model, and historical model its points of departure. It contributes to this body of literature by focusing on the specifics of the deliberative process While much of this work is project might better be described
30 itself can provide for good decision making. This is consistent with the trend in the literature (Chambers 2009, Mansbridge 2007a) which has arguably moved away from (Dahl 1989, Fung 2006, Smith and Ryan 2014) and decision making processes (Ackerma n and Fishkin 2004, Fishkin 1991, Fung 2003, Fung and Wrig ht 2003, Warren and Pearse 2008 ) There is also an underlying normative component that motivates this project: Governance practices reflect what we think colleges and universities should do. If we think they should merely confer credentials then it makes sense to run them like businesses by empowering a decision making elite keen to the science of efficiency and c ost effectiveness. In this case, a relatively narrow range of expertise would be sufficient to solve all problems an institution faces or might face, and we could rest assured that economic principles properly applied will always keep academe afloat and on mission. If, however, we think higher education should be a citizen building enclave (which is the role it has historically played, and which no other social institution is tasked with), then we have an additional reason to embrace deliberative and democr atic governance structures ; viz that rely on participatory world making. In this case, shared governance is important not just for the results that it may produce and for the epistemic value those results may yield, but also for the procedures through whi ch those decisions are made. In this sense, higher education governance speaks volumes about what kind of political cultural, and social values we hold more generally. For these reasons, this dissertation also speaks to students in the emerging field of c ivic studies, and attempts to show that institutional epistemological concerns can
31 play a meaningful role in renewing civic and democratic values. However, the epistemic component that defines this project is unique in this literature. To better understand the contribution my dissertation stands to make in this field, it is necessary offer some comments on the broader push to democratize academe. Civic Renewal and Education Reform t four essential components of citizenship and civic engagement: facts, values, actions, ant, c facts, values, and actions should culminate in a strategy for achieving realistic goals that outcomes and (2013: 25). This question is one that higher education has traditionally been concerned with answering. As many scholars lament, however, in the last few de cades the institution writ large has undergone a shift from creating thoughtful citizens to training workforces (Bok 2003, Brown 2015, Giroux 2004, Nussbaum 2010, Schram 2014, Schrecker 2010, Shumar 1997) In the wake of World War II, colleges and universi ties shifted their curricula away from citizen building and instead began teaching the technical skills that led to high paying jobs and secure, middle class lifestyles. Consequently, the civic and
32 democratic mission of schools suffered, as the institution became seen as a credential role within it. While voices speaking against this trend could always be heard, they have recently reached a fever pitch. We can look to thr ee major events as evidence. In 1999, an influential report called the for a (Boyte and Hollander 1999) it reads, research institutions are being challenged to justi fy what they do and how under review, spurred by calls for accountability, efficiency, and utility as well as by questions about the theories of knowledge embedded in prevailin g rewar d and evaluation systems (1999: 7). Its authors claim the corporatization of academe was a turning point in its civic history. In response, many schools were forced to prioritize profitability over the pursuit of knowledge, for instance by offering course s in responses to market and industry demands rather than according to the needs of a complete education. As a result, training, and American universities, by and large, los t their civic mission (1999: 7). The Wingspr ead Declaration therefore called for a restoration of this mission and served as an early catalyst in the contemporary movement to democratize higher education. Recently the American Association of Colleges and Universities convened a task force to develop a wholly new approach to renewing this mission. The result, a report US democracy and a viable, just global community requi re informed, engaged, open minded, and socially responsible people committed to the common good and practiced
33 Engagement 2012: 13 14). L ike the Wingspread Declarat ion, it explicitl y ties democratic education to civic renewal. In doing so the report seems to counter the kind of thinking that diminished the civic mission in the first place: though the value of a democratic education cannot be measured, both society and the economy suf fer without it In doing engagement movement. This recent push for civic engagement arguably came to a head in 2014, when the White House and the US Department of Education o rganized the Summit on Civic Learning and National Service at Tufts University Tisch College 2014: 2). For the first time it seems this movement has reached a critical mass and that real, substantive change is possible in higher education. Finding Reason to Democratize In researching the democratization of higher education, I found that democracy is defended on the grounds that one might expect: that it imbues civic values like respect for difference, equality, fairness, or reciprocity that make for good citizens. Further, I found these efforts fell into one of three basic categories: (i) those that broach the topic of campus/community relations at the institut ional level (call this community and (iii) those that attempt to model democracy within academic institutions. Community engagement involves efforts to engage schools an d their local communities in mutual problem solving (Benson, Harkavy, and Hartley 2005, Boyte and Fretz 2011, Bringle, Games, and Malloy 1999, Checkoway 2001, Colby et al 2010, Dzur 2008, 2010, Mathews 2005, Van de Ven 2008) Without a democratic focus,
34 ngagement efforts are often pursued as ends in themselves, and engagement becomes reduced to a public relations function of making known what the campus is (Saltmarsh and Hartley 2011: 18) Absent this focus, [univer approach to public problems is predominantly shaped by specialized This elite model assu mes a basic distinction between knowledge producers and knowledge consumers (2011: 20) that undermines the idea of mutual co creation and problem solving. Democratic community engagement would have schools interact with the local (Bringle, Games, and Malloy 1999), meaning they approach problem democratization of knowledge: academ icians are facilitators of expertise who must combine it with local knowledge to craft optimal solutions. The claim, however, is not that knowledge is more accurate when it is crowdsourced and pooled but rather that an epistemological pluralism is fairer a nd leads to more commitment by schools if leaders see themselves as part of the community. The second category involves educating for democracy. This venue has two aspects, what I call showing and telling. Showing Schools show students what democracy means through place based or serving learning: directly placing students into the community to address and help solve problems (Barber and Battistoni 1993, Battistoni 1997, Butin 2010, Hunter and Brisbin Jr 2000, Kezar Chambers, and Burkhardt 2005, Saltmarsh, Janke, and Clayton
35 2015) This has a democratizing effect similar to the one above, but whereas the first on student interactio n. Efforts to instill democratic values will emphasize the experimental 90) find that service and funct Telling Concurrently, democratic values must be explained and taught in the classroom (Ben Porath 2013, Benson, Harkavy, and Puckett 2007, Boyte 2008, Generation Citizen 2015, Hess 200 9, Mutz 2006, National Task Force on Civic Learning and Democratic Engagement 2012, Thomas and Levine 2011, Westheimer and Kahne 2004). While service learning is a powerful pedagogical tool and shows students how to apply democratic values, students must s till be taught how to recognize and appreciate of courageous conversation and action, where the most pressing social, economic, and political needs of the nation and world are identified, studied, debated, and where democracy allows students to recognize and th us appreciate the co creation they are participating in during their place based service efforts. The third and final category involves the ways in which the academy can be made to reflect the democratic principles on which it is based (Boyte 2015, Guinier 2015, Mallory 2010, Mathews 2015) These largely involve efforts like affirmative action
36 that attempt to make the student, faculty, or staff bodies more equitable. They can also take the form of community involvement in things like mission statements. I find that that most, if not all, of this literature relies on the normative defense of democratic participation: it is better to create institutions that reflect the cares, concerns, and commitments of the broader community. However, I did not find any cla im that democratic processes and procedures lead to smart(er), better, wiser, or more accurate decisions. There is thus a gap in the literature: there are few if any arguments that deliberative and/or democratic decision mechanisms are preferable because t hey optimize results. My goal in this dissertation is to complement these procedural defenses by arguing for the implementation of democratic principles and practices on these epistemic grounds. 13 Since doing so concerns the quality of decision making, I ne mechanisms, which is what led me to make shared governance the object of study. In this regard, this dissertation finds a new reason to democratize that supplements the traditionally normative discussions of d emocratic civic engagement. It considers what is actually possible, what is normatively praiseworthy, what requires direct and collective action, and that culminates in a conceptually coherent and practical strategy for action. In doing so, it complements and acts in concert with current attempts through which democracy and democratic action c an enrich higher education. The claim I defend here, that wicked problems are best addressed by diverse groups and that 13 For a fuller discussion of procedural and epistemic defenses of democracy, see Chapter 2.
37 institutions can be designed to promote such results, is relatively small. But this solution is designed to extend meaningful participat ion in governance to faculties in ways that both improve campus culture and the quality of institutional decision making. Chapter Summaries The remainder of this dissertation is comprised of two parts. Chapters 2 and 3 are diagnostic. They evaluate and at tempt to identify what is wrong with shared governance and how we should proceed given these concerns. Chapter 2 explores what shared governance is and why it is considered an outdated decision mechanism. After reviewing the critical literature, I offer a new explanation from the perspective of democratic theory that takes into account the logic of legitimacy underlying the philosophy of shared governance. In contrast to popular critiques, I find the problem is not with the collective nature of decision mak ing. Rather, it is that shared governance (i) frames campus wide problems as disputes of interests among faculty, administration, and trustees, and (ii) considers fairness the standard of good decision making, but (iii) fails to consider epistemic consider ations like expertise, knowledge, justified belief, or wisdom. I agree that it is outdated and in need of change, but not for the reasons that many think. I conclude that governance practices should adopt a standard of accurate decision making rather than the normative standard of fairness. Chapter 3 shows that large, diverse groups are better able to meet this standard than one or a few experts. It shows that the value of the expertocratic argument is showing that the problems facing academe are epistemic in nature and thus require appropriate solutions. However, I demonstrate that while accurate in its diagnosis, the argument fails on its own merits to provide a good prescription. Diversity theory proves
38 that diverse groups are ideal for solving problems under the conditions in which academic decision making takes place. Chapters 4, 5, and 6 comprise the prescriptive second half of the dissertation. Through them, I propose a new theory of deliberation designed to help solve the inefficiencies and ineffecti veness of current governance procedures. In doing so, these chapters are relatively removed from the empirical, on the ground observations of the chapters in the first part. They feature detailed analyses of collective decision mechanisms like deliberation In Chapter 4 I develop a theoretical model of deliberation based on the epistemic standard. My task is to find what specific qualities about the deliberative process makes collective wisdom greater than t he sum of its individual parts, and I argue the co To do this, I look to the fields of deliberative democratic theory, cognitive and social psychology, decision making, and communication science Drawing on work i n these fields, I find that the deliberative problem solving process provides epistemic value if it consists of reason giving and decision oriented argumentation by a sufficient number of diverse but relevant deliberators. If Chapter 4 concerns how to tal k, Chapter 5 concerns when to stop. Epistemic democrats believe that unanimous agreement offers an epistemological advantage over other stopping points like majority rule. In reality, however, unanimity is very rare. So rare, in fact, that we should be wil ling to explore any other option that might viably replace it. Here I draw a distinction between my argument and those of other diversity theorists who defend unanimity on epistemic grounds. I claim that if we can devise a
39 deliberative process that tends t o produce the right or best answers, solutions, decisions, or choices (i.e. if my epistemic model holds water), then a dissenting minority understand why or disagree that it is the best. It thus substitutes majority rule for unanimity without losing an y making in shared governance without sacr ificing such value. To defend the claim that diverse groups produce wise decisions or problem solve well is not to say that all policy making should be the province of a large legislative body arate from the epistemic value that groups bring to the decision making table. Chapter 6 offers concluding remarks about the limits of my epistemic model within the broader confines of shared governance practices, as well as some suggestions for turning th is theoretical model into concrete reforms and best practices. I suggest that this model is beneficial to the extent it offers epistemic benefits. But better deliberation does not simply mean more. Thus I offer guidelines for when it is and is not appropri ate.
40 CHAPTER 2 FROM PROCESS TO OUTCOME: THE DISCONTENTS OF SHARED ACADEMIC GOVERNANCE Introduction What is the doctrine of shared academic governance, why is it considered a failed or failing decision making mechanism, and what accounts for these failures? Shared governance is traditionally understood as a collective and inclusive approach to decision making that considers all relevant parties on campus. While the concept still enjoys paradigmatic status throughout American higher education, it is widely criticized by faculty members, administrators, and trustees alike for failing in practice to deliver its theoretical promises. An increasingly popular explanation for these failures is that complex, and uncertain (i.e. wicked) problems that cannot be adequately addressed through collective means or inclusive governance bodies but instead require quick, decisive, executive action. I argue instead that shared governance fails not because it r elies on collective decision making but because governance procedures and practices often address procedural concerns for fairness at the expense of epistemic concerns for accuracy. In other words, they are trying to solve the wrong kinds of problems. Thi s chapter proceeds as follows: First, I review and analyze shared governance and its failures. Next, I draw on theories of democratic legitimacy to develop a conceptual critique of shared governance practices, arguing that they (i) frame campus wide proble ms as disputes of interests among faculty, administration, and trustees; (ii) consider fairness the standard of good decision making; but (iii) do not account for the demands placed on them by concrete, real world problems. I conclude that while collective decision making may offer inefficiencies, the ultimate factor determining the
41 future success or failure of shared governance is whether or not its procedures can help solve real problems. Governance practices should adopt a standard of accurate decision m aking rather than the normative standard of fairness. Shared Governance: Its Meanings, Complexities, and Failures administration must be empowered to execute as necessary the rules and actions that will enforce that mission. Its faculty requires the academic freedom to create and disseminate knowledge. Shared governance is the collection of structures and processes a school implements to balance these forms of authority (Birnbaum 2004: 5). But shared governance is also an idea. At its core is the intuition that fairness, inclusion, and collaboration are necessary components of good decision making. Since the 1960s, co llege and university faculties have played an increasing role in the running of their schools, in addition to their traditional role in determining academic matters. In 1966, the American Association of University Professors (AAUP) the Ame rican Council on Education (ACE) and the Association of Governing Boa rds (AGB) published the foundational Statement on Government of Colleges and Universities (often called the Joint Statement ), which has set the standard for shared academic governance ever since. 1 The J oint Statement 1 For an historical account, see Gerber 2015.
42 T he variety and complexity of the tasks performed by institutions of higher education produce an inescapable interdependence among governing board, administration, faculty, students, and others. [This] relationship calls for adequate communication among the se components, and full opportunity for appropriate joint planning and effort (AAUP 2015). Joint effort requires that members of the institution who are affected by major decisions should be involved, in some way, in the making of those decisions. Since i ts publication, the idea of inclusive or shared governance has enjoyed paradigmatic status within academe. Hence, shared governance, as I will use the term here, refers to the collective decision mechanisms through which faculty, administration, and truste es (herein referred to as governance units) are involved in making choices that affect them. Joint Statement has been highly influential and has served as a de facto National Center for Education Statistics (NCES 2015), there are over 4,700 degree granting, two and four year colleges and universities in the United States alone Of these, all but six institute some form of shared governance (AAUP 2016). The Joint Statement has also been influential for distinguishing two types of sharing. Call these types inclusive and divided 2 Inclusive responsibilities are those that e at one time or another the initiating capacity and decision making participation what Dzur (2008: joint, mutual, or collaborative decision making that g ives everyone a share in governing. Faculty (usually through a representative body like a faculty senate) administration, trustees, and occasionally students or the church are 2 The wo
43 considered co creators of the campus. The Joint Statement specifically names pl anning, decisions regarding physical resources like the use of buildings or facilities, budgeting and resource allocation, hiring and firing of presidents, and faculty salary decisions as the things that should be subject to inclusive decision making. Beca use power is shared, we can say inclusive governance relies at least theoretically, on the search for agreement or consensus among governance units. This stands in contrast to divided governance. In a divided schema, roles and responsibilities are di vvye d up according to expertise. Whereas inclusion involves everyone making all decisions, divided governance requires discrete groups each with their own decisions to make. Governing boards invest in the financial and philosophical future of the school, inclu vision to execute the academic mission, and establish an administrative body that uses human and capital res ources to this end. Faculty responsibilities include determining curricula (e.g. subject matter and instructional methods), setting research agendas, decisions regarding faculty status (e.g. appointments, tenure and promotion, selection of department chair s), and other educational aspects of student life. In a divided schema, units have exclusive spheres of authority that they do not share, thus eliminating the need for inter unit agreement or consensus. Communication between governance units is largely adv isory or consultative. When units do work together, it is often information, rather than decision making power, that is shared. In addition, the Joint Statement also allows that voices should be included to differing degrees, depending on the matter at hand. In simple terms, this means for any
44 given decision a governance unit can have all, some, or no authority. For instance, at the University of Florida governan ce units use the language of determining, recommend ing (i.e. partially determining ), or consult ing on policy (UF 2016) A unit determines policy when it has full decision making authority. In contrast, recommending policy means having some amount of say in the process but not full determinative authority. Consultation involves a unit having the right to be heard by those with determining power and usually to stay abreast of their decisions, but not deciding power. In many cases, determination and consultation are complementary roles. For instance, if faculty craft academic policy, administrators and trustees will often have the right to voice their opinions. In contrast, recommending policy means that all parties work together in jointly crafting policy. From Inclusion to Division In recent decades, American higher education has experienced a trend in favor of divided governance. In simple ter ms, this means increasingly more campus wide decisions are being made by fewer people. Explanations differ as to why this shift toward divided governance is occurring. Many see it as a consequence of the trend starting in the 1980s to run colleges and univ s ( Birnbaum 1988, 2000, Bok 2003, Brown 2015, Gerber 2014, Ginsberg 2011, Giroux 2002, Guinier 2015, Nussbaum 2010, Schram 2014, Schrecker 2010, Shumar 1997) 3 Schram (2014) points t 3 This is not the same as running a for profit university, such as the Minerva Project or Apollo Group Inc, which advertises itself as the University of Phoenix. See Wood (2014)
45 at risk of losing out in the global economic competition if it did not reform education so as to graduate more more streamlined educational programs that reduced costs while focusing solely on marketable degrees. But it also required that schools be run more like businesses, meaning that cost decision making. Accordingly, an historical shift in the com position of academic administration In the past, most administrative positions were filled by faculty, and usually on a temporary basis. But now administrators are increasingl y hired from the business world, put into permanent positions, and see their school as a business to be run. The result is a body (Gerber 2014: 2) and see themselves as the only experts on campus capable of achieving success. This new generation shares authority only as a matter of legality or convenience: they delegate to faculty the relatively insignificant decisions that have limited importance and /or that t hey would rather not deal with 4 Data suggest this explanation is not merely the grumbli ngs of jaded or cynical faculty. A study of shared governance at research universities during the years 1992 to 2007 finds that faculty authority declined noticeably at the college and institutional 4 Barden and actively discourage) peers and charges to develop the skills, knowledg
46 promotion/tenure their historical sphere of influence (Finkelstein, Ju, and Cummings 2011: 219). While retaining the hold on traditional faculty concerns, shared responsibilities have ance matters have actually remained quite stable over this period, as faculty and administration have strengthened the hold on their position for themselves in between the se units that has absorbed what authority each 5 These positions enforce the managerial principles of supervision and attention to the bottom line at the level of day to day departmental activities without the nee d for heavy handed administrative intervention. Some argue the antagonistic faculty/administration relationship that results further contributes to poor decision making (Birnbaum 2004, Johnson and Pajares 1996, Kaplan 2004a, Kezar 2004, Tierney 1988, Tiern ey 2006, Yankelovich 1999) For instance, Birnbaum (2004: 15) claims that procedural concerns like trust and informal requirements like social capital explain effective governance: Governing alone is like bowling alone regardless of how wise or virtuous t he person or office that determines what the rules should be, social capital within the group may not be increased and, in fact, is likely to be diminished as it transforms a process of reciprocal social interaction in an exercise in unilateral decision ma king. It is the process of shared governance, and not its outcome, that helps to build the dense network of connections that create social capital. 5 For instance, the number of respondents indicating that deans and department chairs were influential or very influential in determining budget priorities increased from 9.2% in 1992 to 42.4% in 2007 (Finkelstein, Ju, and Cummings 2011: 208).
47 Without informal qualities like trust (Pope 2004, Stensaker and Vab 2013, Thoenig and Paradeise 2014) mea ningful deliberation (Greenhalgh 2015, Mallory 2010) or personal relationships with colleagues and administrators (Kezar 2004) the practices of shared governance are seen as empty or symbolic gestures, free of real meaning or efficacy. This echoes studie s of governance more generally, which routinely find that shared meaning (Ansell and Gash 2007) and dialogue (Stains 2014) are crucial components for successful governance. The cultural critique suggests the decline in participation and trust in the shared governance system by all parties may account for its decreasing effectiveness. Shifting Landscapes and Wicked Problems: Explaining the Divided Governance Trend There are, however, other explanations that involve the difficulty associated with consensus b uilding in a rapidly changing environment. In this section I offer some background context often used to justify the model of governance I call the bringing more diffi culty and complexity to higher education (Bahls 2014, Barber et al 2013, Darden 2009, Hirsch and Weber 2001, Kezar and Eckel 2004, Tan 2013, Weber and Duderstadt 2004, 2013) 6 In recent years, changes and shifts in the academic landscape have introduced pr oblems that shared governance is not equipped to handle. In a recent article, Kezar and Eckel identify three specific changes that make traditional forms of shared governance problematic: 6 By complexity, I mean that a problem and/or its attendant solution are comprised of many interconnected elements of components. Difficulty refers to the lack of ease with which a solution is found. For the sake of argument, I consider difficu lty a function of complexity such that as complexity increases, so does difficulty.
48 (1) the need to respond to diverse environmental issues, such as ac countability and competition; (2) weak mechanisms for faculty higher education place great responsibility and strain on institutional leaders to make wise decisions in a timely manner (2004: 371). governance was not designed or intended to solve. These changes generally mean there is less ti me to research, contemplate, and deliberate over decisions in committees or open forums. Consequently, a fundamental shift has been occurring in shared governance, such that responsibilities are being decided increasingly through divided, rather than inclu sive, means. These shifts in the academic landscape have created what Rittel and Webber (1973) solving traditionally consists of two steps: identif ying the problem and then designing a solution (Buchanan 1992: 15) Wicked pr oblems, however, elude this basic formula because either the problem cannot be fully identified or understood, the problem itself changes in nature before it can be solved, or there simply is no solution. However, [w] icked problems are in no way evil; they are simply wickedly difficult to developed the notion of wicked problems problems that their engineering education well prepared them to tackle. Though often very complicated, tame problems such as building a suspension bridge or splitting the atom are solvable by experts who have sufficient knowledge and technical skills. Wicked prob lems, in contrast, cannot be solved by data analysis or even well address ed within a narrow range of expertise. They demand systems level thinking and attention to context owing to the dynamic complexity and particularities that define them. Multilayered social, political, ec onomic or ethical tensions pre clude answers supplied by algorithms, logic, standard operating procedures or simple rules and principles. Indeed, differences in values and perspectives among stakeholders typically preclude an uncontested definition of the prob lem itse lf (Thiele and Young 2016: 35 ).
49 Accordingl y, wicked problems are never really solved But they can be res olved. 7 Put another way, only suboptimal solutions or next best heuristics can be located. For instance, curing cancer may prove wicked if it is the case that no solution exists or that we neve r find one. Or consider the idea of sustainable economic development, which emphasizes increasing economic power without harming the environment in the ways that the US and Western Europe did during the Industrial Revolution. These twin goals often seem at odds with each other, such that advancements in one direction hurt efforts to advance the other. It is thus possible that the only way to raise the developing world out of poverty is through environmental destruction, which itself brings new wicked proble ms. Because they require systems level thinking that involves lots of complex interactive parts, there is little chance that resolutions will be created without bringing together teams of cognitively distinct of people who can simultaneously address the sy nergistic and interactive nature of these problems. While the problems facing academe may not prove as wicked as curing cancer or developing the world they nonetheless contain elements of wickedness 8 that likewise make their resolution extremely difficul they believe will prove wickedly difficult to solve. First, technology has connected and shrunk the global economy such that h igher education institutions now compete on a global market (2013: 9). The internet has transformed the application process and it is now common for high school students to shop around at the international level for the 7 8
50 best quality education they can fin d, just as they might do for material goods or information. However, along with globalization comes increasing inequality, such that the pool of eligible college students is shrinking (2013: 11). Hence the market is becoming both global in scope but finite in number. Further, the price of college is supply is exceeding demand. Finally, the internet has made information free and ubiquitous (2013: 16), meaning there is less need to go to attend a college to learn marketable skills. What was formerly hidden behind the paywall of college tuition is now c heaply, if not freely, accessible on websites such as Coursera, edX, or even Youtube. Accordingly, the technology centric and technology adjacent fields currently in demand tend to care more about what skill sets their employees possess and less about wher e they learned them. perennial problems like budgetary constraints and social pressures. Schools are always expected to produce large quantities of cost effective, high quality degrees But today, and demands to engage the community, business, and industry; to solve social problems and improve the schools; to generate cutting edge research and innov ations to fuel the economy; to develop a more just and equal society by preparing a diverse student body, while having fewer funds, more demands from students, and an chang es mean that colleges and universities are fighting a new and difficult battle.
51 There is increasing and international competition to attract a dwindling number of tuition dollars. At the same time, the decreasing value of degrees and the ubiquity of inform ation might just mean decreasing demand for a traditional college education. Schools must figure out how to make their degrees valuable, affordable, internationally recognized, and worth the cost of attendance in a world of increasing labor surpluses and f ree information. To say the least, university leaders are in an unenviable position. Reactions to Shifting Landscapes In response to these elements of wickedness, many campus leaders have chosen to simply sidestep shared governance practices on the ground s that representative bodies like the faculty senate are no longer able to keep up with the new demands of academic decision making (Duderstadt 2004, 2009, Gerber 2014, Ginsberg 2011, Fish 2007, Kaplan 2004b, Keller 2004, Kezar and Eckel 2004) The sharing of historically administrative authority they claim, has produced overall decline s in campus success because large groups of academics have authority without accountability for their actions or decisions A recent report by the AACU summarizes the situat ion this way: It is generally agreed that the faculty, those with expert authority, should be the ones to make academic decisions rather than administrators or trustees, who have bureaucratic authority. Yet while the faculty are generally responsible for academic decisions, they are seldom held accountable either for student learning or for the fiscal results of their decisions. And while administrators are held accountable for student learning by accrediting agencies, they have no legitimate authority to intervene in the academic programs that are designed to produce student learning; while they are responsible for financial prudence, they again have little authority decisions made by the fa culty (Gaff 2007: 6, his emphasi s). In the wake of the Joint Statement they argue, traditionally administrative responsibilities became subject to inclusive shared governance and thus became
52 otected by the principle of academic freedom and For instance, Duderstadt says the problem with shared governance is that faculty (200 4: 138). While these allow for the creation and dissemination of knowledge, they momentum, high risk, university wide decision environment (2004: 138). Del iberating in groups, inter alia takes too much time, faculty members are too apathetic and those who do participate usually act out of self interested motives freed Consequently, elite administrative lead ers often feel their only option is to subvert formal shared authority and act autonomously. hit, Yale found itself dealing with a $350 million deficit (Gideon 2012) In an eff ort to right the ship, then president Richard Levin and his executive cabinet were forced to quickly make many cost saving decisions that would normally fall under the purview of shared governance. During this time Yale faculty saw, among other things, a c onversion to a shared services business model, a freeze on new hires and faculty appointments, and years of across the board budget cuts. More recently, a plan was approved to open a new and lucrative international campus, despite being opposed by a majori ty of the
53 faculty (Rodrigues 2014) However, it is not always (or solely) admi nistrators who sidestep faculty. And neither is the skirting of standing rules or traditions always intentional. In some cases, structural factors prevent faculy from joining the conversation. Consider, for instance, the recent controversy at the Massachus etts Institute of Technology (MIT) concerning plans to develop commercial real estate on its campus Lik e many schools with significant endowments, MIT has a separate corporation whose sole purpose is to strategically plan for its future. This body, the MI T Investment Management Corporation (MITIMCo) answers to neither administration nor faculty members. Citing a burgeoning tech industry and a shortage of available real estate in the area surrounding campus, MITIMCo proposed building over 1,000,000 square f eet of commercial building space on unused campus property and selling long term leases (up to 60 years in some cases) to technology companies looking to set up shop in the area (King, Perry, and Salvucci 2014, Simha 2011) After learning about the plan, concerned faculty created an emergency task force to better understand the impacts such a proposal would have on campus. However, MIT has no faculty senate through which to lodge formal complaints against MITIMCo. Instead, [f]aculty meetings are chaired by the president, and often lack a quorum of 3 percent of the faculty. Important faculty business is handled through a set of standing committees, in essence joint faculty/administration committees that do periodically report back to the full faculty. Missing from the group of standing committees has been a campus planning committee reporting to the faculty (King, Perry, and Salvucci 2014).
54 Citing numerous concerns including the already critical shortage of graduate student housing, t collaborative campus culture, the lack of space remaining for future academic building 12), an renowned School of Architecture and Planning (which includes an urban planning department) or the Sloan School of Management criterion for pla case, structural factors prevented faculty concerns from becoming part of some crucial planning decisions that will determine the future of the campus. While there is no disempowerment yields the same results. According to leaders like Duderstadt and Levin the only way to keep their schools afloat was to go around the guidelines of shared governance and m ake important decisions with the quickness they required. In the case above there was no faculty involvement, which the urgent economic circumstances may or may not have justified. But even in cases where faculty senates or other shared governances bodies are involved we should not be surprised to hear that their position is still absent any meaningful authority. For instance, instead of working with a faculty sena te committee to write policy, administrative leaders might convene a task force to draft a pr oposal and then send it to a faculty senate committee to be voted on. T his vote might be decisive i n schools with relatively powerful senates, while in others a vote might be largely symbolic (Minor 2004) Either way, faculty are involved at the end of the policymaking pipeline.
55 This shift means that faculty authority is becoming increasingly concentrated in their own fields and departments, but less important at the college or university level. Such disempowerment further reduces the effectiveness of inclu sive practices by fully removing them from the problem solving process. Thus at the campus level one might are given more consideration than their mouths or brains. Hence we can see that the failures of shared governance reach far beyond inefficiency in the faculty senate. Elements of wickedness have become a routine component of day to day operations and the future success of educational institutions will depend on how they can respond and adapt to these conditions. But the strateg y pursued thus far by administrative elites has consequences that might prove just as harmful. Faculty are becoming less willing to participate in academic life at the campus level and administrators are more willing to disempower them in the name of effic iency stands to reason that if inclusive governance were more effective, it would alleviate some of the problems that lead to the shift toward divided governance and possibly repair the cultural damage or increase the social capital that discourages participation in members of all governance uni ts. One thing is certain, however: As governance units become increasingly isolated inclusive decision making becomes less and less frequent. Consequently, t he line between divided governance and consultative expertocracy is blurring. This raises questions as to whether or not divided governance has remained true to the spirit, and the technical guidelines, of shared governance. Divided Governance Through the Optic of Democr atic Theory As a student of democratic theory and collective decision making, I find
56 established argument in favor of limiting the authority of everyday people (Dahl 1989: 52 64). For instance, over two thousand years ago Plato (1945) argued that the demos is well meaning but ignorant and unable to muster, even collectively, the epistemic resources necessary to perceive of the good life and arrange the living world accordingly. The result is a technocracy, in which the wisest are empowered with decision making authority because they know best what is in the common interest. This kind of thinking sits at the basis of such a critique of shared governance. Thus even if logistical c oncerns like short time horizons are not a factor, it may be that we should put in charge only those people who know how to properly run an institution, just as we should not let administrators teach biology if they have never studied the subject. On the other hand, however, there are numerous examples of groups much larger than a faculty senate successfully working together to deliberate, problem solve, and write policies that deliver public goods. Models of participatory governance 9 are increasing in pop ularity as experiments in participatory budgeting (Fung 2006, Fung and Wright 2003) crowdsourced policy making (Warren and Pearse 2008), and deliberative polling (Ackerman and Fishkin 2004, Fishkin 1991, 1995, 2009) support the claim of the best solutions to difficult problems, even when compared to highly skilled problem solvers (Hong and Page 2001, 2004, 2012, Page 2007). This intelligence is maximized when groups are both deliberative and democratic (i.e. inclusive and egalitarian) (Landemore 2012, 2013, Landemore and Elster 2012) If there is any truth to this 9 For a fuller list of organizations dedic ated to public participation, see Fung 2015: 2.
57 research, then we can refute the claim that shared governance cannot work, even while acknowledging that it often does not. Why is it that higher education institutions, comprised of experts and knowledge creators, cannot manage to handle institutional concerns with the same ease that the (Fung 2 006, Fung and Wright 2003) ? My contention is that shared governance practices do not fail simply because of their collective, inclusive, or joint effort aspects. Rather, it has to do with the kinds of standards that these practices are held to. This sectio n leverages the literature on democratic theory to illustrate that the conception at the heart of shared governance (a procedural/internalist conception) measures decisions according to the standard of fairness. However, proceduralism cannot account for th e epistemic demands placed on it by wicked problems. Thus, critics are right to point out the inadequacies of shared governance but wrong to pinpoint its inclusive nature as the source of the problem. This suggests the failures of shared governance lie not in their inclusive or participatory nature, but in how these practices are used in specific settings Duderstadt argued that faculties run their departments well because their expertise and deliberative methods are ideal. But at the broader level these be come inefficient and ineffective. I want to argue he is partially correct but incorrectly assumes the same standards and goals inform decision making at each level Divided governance functions well in situations that require achieving a standard of accura cy or correctness in its decision making, which relies on epistemically warranted beliefs, expertise, and knowledge. Inclusive means are less effective not because they are inclusive or because inclusivity is inefficient, but because they resemble interest group
58 politics, assume a basic conflict of interests, and thus rely on meeting standards of fairness. We this shift in standards accompany the difference between handling departmental to campus wide matters. This difference is masked by the fact that a do uble move occurs: a shift in decision making body and a shift in standard. Duderstadt 10 fail to acknowledge this double move and suggest inclusivity itself is to blame. Thus, the key to improved shared governance is to hol d decisions to epistemic rather than fair standards. To justify this explanation I rely on theories of democratic legitimacy, or the justifications for collective decision making. A theory of democratic legitimacy, as the name presupposes, says that a poli tical decision (rule, policy, law, etc) ought to be considered legitimate when it is made through a democratic procedure (Benhabib 1996b, Buchanan 2002, Chappell 2011, Christiano 1996, 2004, Cohen 1997, Dryzek 2001, Estlund 1997, Manin 1987, Peter 2007) T heories of legitimacy vary and differ along multiple dimensions but all offer normative criteria for determining when a decision should be considered valid, binding, or deserving of an obligation to obey it. Thus, Peter (2007) says that just as justice is 11 10 should determine strictly defined and non overlapping spheres of decision making authority. In Chapter 3, I give a detailed look at the expertocracy as one kind of epistemic argument that relies on elite, rather than collective, decision making. 11 This collectively made attribute is sufficient for a decision to be considered legitimate. For this reason, we ca n apply these theories to cases of democratic decision making that fall outside of political democracies.
59 Fairness as Rightness: The Procedural Conception of Democratic Legitimacy Democracy can be defended on procedural and/or epis temic grounds. The former regards outcomes as legitimate to the extent they were made through a fair procedure. The latter defends the capacity for democracies to make the best or right decision. In both cases there is an implied standard of correctness th at underlies the sense of legitimacy. Procedural theories believe the procedure to provide that standard: so long as a decision is reached in a fair way, it is legitimate. Epistemic theories of legitimacy, however, argue there is a right, true, or at least best answer and so they subscribe to a content independent standard of correctness. Procedural or moral defenses rely on the normative preferability of collective decisions, usually because they are made through procedures that uphold values like equality fairness, mutual respect for individual rights, and autonomy ( Cohen 1989, Dahl 1989, Gutmann and Thompson 1996, 2004, Rawls 1993, 2001, 2002, Saffon and Urbinati 2013) Democratic decisions are thus considered legitimate if they were created by procedure s considered just by everyone obligated to obey them. Proceduralist approaches vary by the degree to which fairness is considered on making proceeds through public deliberation 335). Pure proceduralism is premised on the idea that democratic proceedings, whether aggregative or deliberative in nature, that take place within a milieu of disagreement should hold fairness as the ideal outcome. This idea has been particularly influential in the study of liberal democratic politics where moral disagreement over the basic structure of government is the backg round condition of political decision making. For
60 instance, a ccording to John a massive and often irreconcilable variety of values, beliefs, preferences, and conceptions of the good, which he call him to believe we have no way of judging the substance of a political decision absent from the procedures in which that decision is made. In fact, there is no external or independent standard at all for judging the outcomes of political decision making. Rawls ian procedu ral democracy (Raz 1990) : politics must refrain from appealing to any conception of the good whatsoever and should instead be constrained to only those arguments that are or could be the subject of public approval (Rawls 1 993: 133 172 ; 2001, 2002 ). Under these circumstances, a procedure that fairly and equally considers the interests of all concerned parties is sufficient for yielding a legitimate outcome. legitimacy should be both a fair process and a rational outcome (Peter 2007: 333). That is, the outcome must not only be fair, it must also be one that all parties are willing to accept. Some argue that one implies the other, or that a perfect procedure i s sufficient yield a rational outcome (Cohen 1997, Habermas 1975, 1996). Habermas, for instance, believes that if such a thing as the common good or common interest exists (he calls it s constrained by rules that restrict our speech to the search for such interests. This speech must abide all
61 exclusive motivation of participants (1975: 107 8). He concludes that if under the strict Like the pure proceduralists, Habermas is hesitant to assume that a general interest exists. However, he also denies that the intrinsic value of a fair procedure is sufficient to establish legitimacy (Peter 2010 ). Here, deliberation serves as the vehicle or instrument through which such rules would be identified, weight ed defended, and ultimately proven worthy of becoming enforced as a law. nterests, preferences, or beliefs. For this reason, rational proceduralists prefer deliberative procedures that allow all parties to exchange reasons for and against propositions in order to arrive at a rational outcome that resembles the conclusion of a l ine of making procedure because it is designed to reach fair decisions through inclusive and collective speech (Benhabib 1996a, Bohman and Rehg 1997, Chappell 2011, Cohen 1997, Dryzek 2001, Es tlund 1997, Gutmann and Thompson 1996, Habermas 1975, Manin 1987, Young 2000) Cohen (1997: 74 decisions ; (ii) deliberators restrict themselves only to reasoned speech, meaning propositions that are defended by reasons that can be supported or criticized; (iii) the rules treat all deliberators equally, meaning both that everyone has equal say, and that
62 power relations outside the discussion have no bearing on what is said or not said; and ideal procedure as one in which free and equal participants commit to obeying the decisions that they make on behalf of the public interest. The legitimating force in this process is the deliberative exchange and assessment of reasons by and about all thos e who are involved in obeying these decisions. This explanation of deliberation helps us to understand the collective decision making mechanisms through which shared governance operates: Deliberation enables conversations designed to produce fair decisions that all (governance units, if not individuals) could, at least theoretically, agree to because they involve discussions of what is best for everyone involved. deliberation should be reasons that all could accept, proceduralists look to the kinds of arguments that could be subject to consensus. This does not mean decision makers hold full, unanimous agreement as the direct aim of their deliberations or as the stopping point at which mutual communication ends (Landemore and Page 2015: 3 4). Unanimity is rare, not just in faculty senate committees but writ large 12 Rather, consensus is an ideal that motivates the actors in the decision making process. Thus Cohen famously argues could be the object of free and reasoned agreement among equals (1997: 73; my emphasis). When we want to achieve consensus, we focus our debates and discussions on the type of reasoning and argume nts that others could agree to. We try to find common 12 The problems associated with unanimity and consensus are discussed at length in Chapter 5.
63 ground and make reasonable suggestions. In short, collective decision making is animated by the will to reach consensus. Like Cohen, Habermas (1996) argues that as an ideal, consensus governs the kinds of arguments we make (and refrain from making), what we say, how we conduct ourselves, and what kinds of reasoning we use in order to search for agreement. We the app inclusive decision making in the first place. Equally important is what we omit in our deliberations. We ignore the narrow pursuit of self interest if it conflicts with agreeab le outcomes (Cf. Mansbridge 2007, Mansbridge et al 2010). We avoid polarizing or divisive speech, ultimatums, and strategies to pursue goals other than the expected be free of coercion, except for the motivation of reasons that Habermas (1996: 306) this kind of speech threatens or makes impossible the consensual outcomes. Proceduralists deny (whether intentionally or not) that there is a procedure independent standard of correctness, which means either fairness or the idea of consensus are considered suff must occur the only condition of correctness is an outcome that all agree is fair and are thus willing to consent to. Whether or not that decision is correct according to any other metric is in consequential or beside the point. However, because proceduralism
64 outcomes it produces and to consider that some outcomes may be preferred over others. Or, put another way, it only considers standards like fairness that describe the decision making procedure itself. Moreover, it avoids justifying group decisions on the grounds that they produce good, right, or beneficial results. This makes sense if we assume that a plu ralistic liberal society is incapable of determining what the common good is or if such a thing exists, as Rawls does in the example above. Thus he argues fairness is a second best heuristic for settling matters of moral disagreement in order to answer que stions about what we are to do. Under these premises, it does no good to consider the general will or some other notion of common interest as the standard of a correct decision. One might as well be waiting for Godot. This is not to say proceduralists are un concerned with outcomes. Many believe they produce genuinely beneficial results (Allen 2004, Barber 1984, Bohman 1996, Fung and Wright 2003, Gutmann and Thompson 1996, 2002, 2004, Knobloch and Gastil 2015, Lo et al 2013, Mansbridge 1980) Further, m any b elieve they imbue democratic values that make us more interdependent, trustworthy, and respectful of each other (Benhabib 1996a, Bohman 2003, Eisenstadt, LeVan, and Maboudi 2015, Elster 1998, Fishkin 1991, Niemeyer 2011, Young 2000) Mendelberg (2002: 153) says a properly deliberative process minded and However, they do not rely on outcome s in their defense of legitimacy and they do not (as a rule) include epistemic considerations of the quality, accuracy, or relevance of information upon which those decisions are made as a principle of deliberative substance.
65 Critiques of this political a pproach to academic decision making abound. How, for instance, does aggregation of interests help decide which university president will do the best job? Thus Stanley Fish (2007) argues the problem with this democratic model of procedural justice is that d liberal democracy is possible when the many put aside their personal belief s about the good life to answer questions about what the shop and how and by whom the shop is run are different matters. To conflate them is to turn an intellectual question what is good scholarship and teaching? into a political one process, and governance becomes an impediment (2007: 12). Inclusion of voices and sharing of information is good if it is the means to an end. Without direction, however, inclusiveness become the point of governance rather than a decision making tool. Collective Wisdom: The Epistemic Concep tion of Democratic Legitimacy A problem for proceduralists arises, however, when the subject of a decision (Estlund 1997). r pollution: If we decide that a problem, such as air pollution, is of public interest and coin to decide what pollution laws to enact, even though this would be procedurally fa ir. Rather we will judge the success of democratic institutions according to criteria that are (partially) external to the decision making process: do the pollution laws enacted actually reduce pollution to acceptable levels, at an acceptable cost? ...Wheth er the law succeeds in solving the problem for which it was drafted depends on its external
66 consequences not, or not simply, on the fairness of the procedure by which it was enacted. 13 outcome ( viz an affordable, effective, sustainable, and timely pollution policy), not just one that results from fair treatment. The correct solution is whichever one meets these conditions, and we justify democratic procedures to the extent they actually achieve such a solution In this case, the democratic process is used to solve an epistemic problem. Recall from Chapter some right (or at least best) answer concerning how it ought to be re solved that is 2014: 282). Call this requirement a procedure independent standard of correctness 14 For epistemic problems, such as the air pollution example above, sol ving them requires expertise, wisdom, knowledge, and the accuracy, quality, or relevance of information rather than fair procedure (Ober 2012: 5). It also requires consideration of judgments and beliefs, rather than interests and preferences. The question decision makers ask themselves is not what outcome everyone would agree to; rather, it is which solution is optimal for achieving the right, best, or most preferred outcome. 15 Epistemic democrats defend democratic processes as legitimate to the extent they yield wise, smart, or intelligent decisions (Anderson 2006, Bohman 2006, 2009, 13 say that instrumentalism is a consequentialist account of decision making. Consequentialism, as I understand the term, means that the moral rightness of an action is determined exclusively by the consequence(s ) it produces. In contrast, instrumentalism says that certain external consequences will determine whether or not a solution to a problem will achieve our preferred outcome. To mind, Arneson (2013) features the only fully consequentialist account of democr atic decision making. 14 15 In Chapter 3, I explore at length two epistemic theories of decision making one collective and one elite as they pertain to the case of academic shared governance.
67 Cohen 1986, Estlund 1997, 2008, Gaus 1997, Goodin 2003, Landemore 2013, List and Goodin 2001, Ma rt 2006, Nino 2006, Ober 2010, Peter 2009, Surowiecki 2005). This epistemic conception of legitimacy says knowledge and expertise is widely distributed among the populace and that democratic processes are valuable to the extent they pool this knowledge in to smart decisions. Epistemic democrats thus see democracy as a sort of problem solving process according to which aggregative and deliberative measures may be deployed. The epistemic defense of democracy relies on the contentious claim that such a standa rd of correctness exists, 16 Liberalism Against Populism Riker (1982) argues that populists are incorrect to assume that anything like (1951) having clearly ranked ordered preferences, and that no discernable ration ality could be discerned from analyzing such votes. In response, Cohen (1986) published what has become a foundational argument in the epistemic defense of democracy. 17 He argues that Riker would be correct in his argument if populists merely assumed the r esults of any voting procedure, whatever argument besides the fact that no one endorses such a claim (1986: 27), is that it assumes a liberal conception of voting, in which the purpose is to express preferences or interests. On 16 For examples, see Ingham (2012), Schwartzberg (2015), van Hees (2007) 17 For a detailed history see Schwartzberg (2015).
68 these grounds, which are echoed in the procedural account of voting above, there is no reason to assume that any true sense of a common good can be distilled from the aggregation of preferences. Instead, Cohen argues, a (1) an independent standard of correct decisions that is, an account of justice or of the common good that is independent of current consensus and the out comes of votes; (2) a cognitive account of voting that is, the view that voting expresses beliefs about what the correct policies are according to the independent standard, not personal preferences or policies; and (3) an account of decision making as a pr ocess of the adjustment of beliefs, adjustments that are undertaken in part in light of the evidence about the correct answer that is provided by the beliefs of others (1986: 34; his emphasis). Such an interpretation suggests that such a thing as the commo n good or general will exists, and that, contra Riker, individuals are not expressing their preferences or interests when they vote. Rather, they are expressing what they believe to be in accord with the general will. The third element requires that decisi on making resemble a form of consensus building, wherein all members are able to exchange beliefs with each other, reflecting on the responses given by others and adjusting their own in turn until at least a majority conclusion can be reached. This suggest s, as others have pointed out, but that they are willing to reflect on feedback and reasons that others give and adjust their own beliefs accordingly (Grofman and F eld 1988: 567) process: When in the popular assembly a law is proposed, what the people is asked is not exactly whether it approves or rejects the proposal, but whether it is in conformity with the general will, which is their will. Each man, in giving his vote, states his opinion on that point; and the general will is found by counting votes. When therefore the opinion that is contrary
69 to my own prevails, this proves neither mor e nor less than that I was mistaken, and that what I thought to be the general will was not so. fairness is the standard of correctness against which both political decisions a nd 1762/1993: 277). It is because 18 democratic procedures distill the general will that the outcomes of majority rule voting have political legitimacy. This epistemic defense relies on one application of the law of large numbers called the Condorcet Jury Theorem (CJT) that measures the probability of a majority guessing correctly ( Austen Smith and Banks 1996, Estlund 1994, List and Goodin 2001). The CJT states that if jurors are independent (they do not review evidence in a way that might cause them to converge opinions) and competent (each has a probability of 0.51 or greater of ch oosing correctly), then the majority opinion of a jury is more likely to be correct than any single juror (i.e. >0.51). 19 As group size increases, the probability of a correct majority opinion quickly app roaches 1, such that a jury of 5000 has a probability of over 0.90 of guessing correctly (Landemore 2013: 169). The CJT suggests then, that under these simple constraints, a majority rule is preferred to an expert rule. great extent with comparative institutional design: with identifying the institutions that 18 19 be italicized for emphasis.
70 yielded an understanding in the literature that deliberation can serve the same func tion knowledge and justified beliefs about what is in the best interest of the group and/or how to achieve such ends. 20 Consequently we have seen the basic premise of the CJT applied to deliberative procedures. For instance, Landemore (2013) suggests that democratic decision making with regard to national politics should consist of two steps: (i) inclusive deliberation at the national level that informs, refines, and develops smart individual opinions that is followed by (ii) aggregation of those beliefs that concludes with a majority rule (2013: chap. 4, 6; Cf. Ackerman and Fishkin 2004, Fishkin 1991, 1995, 2009). Insofar as deliberation requires talking and listening to others about reasons, beliefs, interests, and preferences (among other things), the search for institutional design has led theorists and practitioners to discount the independence criterion that characteri zed the early epistemic voting models. In fact, and as I argue in the remaining chapters of this project, deliberation may actually be optimal insofar as it allows for individual group members to adjust the accuracy of their beliefs until a workable soluti on achieves consensus. 21 Aggregation and deliberation are therefore used to the extent they produce epistemic benefits. general will, the general interest, the common good, truth, t (Landemore 2013: 44 20 This is perhaps due to the fact that deliberative democracy was not yet a mainstay of democratic theory 21 One example of an epistemic theory of deliberation is discussed in detail i n Chapter 3.
71 reflects the best interest of the people and is independe nt from their actual, possibly mistaken preferences, beliefs, judgments (2013: 45). In this case, the standard represents any goal that unites a group and serves as the principle for which it is a group in the first place. Cohen (1986: 34) offers four cri teria for determining a general will: (1) the members of the group share a conception of the common good; (2) the members regard the fact that an institution or policy advances the conception as a reason for supporting it; (3) it is fully common knowledge that the conception is shared; and (4) the conception is consistent with the members of the society regarding themselves as free and equal. These criteria are not particularly hard to meet. In simpler terms, they require that a group (3) freely (1) shares a conception of the common good, which they (2) have good reasons for sharing, and (4) it is common knowledge that they share such a conception. independent standard. A strong existing, definitive common good. Democrats of this sort believe that there is a true or right answer that collective decision making is tasked with discerning and thus defend d arguably lump Platonic moral realism or logical positivism into this category as well. In contrast, a weak epistemic version merely acknowledges that for some collective decisions some answers, solutions, and decisions are better than others, and that our standard is whichever available option proves better or would likely achieve the preferred outcome (Landemore 2013: 210).
72 Expressing Interes ts Versus Adjusting Beliefs Whereas the content of procedural decisions is the expression of interests, epistemic conceptions of decision making instead rely on the exchange or expression of d the importance for this interests, such as we saw described above in the Rawlsian libe ral democracy (Mansbridge 1980: 3 5). In contrast, t o to achieve their commonly held goal. find it hard to agree on any principle for (1980: 4). We typically experience adversa ry theory when we participate in political democracy, such as when we elect an official to office or vote on referenda. Based on the description above, these are the conditions that describe decision making in inclusive governance contexts. Adversary theor y is based largely upon laissez faire economic theory, which understands the aggregation of individual preferences as the closest approximation we competition can in princ iple aggregate millions of selfish political desires into one
73 is founded on self interest, justified as the only way to guarantee that some are not unfairly privileged over others. While this allows f or fair process, there is no guarantee that the outcomes it yields will be good ones or will yield a rational outcome (Arrow 1951, List et al 2013). As two procedure is They do not always resolve the tensions that produce disagreement, and they do not Hence, it m ay be the case that adversarial and procedural democracy entails above all (Schauer 1999: 24) equal power to protect their interests agains t one another since each will promote the understanding of the common good or general will, the goal becomes a discussion about how to accomplish or achieve its preferred o utcomes. This requires discussion not about interests but beliefs. Such a process makes expertise important to the deliberative process. Unitary theory presupposes that everyone is willing to forego things like equality of voice in order to privilege those voices that offer the best chances of achieving their mutually preferred outcome. In fact, equality may impede the process
74 insofar as it gives equal weight to all information, some of which may be unhelpful or irrelevant in achieving the common goal. Thi s is not to say equality is unimportant; equal respect is presupposed by the fact that everyone wants to achieve a mutually preferred outcome and achieved because all parties freely share in that conception. This allows for the expression of beliefs throug h deliberation and the reflective adjustment to them in light of hearing agreement with, at least willing to be concluded by. Accordingly, unitary democracy involves decision making through consensus rather than majority rule (Mansbridge 1980: chap. 14). Thus it may be justified to privilege some voices without requiring equal participation from those who would remain silent. These are the conditions, I contend, under which divided governance takes place, because they allow for epistemically warranted beliefs about correct decision making. antly common or conflicting agreement, and therefore rules of consensus, are only possible in circumstances in which all members share common interests and privilege the achi evement of that goal above other normative values like equality. In simple terms, when we all want to achieve the same goal, we are willing to let the wise or knowledgeable speak on our behalf, and we are likely to allow their opinions, choices, recommenda tions, or decisions speak for the group because we think that doing so will yield the outcome that we all want. However, in adversary conditions, the only thing we can agree on is a fair decision
75 because there is little or no common ground on which everyon e can agree. Thus, if ex ante agreement to let a fairly substance, because there is a fair and equal chance of my prefer ences being held in the majority. Such adversary conditions, however, do not involve the expression of epistemically warranted or justified beliefs. Shared Governance: Wicked Problems, Procedural Solutions amental distinction in the way that groups approach collective decision making. If a group can agree that there is a right or best way to solve a problem, then its goal becomes finding the optimal solution. Whether we dress that solution up in lofty, abstr act terms like the general will, or merely call it the best available option, it becomes the end toward which group problem solving is the means. Some argue it is not even necessary to know ahead of time exactly what that solution is or entails, just that we have a belief that such a thing exists and that our joint efforts will approximate it to some degree (Landemore 2013: chap. 8; Landemore and Page 2015). However, if we know that such a point of common interest does not exist or if we refuse to consider it, then the only way forward for collective decision making bodies is to count up equally weighted interests and proceed with the most popular one. Inclusive shared governance practices, I mean to suggest, force problems into an adversarial framework. Th e Joint Statement they frame decision making as the negotiation of conflicting but entwined interests that must nevertheless res ult in agreement. The Joint Statement defends collective decision
76 making processes on procedural grounds by averring it is normatively preferable to include all governance units, their interests if not their direct participation, because it helps reach dec isions that are considered agreeable to everyone obliged to follow them. Inclusive governance forces problems into an adversarial framework when it frames problems as disputes of interest, when it relies on rights of participation and inclusion of interes ts, and when it considers the standard of fairness sufficient to resolve such disputes. This is an inappropriate theory of collective decision making, because shared governance is a problem solving mechanism designed to solve concrete epistemic problems l ike budgeting priorities, strategic planning, and the hiring and firing of presidents. These kinds of problems are not, or are not primarily, solved through the becaus e quite often they have an optimal solution that is located or sufficiently approximated with recourse to expertise, knowledge, justified belief, or wisdom. 22 This sensitivity to epistemic problems is largely missing from procedural practices. If any soluti on that results from fair process can be subject to agreement, then there is no guarantee it will be optimal. As Fuerstein shows, epistemic problem only if, of all the possible agreements to be had, only some are properly wanted, and s ome properly wanted more than others. But if it is only agreement that we are after, then that premise is false, since sameness of belief is a property equally well can stand as correct and end the 22 They tend to be so difficult to solve that we cannot even say what will be necessary ahead of time to solve them. This means that although they are epistemic in nature, we cannot rule out ex ante what kinds of information, expertise, and justified belief will help solve them; reasons enough to support collective problem solving and decision making.
77 conversation or debate. Thus, whether intentional or not, the answer we think is most fair will often be presented as the best or most preferred one. But this standard is inappropriate because it is not sensitive to finding the correct, or better, solution. Shared governance practices, such as joint effort decision making among governance units are supposed to be optimal for handling pressing academic problems with difficult, if concrete solutions. These kinds of problems are often epistemic and technical in nature. Put another way, they are not, or not primarily, normative problems (even if they may require normative considerations). And yet, while acknowledging this, shared governance is structured according to a normative theory of inclusive decision making that treats these problems as if they can be solved through the fair and equal treatment of interested parties and agreeable solutions. In contrast, divided governance changes the rules of the game. Here, one or a few experts who share interests adopt unitary conditions and rely on epistemic concerns like expertise, justified belief, and knowledge in order to achieve their common goal. While critics of shared governance like Duderst adt point to the size and composition of the decision making body, it is actually the decision making conditions that determine the success or failure of its practices. Size and composition are not the primary considerations. typology also suggests that consensus has no place as a decision rule when parties hold conflicting interests. Consensus rules, and even stronger unanimity rules, presuppose that one conclusion will satisfy all parties enough that they will be willing to go along with it. But when interests become the object of debate, there is no guarantee such a consideration will happen. Even the rational proceduralists like
78 Habermas and Cohen, who argue that our deliberations should be guided by the will to find consen sus, do not make the mistake of holding it as the actual stopping rule that must be satisfied before deliberations end. Rather, they say bargaining and negotiation will have to play some role in finding workable agreement, if such a thing is to be found at all. Thus agreement models that seek to find solutions that all governance units can live with are likely to be unproductive or ineffective when parties cannot agree on what the outcome should be or what its most important goal is. Epistemic problem solv ing cannot take place when group members disagree in the first place about the ideal goal or solution they want to achieve. If the point of adversarial theory is to find a way for everyone to express their interests without privileging one viewpoint, then the priority must be ensuring that these are counted as equal. However, what has this to do with problem solving? When we engage in either aggregative or deliberative methods of solving problems, the content of our votes and speech is beliefs about what is best, not about interests. This means, for instance, that committee work to develop workable budget priorities should be comprised of people giving and adjusting their beliefs about what aspects of campus deserve funding. If the object of discussion and d ebate were solely about interests, then there would be little hope for reaching a compromise, reason enough to understand why Duderstadt calls conversation, especially if harm stands to be done (Mansbridge et al 2010), but interests cannot alone satisfy the demands of epistemic problem solving. When problems are epistemic in nature and thus have a correct solution, it is in herefore stand as the shared interest
79 and outcome that all are willing to find. Assume, for instance, that the question of whether or not an underperforming college should forego developing original curriculum and instead import a series of massive open on line courses (MOOCs) (Haber 2014) On the one hand, MOOCs are financially beneficial because they both drastically cut costs while allowing schools to register countless more students through online platforms. On the other hand, forcing professors to teach would effectively violate the doctrine of academic freedom. We can assume that there is a right or best solution to this problem, even if we do not yet know what that solution is. That solution is the subject of shared concern. F aculty, administration, trustees, students, and external stakeholders like states, legislatures, or churches (in the case of parochial schools) all share an interest in achieving this outcome. But a decision making model based on producing acceptable agree ment cannot provide guarantees that it will, or even intends to, find the solution that a procedure independent standard of correctness would identify as the right solution. Consider the critiques offered earlier by Duderstadt. He says that faculty are ve ry good at running their own departments and managing their native disciplines but they rarely make consequential decisions about pressing issues. He explains this differen ce as one of ineptitude or lack of concern: faculty are bad decision makers at large scales and/or they are not personally invested with steering the college/university. This explains why he and other former presidents say that the best way to get things d one is to draft policy proposals and then ask the senate for a thumbs up or down rather than
80 becoming more insular and faculty senates are declining from influential to ceremonial bodies (Minor 2004). However, I see a different explanation, and the procedural/epistemic debate offers us an optic through which to understand this bifurcation of effor t: simply, academicians apply a fairness standard to decision making at the campus level but treat matters in their own departments with epistemic considerations like accuracy, wisdom, and expertise. Those faculty senates that remain influential, I suggest are the ones that approach epistemic problems as such. How to address this issue is the subject of Chapter 3. Conclusion Ultimately, the adversarial framework and procedural defense of inclusive participation does not serve as a good model for shared governance. Adversarial theory is premised on the assumption of disagreement and the need to protect conflicting interests It recognizes everyone as free and equal and is concerned with inclusion and consideration of interests In doing so, it focuses on the mechanisms, processes, and procedures that help enable (if only theoretically) fair decision making While normatively praiseworthy, this is simply not the situation that academicians find themselves in when trying to reach decisions about how to run t he campus. What it ignores is evaluation of the outcomes that result and the quality of the input that produces them. Further, it does not provide the tools necessary to handle the kinds of epistemic problems that shared governance is charged with solving. For these reasons inclusive participation premised on the procedural conception of fairness simply misses
81 Procedural concerns for fairness and inclusion should have a place in collective d ecision making. Considerations of what helps or hurts members of the campus should inclusion of many and diverse perspectives allows for easier and better problem solving, a topic I take up at length in Chapter 3 But as I have shown here, discussion and protection of interests and preferences are insufficient to solve pressing, concrete, and epistemic problems. Inclusive governance practices fail to the extent they employ t he standards of fair decision making in the search for solutions to these problems. Divided governance works because it makes the shift from process to outcome. It holds the accuracy and relevance of information, knowledge, expertise, and justified belief as the standard of correctness when finding decisions. Critics of shared governance correctly point out that divided governance is more effective but incorrectly identify the smaller number of problem solvers, rather than the shift in standard, as the rea son for its success. I have argued in favor of holding shared governance to the epistemic standard of accuracy rather than the procedural standard of fairness. That is to say, shared governance should be held to the same standard as divided governance. In Chapter 3 I turn to questions of how best to meet this standard. Given the failures of collective decision making, one could very plausibly assume I point to technocratic solutions. However, this is not the case. Instead, I show that large, diverse groups are better able to meet this standard than elite groups of experts. Thus, I show that what shared governance ultimately needs is better, not less, deliberation.
82 CHAPTER 3 THE POWER OF PERSPECTIVE: COGNITIV E DIVERSITY, EXPERTISE, AND DIFFICULT PROBLEMS Introduction In Chapter 2 I argued that one basic failure of shared governance is the tendency to approach problems as if they were normative and could be remedied by intrinsically fair and inclusive procedure s. In response, I showed that epistemic values like accuracy should replace fairness as the necessary and sufficient standard, and that this standard is met to the extent it relies on wisdom and expertise, rather than the aggregation of interests, to make decisions, choices, or policy (Ober 2009: 2). Insofar as this closes one door, it opens another: Who is best capable of meeting this standard? Should decision making institutions or formal power structures change to meet this standard? Put simply, should academic governance be an elite or collective responsibility? In this chapter, I defend the latter, arguing that collective decision making is better able to meet this standard than models that rely on the intelligence and knowledge of just one or a few de cision makers. I attempt to falsify the claim that an expertocracy is better able to solve problems under conditions of difficulty, complexity, and uncertainty. To the contrary, these conditions warrant inclusive problem solving and collective decision mak ing. To this end, I compare and contrast two epistemic theories a collective model and show that the collective model outperforms the elite, even by the ards. I find that expertocrats correctly diagnose the failures of collaborative decision making and are right to bring epistemic concerns like accuracy, expertise, knowledge, and correctness into the debate over the future of shared governance.
83 However, th ey fail by their own lights to produce a better prescription. Because groups tend to be the best problem solvers, these findings support the idea that shared governance should rely on collective problem solving institutions. he Expertocratic Critique The failures of shared governance in recent decades to provide accurate and meaningful problem solving have led some to advocate doing away with the institution altogether (Bowen and Tobin 2015, Collis 2004, Duderstadt 2004, 2009, Duderstadt and Womack 2004, Fish 2007, Keller 2004, Pierce 2014, Roth 2015, Schmidt 2014, Thompson 2005, Tierney 2004) Though it still enjoys paradigmatic status, these critics see shared governance as an idea that has outgrown its usefulness. They claim decision based nature is simply unequipped to handle these conditions. It is a school bus when schools need a Ferrari. While the re are numerous ways to construct an epistemic theory of decision making, not all are built in the same way. This section explores an elite or non inclusive is a push to do away with shared governance or to redefine the term in such a way that Chapter 2 First, shared governance has two components I have labeled inclusive and divide d: inclusive governance refers to those decisions that are made with joint effort by two or more governance units, possibly with differing levels of authority. This is what the divided governance refers to authority held exclusively by one unit, such as faculty control over curriculum design or
84 trend toward divided governance, such that fewer decisions are being m ade through joint effort. Third, some critics charge that administrative authority largely became subject to inclusive governance while academic authority has remained almost exclusively the province of faculties. In the face of the myriad failures of inc lusive governance, many have called for an alternative decision making regime in which roles and responsibilities are strictly aligned with expertise such that governance units have discrete and non overlapping spheres of authority. Such a model would retu rn authority over non academic matters to administrators by reducing faculty involvement to a consultative or advisory role The expertocracy thus resembles encouraged to consult but are also free to act singularly if necessary to adapt to changing academic landscapes (Bowen and Tobin 2015: 184) In political terms, an expertocracy is a highly skilled oligarchy designed to grant dec ision making authority to those with the most cognitive capacit y It is thus a n epistemic model of governance in that it takes into account considerations like sensitivity to knowledge, wisdom, justified belief, and relevant information. The expertocratic model rejects traditional shared governance practices like cons ensus 2015). Recall from Chapter 2 that Duderstadt says the problem with shared governance ge accumulation (2004: 138). While these allow for the creation and dissemination of momentum, high risk, university wide decision environm ent
85 (2004: 138). That is, academe now requires quicker and more adaptive decision making that the glacial pace of faculty senates cannot accommodate. As a result, shared governance structures like faculty senates are venues for perpetual talk rather than a ction. Further, Duderstadt claims, faculty are too numerous, too apathetic about campus wide issues, too self interested when they do participate, and too unfamiliar with managerial decision making to alleviate the changing needs of colleges and universiti es as they navigate the unknown and unknowable terrain of the twenty first standing tradition of shared governance, in which power is shared more or less equally among all potential decision makers, is cumbersome and awkwa The Expertocratic Case for MOOCs In response to these perceived failings, expertocrats suggest that deliberation and collective decision making have become a detriment to schools. To illustrate their point, consider the argument Bowen and Tobin (2015) give for adopting massive open online courses (MOOCs). 1 They claim that online platforms a re bound to become more than just so many tools to improve the way we teach cont ent and run higher education; they will eventually be the very foundation on which all its institutions are built. MOOCs are a cost effective way to reduce overhead while provi ding high quality instruction and materials to growing numbers of students, many of whom now attend college online. As schools move increasingly into the digital domain, MOOCs may prove to be the future of teaching itself. 1 c ontroversy surrounding them. For a good overview, see Haber 2014: chap. 4.
86 In 2013 faculty at San Jose Stat their course listings for the upcoming year. This addition, among many others, was part of a cost saving measure by upper leve l administrative leaders who replaced or known MOOC providers like edX. The move, they claimed, would lower expenses without sacrificing quality, because the courses were recordings of elite professors teaching at elite universities. However, reactions to the expertocratic action were not well received. Currently, it is common practice to delegate authority over teaching methods to departments, meaning administrators do not control if or how much a school adopts MOOC technology. Most teachers prefer to teach their own lessons and will use technology to the extent that it enhances direct interaction with students. However, with man y ( Bowen and Tobin 2015: 116). 2 having a scholar teach and engage with his or her own students is far superior to having those students watch a video of another scholar engag ing his or her students (Kolowich 2013) From their perspective, no streaming video could substitute in class student interaction. The move, they feared, was made with haste a nd bottom line thinking, and without regard to what they saw as an obvious reduction in instruction al quality. In response to this negative feedback, the administration ultimately 2 (2014) Academe A professor who does not choose and present educ ational content in his or her courses is no longer a professor in the traditional sense of the word, becoming instead a glorified teaching assistant
87 walked back their move content with their own (Kolowich 2013). Additionally, many schools have failed to adopt MOOCs or similar platforms because that decision is left in faculty hands. This was the case in 2014 when the University of Florida asked a number of its departments to decide whether or not to contribute content to a new online branch of the university called UF Online. After a period of research and deliberation, the Political Science department voted decisively against joining the program. While the decision was not u nanimous, two thirds of the faculty felt that adding an online major would essentially create a digital copy of the brick and mortar department. And because the online school targets traditional students (i.e. undergraduates) rather than non traditional st udents, there were feelings that such intra university competition (Straumsheim 2014) Furthermore, MOOCs often require that faculty members relinquish control over instructional method to another gove rnance unit (or worse, an external unit like a state legislature) This is a source of real concern among the teaching community with many declaring that onli ne programs comprised of MOOCs like UF Online direct ly threaten academic freedom. As a separate a rm of the school, online programs do not belong and are not responsible to their corresponding departments. So, for instance, content provided by political science professors would be nearly identical to the lectures they give in person but would remain th e property of the university. This poses a potentially existential threat to departments: if enough students opt for the more convenient online option, departmental enrollment (and funding) will dwindle Furthermore, online
88 programs can hire their own adjunct lecturers that essentially present and expound on the prerecorded content provided by faculty. Thus, b y unilaterally deciding on teaching methods, it is plausible that administrators may gain back door access to curriculum, a province traditionally and exclusively reserved for faculty. 3 In response, professors and teachers have fought vehemently to retain the right over teaching methods. From an expertocratic perspective, these acts of refusal are myopic and narrow minded. According to Bowen and Tob in online technology will eventually be the very foundation on which all its institutions are built replacing brick and mortar classrooms The MOOC controversy, then, has all the makings for catastrophe. Those that cannot respond and adapt will inevitabl y fail to succeed in the twenty first century. Thus faculty in both the San Jose State and Florida cases are not just delaying the inevitable, they are also suffering the opportunity costs of not adapting at the right time. Technology is just beginning its inexorable crash upon the shores of academe; there is no telling how it will affect specific institutions or higher education writ large However, faculties are empowered through mechanisms like shared governance to ignore the warning signs and thus refus e to adapt. The MOOC example also illustrates the theme of information sharing. A mainstay of the expertocratic argument is the full and open sharing of information between governance units. C ollaboration and deliberation should be consultative rather tha n consensus building A dministrators should seek out input from other units as much and 3 It is telling that among the six schools sanctioned by the AAUP for infringement of generally accepted standards of shared governance, faculty authority over curriculum is not the issue. At Lindenwood University, e.g., curriculum was the only matter that was left to faculty. See AAUP (2016)
89 of the administration (2015: 207) Through consultation, faculty would help d etermine how (rather than if ) MOOCs will be used. They would play an instrumental role in designing courses to fit department and institutional needs, something that only their expertise makes possible On this model, faculty input is still sought out, and leaders ants rather than as adversaries (Roth 2014). grudgingly, any claim to sole authority ove r teaching methods of all kinds ( Bowen and Tobin 2015: 173). They see faculty much as Plato (1945) sees the demos bound to destruction by the pull of innocent yet misguided self interest. They fail to understand that teaching methods have become so much more than a compone nt of academic freedom; they are now tied to the fate of the institution. An enlightened faculty would not only embrace the coming technological era, it would happily do whatever it could to secure it. If only we could place the locus of authority in the h ands of the knowledgeable few who know how to respond, we could avoid disaster an d enter a new academic era. In sum the MOOC controversy illustrates the ways in which expertocrats redefine takes full advantage of faculty expertise but that leaves final authority for most of these complex matters w ith administrators and trustees (2015: 175). In response to this unknowable and yet fast approaching future, expertocrats argue that schools must exaggerate the importance of having educational institutions (and their offspring) benefit from the ability to make decisions promptly, and then to change course if need be.
90 Nimbleness is exactly what current shared governance practices lack. Practices like consensus driven deliberation are time consuming, require extensive discussion, and often lack any accountability (D uderstadt 2004: 241 2). Hence the solution is an elite body of expert problem solvers who can act quickly and effectively to accommodate the kinds of changes required to survive. n[ing] decision (Bowen and Tobin 2015: 186; Cf. Thompson 2005). Administrators, faculty, and trustees should consult with each other as much and as often as is necessary to make sound, adaptive decisions But ultimately each unit is held responsible for successfully overseeing its unique portion of campus responsibilities. Using this model, adherents claim, would retain the epistemic benefits of deliberation ( the dissemination of knowledge) while eliminating the drawbacks ( the stagnation of consensus building). In summary, the nature of twenty first century higher education is changing. Forces like technological innovation are the increasing difficulty, complexity, and uncertainty of academic prob Expertocrats are those who feel the best course of action is to replace shared governance with discrete and non overlapping spheres of decision making authority accompanied by the free flow of infor mation between units to facilitate decision making. Cognitive Diversity: An Inclusive Approach to Wicked Problems Though expertocrats correctly diagnose the failures of inclusive governance, I find their prescription for improvement is not without its own flaws. For one thing, it is not immediately evident that the wisest individuals will unproblematically make the best
91 decisions. For instance, Estlund (2008: chap. 11) points out that in a hypothetical till suffer from everyday biases, such as favoring their own race or gender, even if they believe they are promoting the good or making the best choice. These biases would result in epistemic harm and yield suboptimal rulings, despite the rulers being empi rically the most qualified. Thus, if we are concerned with finding the best solutions possible, one necessary requirement is accurate information. If we incorrectly assume the wisest produce the best decisions, then we risk making two concurrent errors: cr afting a suboptimal solution, and failing to recognize its suboptimality. In other words, we run the risk of mistakenly thinking that one solution is best when in fact other, better options are possible. So we have to keep open the idea that the right or b est solution is not necessarily the product of intention. By contrast, I leave open the possibility here that it is instead an emergent property of collaboration. In this section I consider the implications of the expertocratic argument through the lens of diversity theory. I find that by assuming experts know best, the expertocracy cannot account for: (i) the difference between risk and uncertainty; and (ii) cognitive homogeneity. Diversity Theory In recent years, Hong and Page have developed what they cal (Hong and Page 2001, 2004, 2012, Page 2007, for historical evidence, see Ober 2009) They find that outperform groups o f high ( 2004: 16385 ). These find ings come from a set of formal models in which individuals of varying intelligence (IQ) were randomly assigned to groups and given problem solving tasks. Intuition would suggest that the smartest individuals and the groups with the highest average IQ would be more
92 likely than average IQ groups to successfully solve problems. However, the y found that diversity a better predictor of successful problem solving than was intelligence. C ognitive diversity is defined as the difference in pers pective s, heuristics, interpretations of information, and predictive models that individuals use to analyze a problem and contribute to finding an optimal solution (Page 2007: chap. 1) The more cognitive diversity, the more ways the problem can be viewed, understood, and analyzed. to face interactions information, judgments, and knowledge into one solution ( L andemore 2013, Landemore and Elster 2012, Landemore and Page 2015 ). Any (which accounts for suboptimality) is likely to be cancelled out by the correct information of one or more other members while amplifying correct infor mation. This process yields what the authors a sufficient degree of cognitive diversity. Suppose, for instance, that a problem is so difficult that even the best problem solver cann ot reach the optimal solution. According to Page (2007), collective intelligence occurs when improvements are continually solution. Hence each person is able to contrib ute pieces of the puzzle that collectively fit together. Through this negative correlation in judgments, the right or best choice is located. By viewing a sufficiently difficult problem in different ways, a diverse group can start with a suboptimal solutio n and continually map small improvements onto it until
93 they reach the best solution. Ideally, deliberation ends when all problem solvers agree, despite their differences in reasoning, that an optimal solution has been located. Four conditions must be met i n order for diversity to trump ability in the problem solving context (2007: 159 162): 1. Sufficient difficulty : problems must be difficult enough that even the best individual problem solver cannot find the best solution on his/her own. 2. Relevant problem sol vers : each group member must have some kind of relevant knowledge about the problem at hand capable of producing a solution. As Page puts it, adding a poet to a cancer research team is unlikely to help ( 2007: 164). 3. Sufficient diversity : there must be enou gh perspectives and heuristics in play that the current solution can continually be improved upon until the global optimum is reached or it is determined there is not one. 4. Large population : the population from which the group is selected must be large eno ugh that any randomly selected group would be diverse. Additionally, the problem solving group must also be large. These conditions, I argue, are all easily met by the wicked problems facing academe. Thus, a diversity theory approach is better suited to s olve these problems than is an expertocracy. To see why, let us look at the components of wickedness we discussed in Chapter 2 : uncertainty, difficulty, and complexity. Uncertainty, Risk, Democratic Inclusion It is not clear that a nimble elite is best abl e to handle the difficult, complex, and uncertain conditions imposed on academic decision makers. There is certainly something to be said for giving experts the room and autonomy to make use of their knowledge, and I do not mean to question that aspect of the argument. But there is something incongruous about the idea that a single person or small cognitively homogenous group will be able to guide an institution through the future, despite admittedly not knowing what that future will hold. Experts are gener ally the people we
94 hire to fix already identified problems. If the future really is unclear, then we do not yet know what problems we are likely to face, much less how to expertly resolve them. ly able to remedy them (in the same way that general practitioners are good at diagnosing diseases but refer patients to specialists for treatment). If the future is veiled by uncertainty, then why bet an elite group, which knows a lot about a few things, will be best suited for identifying and resolving those problems? How do we know their deep yet narrow perspective will not limit or distort how or what they identify as problems in the first place? Why not maximize the number of perspectives looking for t he problem? If uncertainty is smoke obstructing and clouding our view, then we may argue instead that the more hands clearing it away, the better our view will become. In other words, if we grant that diversity trumps ability, at least in theory, then the epistemic case for collective problem solving is actually strengthened under conditions of uncertainty. To understand why, we must first differentiate more clearly betw een two concepts: risk and uncertainty. Under conditions of risk, we know what the potential outcomes are but cannot say which is most likely to play out. For instance, a game of roulette under normal conditions is subject to risk: the roulette ball will f all into a finite number of slots. We know ahead of time the number of possible outcomes because it corresponds to the number of slots on the wheel, but chance ultimately determines which one it will be. Again echoing Rumsfeld, we can say risk presents kno wn unknowns. In contrast, uncertainty brings unknown unknowns. Under conditions of uncertainty, we have no knowledge of potential outcomes and so cannot predict which one is most likely to occur. Uncertainty, in other words, is prior to risk. This is all t o say I
95 consider uncertainty the condition of figuring out what is to be done, and risk the subsequent condition of determining whether a solution will work. In studying social movements, Ganz (2004) considers why it is possible for small, David like insu rgencies to sometimes defeat large, institutionalized Goliaths. One might point to the swift rise of ISIS in past years despite the best efforts of global powers like the US. Ganz finds the reason has to do with how flexible and adaptable each one is to ch anging (social, economic, political) conditions; that is, how likely each is to embrace and react to uncertainty (2004: 182 find new solutions: [U]n der conditions of uncertainty, the capability to gene rate new known algorithms, no matter how expertly. In other words, under conditions in which rules, resources, and interests are highly institutionalized and links between ends and mean Strategic capacity is thus more useful in explaining outcomes in turbulent environments where rules, resources, and interests are emergent and links between ends and means are uncertain (2004: 195). In other words, we can only rely on the expertise that brought past successes when we are certain that there is a predictable relationship between the problems that we face and the cognitive resources that success fully solve th em But under conditions of uncertainty, success is instead determined by the ability to find new answers to emerging problems rather than relying on what has worked in the past. Landemore (2014) argues that uncertainty requires democracy, while risk requi res expertise. When it comes to problem solving, risk and uncertainty present two distinct situations. When faced with uncertainty, the task should be identifying options and potential outcomes; that is, getting to a situation of risk. Call this the proble m
96 identifying step. 4 When faced with risk, the task should be increasing the odds of achieving the most desired outcome. Call this the problem solving step. 5 It is important not to confuse these two conditions: [I]f the bundle of political issues any huma n group faces over time was complex but merely risky, that is in the realm of calculable probabilistic certainty, then we would be better off weighing more heavily the voice and exper new game. Uncertainty means that we do not know and cannot know what the future will be like, whether on the short term or the long term, and whether, in particular, it will be like the pas t or radically different (2014: 168; her emphasis). When problem solving under these conditions, it is impossible to say what kind of information or skill set will be required, as there is no knowledge of what obstacles or problems may arise. From this pe rspective, pure expertocracies are problematic, because they do not take into account information levels under conditions of uncertainty. By this logic, Bowen and Tobin acknowledge uncertainty but fail to accommodate for it in their prescription because th ey assume elites have perfect information. Thus they see inclusive governance as an impediment because they tacitly assume a (relatively certain) risky situation. When we do not know what to expect, we cannot say ahead of time what kind of knowledge, expe worse, history cannot reliably tell us whose expertise to seek out because, by definition, not know, ex ante, who among the included will turn out, ex post, to have the kind of 4 5
97 knowledge or bring the kind of perspective needed to solve the problem because we simply do not know enough about the world to be able to predict what problems we are going to face as a group in the future, no matter how much knowledge of the past we may have (Landemore 2014: 174). It makes little sense to willingly restrict the skills and solving. Expertocrats often avail themse lves of ship piloting metaphors; the best captain should be at the helm during the worst conditions, because he has a proven record of excellence. But what if we cannot even be sure we will need to travel by ship? In this environment, democratic inclusion is faces, both about the kind of issues they will encounter and about the nature of the knowledge and cognitive skills required to address them that makes democracy epistemically superior decision hen we do not know what to expect past accomplishments or expertise may not be a good indicator of optimal future outcomes prov ed to be experts in the past. Difficulty, Complexity, Cognitive Diversity Diversity theory also responds better to the difficulty/complexity condition. Recall that by complexity, I mean that a problem and/or its attendant solution are comprised of many int erconnected elements of components. Difficulty refers to the lack of ease with which a solution is found. For the sake of argument, I consider difficulty a function of complexity such that as complexity increases, so does difficulty. It is helpful to reme sufficient difficulty requirement, meaning that the problem surpasses what can be
98 of future decision making cond itions in academe as uncertain and unprecedented (Bowen and Tobin 2015), this project assumes these problems meet the sufficient difficulty requirement and are thus too difficult to be solved with the simple application of already known expert knowledge. P olicymaking often includes so much complexity that existing knowledge (and a fortiori what can be known by just one person) is insufficient to find the best solution. Additionally, it may be the case that existing solutions are obviously suboptimal but no better alternative has been found. This is the case, e.g., with the swelling number of adjunct professors and instructors hired to counteract the rising costs of college tuition. While contracted teaching lowers labor costs, it also carries with it, inter alia reductions in education quality and threats to the tenet of academic freedom. These elements of wickedness make problem solving so difficult that no existing solutions completely resolve them. Adjunct labor may reduce operational costs, but very rare ly does it constitute a complete financially stabilizing solution. To be clear, these cases differ from situations where solutions to problems exist but are merely not applied, such as when a financial advisor shows me how to invest wisely or a doctor pres cribes me drug that will cure my illness. The problem is a lack of expertise (financial knowledge or medical treatment) that can be easily resolved by finding and applying the right existing solutions. The more complex the problem, the greater the need for cognitive diversity. To understand why, consider the opposite: cognitiv e homogeneity. Page argues that diversity succeeds because like minded people will tend to find the same solution (2007: 164). When a problem is sufficiently difficult that the best pr oblem solver is unable to find the best solution (i.e. when s/he locates a local but not the global
99 optimum), adding others who see from a similar perspective and apply similar heuristics will serve only to cluster them around the same solution. 6 The persp ective and heuristic(s) they bring to the conversation are already represented. For instance, if I that the best pie will likely be one of the priciest. This heuristi c helps me narrow a massive list by ranking them in order of price and then starting at the top. If I call friends to help in this important research endeavor and each applies the same thinking, they are likely to try the same few pies that I did and come words, we are likely to rank one above the rest and find ourselves stuck at this peak (but unable to tell if it is the highest peak). But it could be the case that other heuristics we omitted (say, quality of ingredients, number of years in business) would be more beneficial to answering the question than repeated iterations of the same heuristic, and that figuring those into our algorithm would produce a better ranking By this logic, unlikely to be effective if its diversity level remains static. In this example, cognitive homogeneity yields a ranking of pies that will likely lead everyone to cluster around one of them. I suggest that without sufficient diversity, this tendency to clus ter around one solution could easily become myopia if problem solvers believe that the solution they find will be optimal, simply because they cannot see another way. In this case, all are likely to have the same errors in judgment, wherein they jointly be lieve something that is false or wrong, but fail to see it that way. Attempts to problem solve will thus incorporate these incorrect pieces of information. Diversity 6 To borrow from Wittgenstein (1958) when we establish credibility or facticity b y repeatedly using the (Â§265)
100 by the correct information of one or more other members. Through this negative tend to be the one that diverse people converge on despite their myriad differences in reasoning and perspective. In short, when a problem is sufficiently difficult, the right or best decision or solution will tend to be the one that many minds agree on despit e differences in reasoning. To say diverse groups outperform smart groups does not mean we should aim for large dumb masses if it is possible to have both. For diversity theorists, intelligence is an individual attribute and diversity a group attribute. So increasing both tends to on solving difficult problems, and (ii) ar e not single individuals but rather groups of intelligent problem solvers whose size is determined by both the difficulty of the problem and the cognitive diversity needed to address it. Because cognitive diversity is more important than individual intelli gence, the best way to produce smart decisions is to maximize deliberative inclusion. Epistemic democrats thus argue the best way to ensure cognitive diversity through democratic decision mechanisms that include large groups of people. 7 Landemore, e.g., pr oposes the political 7 Though there are diminishing returns, namely when additional group members would start echoing already stated perspectives. See Chapter 4 for a discussion of optimal group size.
101 what matters most to the collective intelligence of a problem solving group is n ot so In sum diversity theory reveals two flaws in the expertocratic approach to wicked problems: first, despite acknowledging uncertainty, it proceeds as if the future w ere merely risky; that is, sufficiently predictable that we can say ahead of time what kind of expertise will solve future problems. Thus, while they acknowledge uncertainty in their diagnosis of shared governance but they make no room for it in their pres cription. Second, expertocrats undertheorize the ability of cognitive diversity (and thus overstate the role of cognitive homogeneity) in solving difficult and complex problems. Locating an optimal solution to suffic iently difficult problems like whether o r not MOOCs are a saving grace r equires first answering economic, technological, administrative, legal, ethical, and logistical questions, among others. But no one expert can achieve the optimal solution under these conditions. In other words, the expertoc ratic approach fails to account for the fact that the knowledge necessary to solve wicked problems is so widely dispersed that finding and aggregating it becomes an additional factor in solving them. ce Work? At this point we face a conundrum. Diversity theory suggests we should find collective wisdom in group deliberative settings like the academic faculty senate. But one need not be an expertocrat to agree with Duderstadt that they have become ineffe
102 making. 8 If groups are so wise, why do they fail to perform well in the context of shared governance? To answer this question, we must differentiate between signaling and explai ning. To signal is to express a preference or position or to name what outcome we would like to see. In order to know something about the meaning of that signal, it must be explained; that is, we must give our reasons. group that is required during a decision making exchange. Deliberation of degree zero the most minimal type, consists merely of signaling a piece of information. When we vote by raising a hand or casting a ballot, we signal at this degree. Degree zero deliberat ion is signal is produced. If the task at hand only requires collecting this information, such as when we vote in a national election, then further deliberation i s unnecessary, because the reasoning behind that piece of information does not contribute to the task. Deliberation of degree one includes an exchange of the accuracy of signals. This degree, too, can be satisfied nonverbally, such as when we recognize so reputation for success in past instances. A more descriptive type, deliberation of degree 8 that faculty senates have generally devolved into a venue whose purpose has become talk rather than Chapter 4 that differs from the way he uses it. I use the word debate to mean weighing the reasons for and against a proposition or set of propositions. I also consider reason giving and decision orientation to be components of proper debate. We generally do not see these qualities in the kind of communication Chapter 4 addresses these issues at length.
103 two This step requires some amount of substantive information exchange. deliberation of degree three interlocutor s provide one another with exhaustive accounts of what kind of models they used, why they resorted to such and such variables, based on what data, and even the reasons that they conceptualized the problem in a certain way. This kind of deliberative exchan ge makes available to all participants a full toolbox of concepts, perspectives, data, and variables to complicate 12). Deliberation is therefore a type of reason giving exchange that adds meat to the bones of signals (see Chapter 4 ). Suppose, e.g., we vote on a referendum to increase the city sales tax. A purely procedural account of voting would only be concerned with guaranteeing fair access to equally weighted votes. The intrinsic fairness of this procedure justifie s whatever outcome obtains. If our decision procedure is a majority rule vote, my reasons for voting do not have a place in the process itself. In fact, it would not make a difference if I had no reasons whatsoever. But an epistemic account of deliberation requires that interlocutors have and exchange reasons and the beliefs underlying them. Thus, if I favor increasing taxes and you do not, my goal as a deliberator should be evaluating your reasons from my perspective. If I can adopt your reasons, then perh a reason for my vote. My interlocutors would only know that from full exchange. For this reason, simple voting does not make use of cognitive diversity, and thus does not yield epistemic value In order for deliberation to produce epistemic value, deliberators must
104 be able to weigh and evaluate all of the information ( beliefs, facts, knowledge interests, opinions, preferences, etc ) brought into the conversation. We conclude that if gr oup deliberative procedures do not yield these results, then full exchange of signals and explanations must not be a functional component of the decision making process. If the essential function of a faculty senate is to allow faculty a thumbs up or down on o utputs (solutions, decisions, policies, etc) that are proposed to them rather than being the mechanism through which they contribute to crafting those outputs, then deliberation has become an additional step used to evaluate policy rather than craft it. St rategically, inclusive governance practices allow faculty to feel they have been heard and thus that shared governance is alive and well. But when it functions in this capacity, shared governance is not primarily about collaborative decision making. Conclusion Many arguments in the academic governance literature rely on nautical of Michigan is called The View from the Helm In many of these examples, the point is to illustrate the need for a decisive and experienced leader who can steer the ship to safety when the waters get rough. Yet these metaphors reveal certain misconceptions. First, ships are not operated by just one person. They rely on the coordination and communication of many people with d iverse areas of expertise. The captain ranks supreme, gives authoritative orders, and commands the crew; this is true. Bu t when it comes to navigating he is more like first among equals, in that he coordinates the intelligence, expertise, information, and actions provided of everyone else on the bridge. Though he charts the course, the ship gets there because of joint effort s more,
105 captains do not steer; h elmsmen do. In other words, the captain relies on the work of others; he is an integral part of a system without which he would just be a lone man at sea. A second and more important truth is revealed in their usage: elitist leadership is a product of a simpler past. These metaphors recall a navigational era when it was possible for just one man to steer an entire ship if necessary. This proves problematic when they are used to justify why the complexity and uncertain ty of future decision making requires an elite expertocracy. To solve the problems of tomorrow, they are calling for the ships of yesterday! In other words, they claim that the future will be different in an unknowable way, yet they harken to the leadershi p of a bygone era. We are thus led into a contradiction where both sides of the argument cannot be true. If tomorrow will always be more uncertain than yesterday, then we need governance that will always prove successful, then what justifies the claims of uncertainty? How can we be certain of its uncertainty? In other words, why do they treat unknown unknowns as if they were known knowns? My point is not to say we should throw out the entire e xpertocratic argument. In this chapter I have relied on the empirical and anecdotal evidence provided by these academic leaders who are frustrated with the very real inefficiencies and ineffectiveness of collaborative decision making. But I see the experto cratic response as reactionary above all else; as this collaborative thing and we all believe it should work, but it
106 9 But if their diagnoses are accurate, their prescriptions seem problematically antiquated. I am cure the very real failures caused by new kinds of problem s Put simply if we can only accept one side of the uncertainty difficulty, and complexity And if those are the defining characteristics, then we need to adapt accordingly. A s I have shown, the best way forward is through a diverse, inclusive system of deliberation and collective decision making that can pool individual expertise into collective knowledge. T his project takes very seriously the diagnoses but disag rees with their prescriptions. It argues for governance through collective decision mechanisms that address epistemic problems with epistemically warranted solutions, in addition to considering procedural concerns like inclusion and fairness. Finally, hav ing explored what structurally prevents current shared governance practices from meeting this epistemic standard, we have cleared the ground and made room for developing a new theory of deliberative problem solving. In Chapter 4 I pick up this thread by s howing that deliberation produces epistemic benefits if and when it consists of reason giving and decision oriented debate for and against a proposition or set of propositions by a large group of relevantly diverse problem solvers using democratic rules of participation. 9 Thank you to Henri Beunders for this expression.
107 C HAPTER 4 WHEN TO TALK IN SCHOOL: AN EPISTEMIC THEORY OF DEMOCRATIC DELIBERATION IN ACADEMIC SHARED GOVERNANCE The Problem o f Group Decision Making In Chapters 2 and 3 I argued (i ) in favor of holding academic shared governance to the standard of epistemic value rather than consensus, and (ii) that large, cognitively diverse groups will meet this standard better than small groups of homogenous elites. In this chapter I defend these claims by showing that collective intelligence produces optimal problem solving but that this is only achievable when deliberation is done well When it comes to decision making, the role of representative bodies like a faculty senate tends to fall on a s pectrum between advisory and legislative. In order to justify wielding legislative power, one must be capable in the first place of making good In the context of institutional governance, expertocrats have been successful at warranted outcome: there is often a right, correct, or at least best solution to most any problem a school faces. T hus t he expertocratic argument says exper tise and knowledge (that is, the epistemic standard) should be the basis of academic decision making Expertocrats are right to point out that consensus or agreement models among governance units often produce unrealistic requirements, and right to be conc erned that procedural justice does not always yield good decision making. This is aided by the concrete observation that shared governance bodies like faculty senates often seem unconcerned with or even incapable of handling pressing financial concerns. Bu t they also append to that definition the assumption that elite administrative experts are best able to meet this standard.
108 In this chapter I reject the second part of their argument. Whether shared governance bodies are ineffective because they have been undermined, or undermined because they are ineffective is a chicken and egg discussion I do not take up. Rather, my concern is showing that large deliberative bodies can produce the kind of solutions that meet the epistemic standard. Doing so turns on its head the assumption that only a small group of administrative elites is the best decision making unit to meet the epistemic standard. When problems are sufficiently difficult that a single person cannot resolve them, what good will it do to consult a few m ore similarly minded thinkers? To this end, I must show not only that deliberative groups are good decision makers, but also that they are better able to meet the epistemic standard than elites. Before this can happen, however, we must start with the quest ion: U nder what conditions does deliberation make for epistemically better decision making? Here, I am concerned with the conditions under which the deliberative process improves the quality of information that deliberators evaluate and leads them to produ ce the quality of information and of the decisions that are made with that information. nitive (Andler 2012: 73) in a way that allows them to produce wise or smart decisions. Group deliberation meets the epistemic standard when it produces better information and decisions than its members would have m ade as individuals (Hong and Page 2012: 56) This is sometimes referred to as collective wisdom or collective intelligence (Landemore and Elster 2012)
109 While many problems can plague a deliberative procedure, the ultimate failure at least according to the expertocratic c ritique we explored in Chapter 2 is an inability to produce good decisions when pursuing the standard of consensus. This is further exacerbated by inefficiencies like protracted discussion, self interested bargaining between faculty and admi nistration, and short time horizons. One possible explanation is that large groups are generally unable to parse in a timely fashion the abundance of information and knowledge their members possess. Just as direct democracy becomes impossible with a large enough population of citizens, there is simply no easy way to give all (or even most) faculty the opportunity to produce and evaluate valuable information in a way that meets the consensus ideal of shared governance. 1 However, for deliberators to make wise (or optimal) decisions the process must avail them of all necessary and relevant information and the means to evaluate it. Absent this evaluative process, a group is likely to get stuck on disagreement or to decide o n partial information, all of which could be irrelevant. Thus, when being held to a consensus standard, one challenge for collective decision making units is avoiding what Moore (2014: 13) on, this chapter is concerned with identifying how groups can deliberate in epistemically valuable ways; that is, how they can improve the quality of information in order to make Can groups 1 Here, I am thinking specifically of the basic assertion in the American Association of University viz. that everyone affected by campus policy should have some say in determining that policy.
110 2 In what follows I develop a theoretical model of deliberation based on the epistemic standard. My task is to f ind what specific qualities about the deliberative process makes collective wisdom greater than the sum of its individual parts. To do this, I look to the fields of deliberative democratic theory, cognitive and social psychology, decision making, and commu nication science Drawing on work in these fields, I find that the deliberative problem solving process provides epistemic value if it consists of reason giving and decision oriented argumentation by a sufficient number of diverse but relevant deliberators Whereas this chapter explores how to talk, the next concerns the optimal use of stopping rules (e.g. unanimity, disagreement) and decision mechanisms (e.g. consensus, majority vote) and the criteria necessary for choosing them. When Deliberation Works I n 2012 the Russian Studies major at Macalester College faced being dissolved when an administrative policy committee found that it was underperforming and had become a drain on financial resources (Klein 2012, Ruiz 2012) The problem, they argued, was insu fficient demand; students were no longer interested in the program. The Macalester faculty is relatively powerful compared to other schools; they have both jurisdiction in budgetary matters like this one, and their role tends to be legislative rather than purely advisory. A committee recommendation of this kind will pass unless a two 2 circumstances is democratic decision making making by experts appropriate and thus potentially legitima one offered here. This difference is explained, I think, by the fact that she is constructing a theory of political legitimacy, whereas mine is a theory of deliberation.
111 thirds majority of all faculty (visiting, tenure track, tenured, and emeritus faculty) votes to overturn it. In the months preceding their vote, the faculty responded by holdin g a series of townhall style deliberative meetings in which they debated the pros and cons of closing the program. From all accounts, those involved gave serious consideration to the reasons for and against closure. On the one hand, faculty members were he sitant to turn on each other and set a precedent of promoting program closures. On the other hand, closing the Russian Studies program meant more money for the remaining departments. While initiated by the faculty senate, meetings were open to all campus s takeholders, including students. After listening to a number of current and would be Russian Studies students, they determined the problem was not lack of demand but rather lack of funding and insufficient supply of interesting or important courses. The de mand was there. T he committee had missed this point because it only considered revenue production. In this case, the distribution of relevant knowledge included students, whose opinions (and dollars) became crucial to the decision whether or not to keep t he program afloat. This knowledge was only uncovered through deliberation designed to establish both reasons for and against closing the major. The committee failed to look at the causes of enrollment decline and instead only measured outcomes. In doing so it established a semi realistic understanding of the situation. Yes, the major was underperforming. But it also missed the fact student interest was alive and well. In doing so it saw as a failed investment what was actually an underperforming asset and reached premature consensus on the decision to close it.
112 This Macalester case shows an example of deliberation producing smart decisions. In what follows, I draw out three components of the deliberative process that were at play in this example and argue t hat they are necessary for producing epistemically valuable, as opposed to procedurally legitimate, decisions. The first and second are its decision oriented and reason giving qualities. These immediately reaching the best viable solution rather than academically discussing the components of the problem. The third component concerns qualities or attributes of the group. Though other deliberative standards are considered ne cessary for procedural legitimacy, and may even help produce conditions conducive to collective wisdom, they are unnecessary from the purely epistemic perspective and have been omitted Epistemic Requirements For Deliberation This section outlines in theor y 3 what deliberation would have to look like in order to produce the kinds of good decisions we are concerned with. I assume a necessary precondition of epistemic deliberation is the perception of importance. Deliberation can be a slow, exhaustive, and som etimes arduous process. It requires time in a field where seriously if we think the outcomes are trivial or do not matter. Deliberators must therefore have good reason to beli eve their time is worth spending. They must have reason for making good decisions and finding smart solutions to problems. This means that problems must be significant, the potential outcomes being deliberated must vary to 3 The
113 some degree, and that the group h as common reason to prefer some potential outcomes or solutions over others. Unfortunately, this precondition is often unmet in much of the decision making that takes place in shared governance practices today. A recent survey (Finkelstein, Ju, and Cummings 2011) finds that faculty do not consider themselves very influential outside of their home departments. Only ten percent reported having influence at the college or school level, and only three percent felt they were influential at the institution al level ( 2011: 209 10). Despite this dearth of influence, faculty in that same survey reported spending on average, 7.55 hours per week to administrative activities. This means faculty give almost one day a week to governance activities that they believe have little or no impact on campus decisions. A consequence is that many offer trivial discuss ion of campus problems becau se they have full awareness that their efforts will receive little if any attention by decision makers (Duderstadt 2004: 149, 2009: 334 5) (2004) We can imagine a typical situation in which faculty perceive themselves as powerless or without influence. 4 Suppose, for instance, a case in which the faculty senate has the r ight to be heard and consulted o n academic, 4 See Chapter 1 for a fuller discussion of the problems of shared governance, including the question of whether senates are made powerless because they are ineffective or made ineffective because they become powerless.
114 campus wide operations and projects) After months of research and time spent debating the relative merits of options A, B, C, and D, the senate committee(s) recommend s option B. However, the administrative decision makers choose option A, the one they had advocated all along and which the faculty senators knew was most likely to be chosen This kind of interaction is all too common in divided governance schemas. An unfortunate and extreme case comes from the 2000 2001 University of Notre Dame faculty senate, a purely advisory body without any decision making authority. Frustrated by its po werlessness, it voted to dissolve itself, this vote perceived what they view as an insufficient voice in the decision (Kellogg 2001) How ever, without any proper authority, senators were told they had no right to disband and were thus commanded to get back to work. The faculty senate was so powerless it could not even stop doing what it considered pointless work. These senates likely do not meet our importance precondition. Here I am interested in how we can build an improved theoretical decision making model that avoids these problems. Deliberation as Decision oriented Debate The Macalester faculty convened their deliberations in order to reach a decision as to whether or not they should vote to overturn the bid to close the Russian Studies department. This gave their meetings a central purpose and focus that determined what kind of speech was and was not helpful for reaching their decision This meant, far from being a place to air grievances against the administration or its policies, the whole endeavor was oriented toward reaching a final and conclusive decision as to whether the program should be saved.
115 If faculty senates or any similar bodies are to avoid devolving into the ineffectual and unmotivated bodies Duderstadt calls make meaningful contributions. This means individual collaborators must be able to mutually find the kinds of suggestions, recommendations, and policy proposals that making as well as, if not better than, the expertocratic alternative must be to (Thompson 2008: 502 ) when we want to find the kinds of decisions that could be implemented or put into action. In contrast, aggregative measures like polling or voting average independent pieces of information into one majority conclusio n. While this may be fair, there is no guarantee majority opinion will be a good one (Urfalino 2012) Deliberation, however, makes individuals merge their many opinions and information into one mutual opinion. This requires weighing and evaluating informat ion rather than treating every argument as equally valuable. From an epistemic perspective, this orientation provides the means to find a smart solution. This orientation does not mean that communication only counts as deliberation if a binding decision re sults from it. Whether or not that decision is ultimately implemented may be crucial to the success or failure of the issue at hand, but it does not affect the quality of the decision making process. When many work toward one solution or answer, they neces sarily look for a decision that all group members are at least willing to consent to, if not unanimously agree upon. What matters more is that everyone involved is willing to accept it. In short, smart group decisions begin with the will to a binding decis ion. Thus in the way we define it here, deliberation is first and foremost decision making communication.
116 A decision orientation requires non trivial common concerns. Tetlock (1983, 1985, Lerner and Tetlock 1999) has found that when the stakes are high, de cision makers are more likely to thoroughly examine all sides of an argument rather than just reinforce their own preferences or opinions. Evidence of this behavior also suggests why presidential candidate popularity rarely correlates with election outcome s (Silver 2015) When a primary or presidential election is months or even years away, brash or bold candidates tend to enjoy favorable numbers in the polls. However, once elections near, the need to make an important decision becomes real and felt. Voters (Wattenberg 1991, 2004) Subsequently, electability (how likely other voters are to also choose the candidate) and suitability for office (would this candidate do the job well) b ecome the criteria for candidate selection. That is, voters start to think differently about what makes a good candidate and thus whom to choose. They analyze with more scrutiny. Similarly, in deliberative groups where the stakes are higher for some than f or others, those with more invested are more likely to try and sway the conversation in their favor, and the less invested are more likely to be convinced by persuasion and talking than by good reasons (Sunstein 2002) Groups should therefore deliberate so mewhere between these two polls: in cases where important decisions have to be made but absent the incentive for individual group members to alter the outcome in favor of his/her own preferences. The Macalester case features such non trivial common concer ns. Here, the stakes were high enough to give faculty members pause and consider all sides of the situation equally. Fewer programs could have meant more funding for the remaining
117 departments. But because the decision would set a precedent and help determi ne future faculty relations, they were also hesitant cannibalize a program without good reason. It was obvious this decision would have significant and long ranging consequences for the college, even if they were unknown. This fact was enough to not only m otivate faculty to organize extensive deliberation in their off time, but also to evenly and thoroughly consider which decision would have proved best for the entire school. The epistemic property of a decision orientation can be seen in another instance o f departmental closure: In 1992, the University of Maryland at College Park was facing resident and provost decided to close a number of academic departments, programs, and units. While the process was initiated by administrative elites, faculty members largely determined the final decision. They began by organizing a committee to develop th e dean to prepare a list of potential cuts using these criteria (Eckel 2000: 21) Using these findings as a starting point, en organized a series of campus wide subcommittees that were asked, not to address the validity of the recommendations, but rather to investigate the implications of closing the unit on the campus and the state Each subcommittee was comprised of faculty from various departments within and outside of the college and at least one faculty member from the targeted unit. Approximately 120 faculty became involved in the discontinuance process through these subcommittees, equivalent to 10% of Maryland College Pa time, tenure track faculty (2000: 21; my emphasis). As a result, faculty changed the initial list, both adding new units and removing some that were initially proposed for closure. In this case, the subcommittees were able to
118 identify the o ptimal set of closures because their mission focused on identifying the implications of closing or retaining each department or program. In other words, they were tasked with making the final decision regarding what was best for the college as a whole. 5 No tice also that this decision could not have been sufficiently reached with just a handful of people. In order to know how a department would be affected by a partial shutdown, those directly affected were necessary for making a smart decision. In situatio ns without this orientation, there is no reason to critically and systematically evaluate information or bring it together into a coherent whole. For (1995, 2009) participants deliberate in order to learn more ab out a popular issue and to develop better, more informed preferences. When this works, the participants walk away as more educated and critically aware voters. However, critics like Richardson (2010) argue that without forcing participants to merge their p references (e.g. go to war, cut taxes) into a cohesive whole, the process (Cf. Fishkin 2010) Others argue that without the need to critically evaluate information, the lack of decision orientation can diminish the quality of the very preferences they are designed to improve (Thompson 2008: 503). In both of these cases, a lack of decision orientation is blamed for a reduction in overall knowledge and the settling on suboptimal policy or candidate preferences. In sum, we have g ood reason to argue epistemic value is produced by decision oriented debate over matters of common concern. 5 From this perspective, whether or not this committee and its subcommittees actually possessed the authority to enact its decisions is inconsequential to a decision orientation. In this case, Eckel reports the final closure recommendatio ns required approval by the faculty senate, then the president, and finally the board of trustees.
119 Deliberation as Reason giving Debate A decision orientation is also important because it constrains the kinds of information presented during delibe ration. In the Macalester case, the impending vote (and its binding decision) required that speech was restricted to the kinds of things that would help the group reach a decision. Because this vote would likely set a precedent for future program closings, faculty had to consider what was in the best interest of the school as a whole. Thus a second feature of the Macalester deliberations was the serious consideration of all reasons for and against closing the Russian Studies program. Pure procedural delibe ration concerns the opportunity to participate equally in making a decision. Outcomes are considered fair insofar as everyone involved is able to contribute to the conversation. However, if I want or need to convince the group that one outcome or measure i s better (i.e. more epistemically warranted) than others, I need to be able to explain why that is the case and why all preferences should merge decision has to have We give reasons in order to determine (i) what is of common concern and/or (ii) how to best achieve desired outcomes. The first case is defined by uncertainty, when what is to be done cannot be determi ned because a best or desired outcome is yet aims, and ideals that comprise the common good are those that survive deliberation interests that, on public reflection, we t hink it legitimate to appeal to in making claims on problem solving process whereby a group tries to maximize the chances of achieving its
120 desired outcome(s). For instance, i n Porto Alegre, Brazil ordinary citizens convene annually to tell the state what publicly funded projects they believe will encourage the common intere st and protect quality of life (Fung and Wright 2003) In Chapter 2, we used the example of air pollution regulations (Anderson 2006: 10) to argue there may be good reason to rank outcomes that are equally just by procedural standards. We would not prefer legislation that is prohibitively expensive, does not make a noticeable difference for 100 years, or requ ires everyone to stop driving cars, if we thought some better solution might be possible. In these cases, we would deliberate over the preference of outcomes in order to determine which one is best. In some cases, both (i) and (ii) can occur. In 2004, the satisfactorily and to also propose changes necessary to correct any problems or imbalances it perceived (Warren and Pearse 2008). For this reason, Manin (2005) describes deliberation in terms of debate rather than discussion. One way to understand the difference is to look at the distinct cognitive tasks that take place during each: We deliberate about a given course of acti on when we suspect that there might be reasons against it as well as reasons for it. If we did not think that there might be, at least potentially, reasons for not doing X alongside reasons for doing it, we would use reason in a different way. We would see k to prove, or at least to establish, that X is the right course of action by supplying solid argument(s) for it. We would not actively seek counterarguments. It is the seeking, and the weighing of pros and cons that distinguishes deliberation from other f orms of reasoning (Manin 2005: 15; my emphasis).
121 For our purposes, and following Manin, we can classify as discussion any mutual communication that lacks this requirement. 6 This is not to say that other forms of communication are not supportive of epistemic and/or democratic value (Delli Carpini, Cook, and Jacobs 2004, Walsh 2007) But our purpose here is to formulate a theoretical model of epistemic deliberation and so we m ust omit other communicative or speech acts. In the practice of deliberative democracy, reason giving has traditionally served a legitimating function (Benhabib 1996b, Bohman 1996, Bohman and Rehg 1997, Cohen 1989, 2007, Dryzek 2001, Elster 1998, Gutmann a nd Thompson 1996, 2004, Habermas 1996, Manin 1987) On this account, we give reasons to justify to each other what laws representatives ) test their interests and reasons in a public forum before they decide. The deliberative process forces citizens to justify their decisions and opinions by appealing to common interests or by arguing in ensuing collective decision should in some sense be justified by public reasons that is, reasons that are generally convincing to everyone participating in the process of deliberation (Bohman 1996: 5). If a particular use of public doing it should be obvious to everyone upon hearing them (or at the very least should win out over competing reasons). The people can therefore justly demand of legislators and of each other th at they give reasons as to why the state would impose some rule on them. The people are then, at least theoretically, able to deliberate over those reasons 6 mean perpetual talk without action, and assumes that a label for what faculty senates should avoid becoming, I mean it in the sense Manin has here de scribed hopefully forgivable) given the comparison.
122 in order to decide ( viz through a vote) whether or not they ultimately support them. For this reaso n, Habermas framed legitimacy in terms of rational discourse over common interests aimed at the regulative ideal of consensus. This ideal is only possible, he avers, if interlocutors are determined to provide reasons why a certain proposition is or is not in the service of the common interest (Habermas 1975: 107 8; See also Cohen 1989: 22). In short, legitimacy hinges on discourse comprised of the kinds of reasons giving has historically served as the procedural and normative basis of political legitimacy. However, mutual and deliberative reason giving also serves a cognitive function, and this is what is of interest to epistemic democrats. To mutually find and evaluate reasons for and ag ainst a proposition or set of propositions in order to reach a decision, we engage in a particular type of thinking. Deliberation must therefore be comprised of rule guided interpersonal speech acts and the comparative weighing of reasons. 7 If we are to co nclude that deliberation makes for better information gathering and smarter decisions, we must first show that 8 this cognitive process allows groups to avoid and paves t he way for eventual collective intelligence and good decisions (shown in the next section). A corollary to reason evaluation is sensitivity to the persuasiveness of good information, or what might be called reflectiveness on the part of deliberators. As 7 nothing else but a weighing, as it were in scales, the conveniencies, and inconveniencies of the fact we are attempting; whe re, that which is more weighty, doth necessarily according to i (1642/1983: 63) 8 ed for emphasis.
123 in dividual deliberators seriously consider every side of a proposition or set of propositions they must be willing to change their positions as new and better reasons are given. This requirement demands that deliberators be sensitive to what Habermas (1996: Dryzek (2000) refers to rhetoric, though some theorists argue there is a place for rhetoric (Bohman 1996, Cha mbers 2009, Cohen and Rogers 2003, Dryzek 2010, Young 2000) Nor is it merely debate o f these pros and cons in order to understand if (and why) a certain outcome is preferable to another or why a decision (or solution, policy, etc) would achieve a mutually preferred outcome. Reason giving enables deliberators to put their information (prefe rences, opinions, facts, knowledge, interests) into a debatable and evaluate form. n 2). floor that blocked the renewal of a patent on a Confederate Flag insignia (Guttman and Thompson 1996: 135). Believing that her valid reason s would not be seriously considered unless she forced attention on them, she argued for hours, threatening to filibuster the bill. Braun, at the time the only black senator, used personal and subjective speech rather than di spassionate rational statements to convey the normative, ethical, and social importance of blocking this renewal. This kind of speech, Guttman and
124 Thompson argue, was justified because her position was both reasonable and neglected by her peers. Relatively extreme (yet non coercive) spee ch provided a vehicle for a reason been impassioned, but it made substantive points, to which members could respond in the way Dryzek describes above. In the end, she won that battle by providing a merited and persuasive argument. This kind of speech can also be seen in the Macalester case. Because the closure decision hinged on student demand for the program, bringing in and listening to both current and potential Russian Studies students became cr ucial to making a good decision Here, student opinions became reasons to support or reject the motion. Students increasingly voiced their willingness to enroll if on ly there was a wide scope of courses, which itself provided a persuasive argument in favor of saving the program. Both of these cases show that subjective (as opposed to strictly speaking rational) debate has a valid place in deliberation if it is used in the service of communicating reasons that everyone can agree to. In short, the reason giving e amenable to evaluation, and (ii) sensitive to merited persuasion. In fact, recent work in cognitive psychology suggests that reasoning is a social and deliberative, rather than internal and personal, evolutionary artifact (Mercier and Landemore 2012, Mercier and Sperber 2011, 2012, Sperber et al 2010, Sperber and Mercier 2012, Cf. Bahrani et al 2010, Koriat 2012) According to the argumentative
125 e others 2012: 369). Humans have always relied on information to make decisions. The more we have, the better our decisions will be. Most of this information comes from othe rs. This not only makes us social creatures, it also makes interpersonal communication invaluable. In this regard, human communication is similar to that of many other m evolved a keen awareness that others can manipulate us by giving self serving, misleading, dangerous, or simply unnecessary information. Thus, theorists argue, humans needed also guarding against the inherent vulnerability of relying on them (Sperber et al 2010). The up shot of the argumentative theory is that humans developed reasoning convince or persuade others in kind. Reasoning, in other words, evolved from human interdependence. Conside r a case in which you have bad information, say you are about to consume a poisonous berry you think is harmless. You are my superior so I cannot just order you to drop it, and food is scarce so you may think I am tricking you to get your berry. How am I t o properly convey my information to you? In this case, I need to develop a persuasive argument in order not just to say, but to convince you, that eating that berry may kill you. This is the kind of situation in which argumentative theorists claim reasonin g was born.
126 This cognitive account supports the definition held here that any activity is considered deliberative to the extent that it involves finding and evaluating the reasons own is relatively limited. As Mercier and Landemore (2012) show, reasoning requires information that allows each party to evaluate its own argument from the perspective of person A uses reasoning to make an argument from opinion a ; person B uses reasoning to examine this argument from opinion b ; person B then uses reasoning to create an argument that relates to the previous argument often a counterargument from opinion b ; A uses reasoning to examine this argument from opinion a (2012: 246). Here, person B uses opinion b as a sort of touchstone against which to evaluate opinion a The result of that ev aluation is then relayed to person A, who similarly judges that information against opinion a in which each party knows that its opinions have been heard and weighed by the other parties. This is on ly possible to the extent that communication is mutual and that each speaker is also willing to listen empathically to the reasons given by others For this reason, we cannot look to aggregative measures, like voting or polling, as having the same potential for epistemic value, because they do not have this evaluative mechanism. While reasoning can take place outside of a deliberative context, one p arty can only deliberate to the extent that it evaluates the reasons for and against its own opinions and those of all other parties. So far we have examined the requirements of the deliberative process imposed by the standard of epistemic value. If we ar e correct, then we have good reason to believe that talking together helps create better or more accurate information. We have
127 yet to address how these requirements guard against the problems inherent in deliberative groups. Perhaps the most significant pr oblem that can impede the epistemic value of deliberation is confirmation bias 9 whereby individuals are inclined to interpret facts and evidence so as to reinforce already held thoughts, beliefs, cares, or commitments (Hergovich, Schott, and Burger 2010, Koehler 1993, Nickerson 1998) It is a sort of tendency toward the cognitive status quo. Confirmation bias helps explain phenomena like group polarization. For instance, Sunstein (2002: 3 4) argues that rd a more extreme point in when our tendency toward confirmation bias is reinforced by limited or no exposure to dissimilar viewpoints. Discussion of a common belief enc ourages the group to double down on that belief. However, research in political psychology finds participants in group alternative outcomes aside from the one each member individ ually preferred (Hirt and Markman 1995) process involved careful consideration and led to more realistic understandings of the topics. The argumentative theory explains these findings in te rms of partial versus full application of the reasoning process: confirmation bias results from reasoning in one direction (reasons for an already held belief) and evaluating only information that rizing groups did not deliberate in the way defined here (Cf. Manin 2005), and therefore we should not be surprised to see in them a lack of collective wisdom. 9
128 If we are to believe the argumentative theory of reasoning, then not only do these requirements allow for the possibility of creating good decisions, they also protect against some of the major problems that impede groups from making them. Reasons are, in this sense, the language of knowledge. We are now in a position to understand why Goodin (2008: 4) people take the positions they do, and sometimes that gives us more reason to in getting all the relevant a shared beliefs and positions. Rel evant Diversity, Democratic P articipation solving leads to better solutions, we still have ceteris paribus for a few experts to make decisions? Consideri ng the well documented failures of shared governance practices, we have no reason to think that deliberative groups are necessarily better decision making units. However, we have so far defined deliberation in a way that is not commonly observed in shared governance. This should, at the very least, allow us to go on the offensive when addressing this question: Why not larger groups? Consider, in contrast, an expertocratic committee or task force, the members of which share the belief that a school should be run according to free market principles. They may tend to agree on one proposition without equally weighing reasons for and against other competing propositions. If they are correct in their thinking, this makes for
129 a quick and efficient decision making p rocess. If not, however, confirmation bias can cause the cognitive myopia that leads to premature consensus. In this section I turn to another problem: even if the group has a decision orientation and debates reasons in the way outlined above, what is to say they have sufficient information for making an informed, optimal decision? As we saw in Chapter 3 diversity theory suggests the best way to guarantee this requirement is to include enough perspectives that cognitive diversity can, at the very least, i dentify and discount errors in judgment that could lead to premature consensus. Thus, we find the previous two requirements necessary but insufficient for an epistemic theory of deliberation. It is not enough that a group works toward finding a binding dec ision or that it seeks the best reasons for adopting one choice over another. It must also make sure that it has all the cognitive resources that it needs to make the best decision. So, I argue here, there is no reason to assume by default a small elite gr oup will outperform a more numerous option. In practice, deliberation ends when the group believes it has achieved the best solution possible or that it is able to find. This means that the quality of outcome relies heavily on the perceptions of the group and the standards they set for evaluating that outcome A group (of any size) achieves premature consensus when it is unable to locate all the reasons in favor of the best solution. If that group is also cognitively homogenous it may mistake unanimity wit h quality of decision; that is, it may assume its reasons are good ones because everyone agrees on them. 10 Theoretically, the best way to avoid the epistemic errors associated with cognitive homogeneity is to maximize 10 The epistemic role of unanimity is taken up at length in Chapter 5.
130 the number of viewpoints that participa nts bring to the debate. One way to accomplish this is through a sufficient number of diverse thinkers. Thus, Bohman (2006: 179) argues: [T]he relevant aspect of diversity that is necessary for improving the process of deliberation is not the pool of reaso ns as such but the availability of the perspectives that inform these reasons and give them their cogency. The pool of reasons can be increased even while still leaving out relevant perspectives (his emphasis). Perspectives are the viewpoints through which we look at, process, and make sense of the world. When a problem is viewed from many perspectives, it allows a more complete understanding, if not of the perfect solution then at least of the nature of the problem itself. In another sense, having many per spectives brings more information to the table when problem solvers are finding a solution. Increasing perspectival diversity therefore makes it easier to avoid premature consensus. This creates both an efficient an d elegant deliberative theory. With Page (2007), w e can call a sufficient number of perspectives cognitive diversity 11 Consider again the case of program closures at the University of Maryland. Final recommendations came only after multiple rounds of committee and subcommittee meetings designed to get feedback at the ground level. The discontinuance committee recognized that formulas for making such important decisions are only as good as the information that is fed into them. By continuously consulting with and reformulating the list of potentia l closures, the group was able to get access to the kind of information that a high level committee did not originally have, viz. what implications these closures would have on everyday operations at the departmental, program, and state levels. 11
131 The second diversity in public deliberation, and for this reason he is led to recommend maximizing such perspect ives interact and inform each other, and in that way open up deliberation, the sheer number of topics that could come up means we cannot say ahead of time which per power that tend to silence minority opinions and favor the status quo (Sanders 1997) (Bo hman 2006: 176) leads him to a rule of maximal inclusion with the hope that a sufficient number of relevant perspectives will be present if all or many are able to threatened by insufficient diversity and the effects of power on perspectival representation, both of which may be prevented with inclusive participation. However, our model differs in that we are attempting to implement a system of deliberative decision making rather than a theory of deliberative democracy. Bohman is concerned with determining what is the common good whereas we are concerned with such things as policy proposals. There is no denying these issues should rely on expertise and technical knowledge. My point however, is to identify who has relevant knowledge. Expertocrats argue one small elite group of maybe half a dozen or fewer administrators possesses the expertise to solve all problems. Our model, however, assumes a distribution of knowledge that necessa rily changes depending on the
132 other words, lots of different opinions, viewpoints, perspectives, and skill sets will help solve the problems a decision making body is likely to face. Relevant expertise is determined by need for cognitive diversity. In the Macalester case, this was satisfied by including current and potential students whose opinions, after all, were the subject of the closure in the first place. Had t he decision been of, say, financial solvency, then perhaps accountants would have provided sufficient expertise to decide whether or how to save the department. But because the decision hinged on the wants and desires of nyone with information or knowledge about those wants and desires, viz. the students themselves. My model suggests whoever can provide that information should be included in the problem solving process. Limiting Inclusion and Diversity To take stock of our argument thus far, we have seen that groups are requir ement above. The question still remains whether or not this is democratic. Before proceeding, a helpful distinction to make is between rules of inclusion and rules of participation. Inclusion rules determine who can be considered a deliberator, the questio n we just addressed. A deliberative group is democratic to the extent that its inclusion rules afford everyone the opportunity to participate. Participation rules govern the interactions among participants within the deliberating group. A group has democra tic participation to the extent it recognizes members as equals. Diversity theory gives us good reason to favor restrictions on inclusion. On one hand, the perspectives and heuristics brought to the table must be able to converge in a way that produces co ntinuous improvement to given proposals until the best possible
133 solution is found. This suggests the need for a large number of people. Landemore, e.g., says that in matters of national elections, policymaking, or constitution writing, the requirements of cognitive diversity will be met by including large numbers of citizens and allowing decisions to be made at the national scale. Indeed she renames this the Numbers Trump Ability Theorem (2013: 104, 2012) based on the need to handle matters of public interest, matters where we are already inclined to side with the procedural argument that everyone deserves the respect of determining the decisions that affect them. On the other hand, if we ignore the need for relevant diver sity, then we risk adding voices that simply do not contribute any information of value to the deliberation or that echo already represented perspectives. This threatens epistemic value by bogging down the process with too much information to sort through. As noted above, Page argues cognitive diversity should be the factor limiting inclusion: adding a gifted poet to a cancer research team will likely not improve its endeavors (2007: 164). Diversity in problem solving is beneficial precisely because new inf found. On this account, extraneous information simply falls by the wayside; that is, non relevant points of view do not get us very much. But de liberators must be able to figure out which information is relevant or not. In sum, we have an argument against democratic inclusion if the task at hand is not benefitted from the simple addition of numbers, and an argument for it if the subject is the co mmon interest of all. This suggests a group should increase diversity until doing so no longer contributes to improved solution finding. In other words, a solution is likely
134 the best when increasing cognitive diversity no longer improves upon that solution Thus, restricting inclusion is not problematic, epistemically speaking, if enough perspectives are represented to achieve cognitive diversity. The need for relevant but bounded diversity should establish the minimum number of group members (the floor), w hile the need for an exhaustive search and evaluation of reasons should set the maximum (the ceiling). This finding is in line with both the literature on deliberative democratic theory and with the common sense understanding that too many voices at once s imply is not conducive to good conversation. However, given that finding an ideal number may very well prove impossible, it seems best to side with greater rather than fewer contributors. To be sure, a too large group may prove impractical by providing t oo much information, which may in turn inhibit epistemic value. But diversity is so crucial that w ithout sufficient numbers, there is no guarantee that all relevant perspectives will be represented and thus no guarantee all sides in the debate will be cons idered. Plus, it recognizes that there is an important place for procedural values like inclusion and that, in the end, even a more efficient democracy is still likely to be messy. emature group emergent property. T enough to produce relevantly diverse thinking and perspectival representatio n Because this chapter lays out a theory of epistemic deliberation, the question of how exactly we determine this number has been intentionally unanswered (though we return to it in Chapter 6). For instance, in the Macalester case, the group was insuffici ently large until
135 it included students. T heoretically a handful of students could provide more cognitive resources than even a large number of non students would be able to. This may sound counterintuitive, given that diversity theory claims numbers trump ability. But these students were the source of good information, not the problem solvers. In this case, they provided the raw or brute data that was then evaluated through deliberation. Once sufficient information was available, diversity theory says we c an expect to see better decisions result from the large open forums than from a handful of administrative elites. While the latter may find very good solutions in some cases or the best solutions in a limited number of cases. But, according to diversity th eory, it is unlikely that they will find the best solutions in most cases. Conclusion Recall from Chapter 1 that whereas some elements of democratic deliberation like giving public reasons or respecting the opinions of others are constitutive of the delibe rative process, epistemically warranted decisions are the outcome of that process. This chapter has considered the theoretical requirements, standards, and venues that are necessary and sufficient for deliberation to produce these decisions To this end, I have not been concerned with the question of whether or not such a conception is extant or even possible. It is sufficient for our purposes to flesh out what such a conception would have to look like before we could ever hope to evaluate its plausibility. As opposed to a minimal definition, which sees deliberation as merely the mutual consideration of reasons, I have argued that to meet this standard deliberation must involve a group mutually finding and evaluating reasons for and against a proposition or set of propositions in order to reach a decision. While not requiring unanimity, the presence of a decision signifies an evaluation of reasons was thorough
136 and exhaustive enough to at least earn tacit consent, if not unanimous approval, of all group member s. To reach this point, a group must be large enough to represent all relevant arguments for and against the issue at hand. Without sufficient numbers, there is no guarantee that all relevant perspectives will be represented and thus no guarantee all sides in the debate will be considered. However, a too large group may prove impractical by providing too much information, which may in turn inhibit epistemic value. What matters is the presence of all relevant perspectives, not the inclusion of large numbers. Thus the need for relevant diversity should establish the minimum number of group members (the floor), while the need for an exhaustive search and evaluation of reasons should set the maximum (the ceiling). But deliberation should be democratic in the way those members communicate, share and evaluate information, and reach a final decision. Whereas this chapter is concerned with who should talk and how, the next is concerned with when talking should come to a close While the epistemic standard may produc e better results, consensus nonetheless plays a significant role in concluding the decision making process. Next I discuss when whether or not a majority rule can offer the same epistemic benefits (e.g. certainty) as a unanimity rule.
137 CHAPTER 5 THE SILENT MAJORITY RULE: RETHINKING THE EPISTEMIC VALUE OF NON UNANIMOUS DECISION RULES IN COLLECTIVE DECISION MAKING Introduction When groups are making collective decisions, does unanimous agreement offer an epistemic advantage over majority agreement? Unanimity certainly has a normative appeal; what better reason for choosing one option over its alternatives than full agreement by all concerned parties? When possible, we should always hope to find the best solutions together and completely. And yet, des pite this appeal, unanimity is seldom, if ever, observed in real world decision making. This unlikelihood is enough for many deliberative theorists to dismiss it as an unrealistic ideal and instead favor more plausible decision making procedures like barga ining or negotiation. However, some defend unanimity as the best indicator or signal of optimal solutions, choices, or decisions. Diversity theorists fall into this category. In this chapter I examine their defense of unanimity. For epistemic problems i.e problems for which there is a best or right solution they contend that only full agreement among deliberators signifies certainty that the right solution has been found. From this perspective, disagreement and a dissenting minority are explained as failu res to realize such a solution. This unanimity assumption, however, places a heavy cognitive burden on individual deliberators. If the condition of sufficient difficulty is met, then by definition these deliberators are dealing with problems that they cann ot individually understand. These kinds of problems can only be conceived of and solved collectively. And yet diversity theory assumes that included in these individual judgments is a realization of the optimality of the best solution. Problem solving deli beration that ends in unanimity may be asking too much.
138 an ideal stopping rule for deliberation. First, I argue that although strength in numbers is normatively and epistemi cally appealing, unanimous agreement is unlikely, unnecessary, and should ultimately be rejected for reasons intrinsic to diversity theory itself. Then, I show that under certain conditions, majority agreement offers the same epistemic advantages as unanim ity. Finally, I offer a theoretical defense of majority rule. Recall from Chapter 3 that the major expertocratic criticism leveled against collective decision too slow, ineffi cient, and ineffective to meet the ever increasing and rapidly shifting demands of the twenty first century academic landscape. In Chapter 4 we looked at the qualities of deliberation that tend to produce the optimal solutions to epistemic problems. We not ed that those qualities reason giving, decision oriented debate by a relevantly diverse group of problem solvers using democratic participation rules can be used to judge the quality of deliberation itself. Here, I put those insights into practice in a way that addresses the expertocratic critique of collective decision making. To do this, I use that conception of deliberation as the basis for an epistemic defense of majority consent: If we can devise a process that tends to produce the right or best answer s, solutions, decisions, or choices, then a dissenting minority has good reason to unanimity retains its place as
139 but non unanimous decision rules like a majority rule can also be effective at solving e pistemic problems. The Epistemic Defense of Unanimity in Broader Context In the study of deliberative democracy, attitudes toward the epistemic value of unanimous consent are mixed. In a recent article, Bchtiger et al (2010) identify a basic bifurcation i n deliberative democratic theory. The two types of deliberation they is rooted in the Habermasian logic of communicati ve action, and embodies the idea of rational discourse, focuses on deliberative intent and the related distinction between communicative and strategic action, and has a strong procedural component. In this view, deliberation implies a systematic process wh erein actors tell the truth, justify their positions extensively, and are willing to yield to the force of the better argument. The ultimate goal of type I deliberation is to re ach understanding, or consensus (2010: 33). In these early theories, deliberative democracy is considered a theory of political legitimacy (Cohen 1989, Habermas 1975, 1996, Young 1996). Thus, Cohen (1989: 22) f asoned, public arguments that could be the subject of agreement by anyone who heard them. Under these conditions, consensus represents the point at which a group has
140 democratically and correctly determined what is in the public interest. Consensus, in shor t, is the will or spirit of deliberation. 1 consensus is a viable, if ideal, outcome of democratic deliberation. Rather, it is more precise to say consensus is considered a sort of litmus test. For instance, Habermas claims that if our goal is to use the logic of communicative action 2 to determine what is in the public interest, and if under those conditions we are able to reach consensus on a decision. To be sure, un animity is normatively attractive. However, it seems downright impossible in circumstances where the rules and logic of communicative action are not observed, like we often see in the polarized and vitriolic nature of contemporary US lawmaking and politics These cases call not for unanimous consensus but for of the political to the 1 In deliberative democratic theory, consensus is defined and used in various ways. It can be understood to act as a goal, outcome, ideal, or actual stopping rule of deliberation. While Habermas and Cohen explicitly define unanimous consensus as the ideal o f deliberation, others, like Young (1996: 122), define it as the actual goal of deliberative bodies. The importance of each is discussed below. 2 of com munication that is removed from contexts of experience and action and whose structure assures us: that the bracketed validity claims of assertions, recommendations, or warnings are the exclusive object of discussion; that participants, themes and contribut ions are not restricted except with reference to the goal of testing the validity claims in question; that no force except that of the better argument is exercised; and that, as a result, all motives except that of the cooperative search for truth are excl (Habermas 1975: 107 8).
141 forms of discourse, more emphasis on outcomes versus process, and more attention to 33). Accordingl consensus rule in favor of more easily achievable or realistic alternatives like bargaining or negotiation (Estlund 1999, Holzin ger 2004, List 2007, Mansbridge et al 2010, 2012, Peter 2009, W arren and Mansbridge 2013). For instance, Warren and Mansbridge suggest decision making by negotiation is normatively acceptable to the extent it justifies agreements that a majority of parties agree to, so long as they are grounded in 92). In the face of recent congressional inaction and a corresponding government negotiations may still seem out of reach but are nonetheless more plausible than full rational agreement. (Chambers 2009, Mansbridge 2007a) However, some democratic theoris ts regard this shift as symptomatic that deliberative democracy has given in to second best consensus rules, ignoring the best for the good. This is especially trenchant when is a right or best (herein referred to as optimal) solution to that problem. Thus, we say group members have epistemically warranted beliefs for agreeing on one choice over its
142 particular decision, rather than the value of simply making a decision at al In sum, they argue that if an optimal answer, solution, or choice exists, then negotiation or bargaining represents a compromise that will often result in a suboptimal solution. unanimity, claiming it offers epistemological advantages like certainty of locating an optimal solution that a less stringent decision rule cannot. For example, Landemore and Page solving contexts, [unanimous] consensus ret ains a strong diversity theory, they believe that deliberation produces two decision making d solutions, deliberative problem solving can also produce synergies, that is, create new solutions On this account, unanimity offers a sort of epistemic strength in numbers: the optimal solution to an epistemic problem is likely one that all problem solvers unanimously agree on. 3 Thus unanimity, aside from its normative appeal, also provides the certa inty of epistemically warranted belief (at the very least, certainty that we have avoided premature consensus on a suboptimal solution). That unanimity can provide such certainty is not at question. But is it the only decision making rule appropriate for s olving epistemic problems? 3 I follow the authors in describing optimal
143 Necessary Conditions of the Epistemic Defense: Meta consensus, Shared Interests, and Oracle Solutions To understand this defense of unanimity we must also step back and look at three necessary and sufficient conditions. The firs t is perhaps characteristic of all consensual and collective decision making: mutual agreement on the nature of the problem itself. This recognition must precede any attempt to find a solution to that problem. Thus, Niemeyer and Dryzek (2007, Dryzek and Ni emeyer 2006, Niemeyer values, beliefs, and preferences. They identify three types: normative meta consensus epistemic meta consensus preference meta consensus refers to agreement on the issues that bear the most importance and on which any decision sh ould hinge (2007: 504). According to Niemeyer and Dryzek, normative meta consensus would be reached if and when, through deliberation, we mutually come to recognize which values should be represented in our search for a solution. Epistemic meta consensus i s subsequently reached when our agreement on values then informs the kinds of knowledge that have validity and the beliefs that have epistemic value in the problem solving process. Referencing Chapter 4, we can say this type establishes what counts as rele vant expertise on the subject at hand. 4 Finally, if we come to agree on what is problematic about a situation and where we should look to find a workable solution, we should also be able to agree on what that solution should entail (i.e. not the solution i tself but what it 4 Recall that, in the Macalester Col lege case, the decision whether or not to close the Russian Studies major hinged on student demand. Here, the opinions and input of both current and would be Russian Studies undergraduates became valid in making the decision.
144 would have to take into account). Without agreement on these background conditions, we would be working towards different solutions and might never come to consensus. Mutual agreement over the values, beliefs, and preferences that inform our perception of problems is therefore a precondition for collective decision making. 5 The second and related condition is shared interests among deliberators. Mansbridge (1980) finds that consensus as a decision rule is only appropriate for ocratic groups, in which all members share the same interests and jointly pursue the same goal(s). 6 Fundamentally, the primary objective as a group is to achieve mutually preferred outcomes. In these situations, shared interests enable the group to work to ward a mutually preferred solution or outcome, and consensus represents the moment of agreement on that solution or outcome, which concludes the deliberative in interest group politics, where interests either are not shared or are in direct conflict: the theory is founded on the assumption of self interest, better than those of others. Each individ 5 Note that meta consensus is a non epistemic requirement; i.e., it does not imply the group will make the right decision. Rather, it suggests they will have to agree on background conditions if they are ever to agree on a solution, but this does not mean the solution is guaranteed to be the right one. I could convince you that tax cuts on the extremely wealthy will allow their preserved wealth to trickle down to the masses and thus benefit society as a whole. That does not make us right, though we would now agree on the policy that we want to create. 6 typology is intended to apply from that size a ll the way up to the polity.
145 the equal protection of interests, guaranteed by the equal distribution of pow er through the vote (1980: 8). to face e the same interests, as they do in a unitary democracy, they do not need equal power to protect their interest against one may be justified to allow those with more e xpertise, experience, or wisdom to have greater say or influence than others who are equally concerned but cannot contribute as much. Ideally, the unitary democracy is an apt venue for consensus building because it is defined by collegial problem solving r ather than competitive struggle (1980: 174). These first two background conditions yield what Landemore and Page call nature of the problem. Upon achieving evaluation consen sus, we are more likely to agree on the solution that will achieve a mutually preferred outcome (what they call that shared background conditions encourage agreement and su ggest reasons why this may be the case (Austen Smith and Feddersen 2006, Coughlan 2000, Hobson and Niemeyer 2011, Lo, et al 2013, Niemeyer 2011, Warren and Pearse 2008). For instance, Coughlan (2000) finds that a deliberative group or committee (e.g. jury) is more likely to share full information when its members know they share interests than more, Austen Smith and Fedderse n (2006) find that when group members either do not know if they share mutual interests (what is ca lled preference uncertainty), or know they have dissimilar interests, they are more
146 likely to lie or withhold information in order to sway the outcome according to their own preferences (Cf. Schulz Hardt et al 2006, Stasser and Titus 1985). This suggests, as Mansbridge asserts, that conflicts of interest are not conducive to consensus building. However, when groups want to find common ground, they generally achieve positive outcomes. Thus, even when meta consensus is not found, reciprocal conversation tends consensus more likely (Hobson and Niemeyer 2011, Lo, et al 2013). The third condition is a self evident best solution. 7 Often times, we may not be sure which solution will produ ce our desired outcome, or which course of action will be deliberators must know or assume as been brought up or evident. Under this epistemological condition consensus represents not hing less than the point at which deliberation has located the optimal solution. Problems with a precise or discrete answer a calculus problem, the meet the oracle condition. But so do cases where each [individual group member] must also rank alternative y above alternative x. Individuals need not k 7 Note that this implies epistemic problems of the sort defined earlier.
147 counts as an oracle. The oracle condition entails two assumptions. The first is that perfect information exists to find the optimal solution. This information provides the means to finding a mutually preferred outcome but is distributed widely and unevenly among deliberators. c reating a whole more intelligent than the sum of its parts. Thus, the assumption of the availability of perfect information allows diversity theorists to demand ahead of time that unanimity is the only appropriate stopping rule. The second assumption is t hat disagreement is the product of cognitive error. you asked this to a friend, she would either reason correctly and arrive at the right answer or not. If she gave the probably not rethink your answer but instead conclude that she has calculated incorrectly or misheard the question. The oracle condition assumes that all difficult problems with optimal solutions can be t reated in a similar way, and that all deliberators are considered epistemic peers 8 capable of finding such a solution. Thus we demand a unanimous stopping rule only by rejecting wholesale the idea of reasonable disagreement. A dissenting minority is consid ered either wrong or irrational. Within these conditions agreement on background conditions (meta consensus), shared interests, and a self evident right or best solution (oracle) 8 their abilities and dispositions relevant to interpr Epistemic peers are, ceteris paribus equally likely of finding the right (or best) answer.
148 Landemore and Page argue unanimous consensus should be held as the ideal outcome stopping rule woul d indeed, ideally, prolong the deliberation to the point where the true or best answer has been identified and agreed upon. Ending deliberation prior to 7). Thus, they ar gue, unanimity offers problem solvers an epistemic advantage over negotiating, which compromises on the best solution in the name of agreement. Consequently, they are skeptical of less demanding decision rules like a majority rule. Because the oracle condi tion means the best answer is self evident, a non unanimous majority means that either the optimal solution has not been found or some of the problem solvers have committed cognitive errors that prevent them from realizing it. Importantly, for epistemic pr (Fuerstein 2014: 282), meaning that agreement is not the point of deliberation. That locating the optimal solution: contestation [like deliberative arguing] has a point only if, of all the possible agreements to be had, only some are properly wanted, and some properly wanted more than others. But if it is only agreement that we are after, then that premise is false, since sameness of belief is a property equally well instantiated at any possible point of convergence (2014: 288). That is, we can either prioritize finding the best answer or finding agreement. Any lunches comprised of ice cream, donuts, and soda is one way to guarantee that students receive enough calories, but even unanimous agreement by the school board on such a solution would not make it a g ood one. So the goal of deliberation must be
149 finding the best answer. Once found, recognition of its superiority will then be the consensus of making the right choice(s). This mean s unanimity should be considered a deliberative ideal and a stopping rule but deliberators should not directly pursue it as a goal. Problems with Unanimity democratic theory. Ho impossible to achieve. I am largely sympathetic to this epistemic defense. However, I contend that diversity theorists conflate unanimity and certainty. It is not unanimity itself that is valuable but the assurance of epistemological certainty that it presupposes. In other words, it is problematic to assume that we can have certainty of the optimal solution to a n epistemic problem only when 9 we reach full rational agreement. Consequently, I am not convinced that unanimity offers an epistemic advantage over a majority rule. In this section, I argue the unanimity requirement is both unrealistic and unnecessary for certainty. First, I introduce several objections to the unanimity rule. on such a stopping rule is mistaken for two reasons: (i) its ideal typical deliberator may fail to understand that the right or best sol ution is the right or best, despite contributing to its identification; (ii) it fails to take into account that disagreement can occur even under ideal deliberative conditions. The unanimity assumption becomes epistemically harmful, at least potentially, b y failing to 9
150 recognize these reasons. When majority consent can do the trick, it is actually unanimity ines a majority rule with the epistemological assurance of unanimity. Non epistemic O bjections : Agreement Through C onformity There are plenty of non epistemic reasons why o ne might object to the unanimity requirement. One reason is impracticality and ineff iciency. Elster (1997) argues that full rational agreement ( a la Habermas), while a worthy ideal, is unlikely to be realized in practice because it requires complete unity about what choice should be made and the reasons for choosing it over others. Deliberative democracy arrives at decisions by transforming preferences through argumentation and is therefore an alternative to aggregative theories of democracy. But if it cannot ever fully transform preferences to the point of unanimity, any model with this stopping rule would, at best, only partially transform them and then have to rely on an aggregation of preferences ( viz voting) to reach a final decision (1997: 14). Thus in public applications, unanimity rule deliberation could, at best, only supple ment the very form of political decision making it is designed to replace. Another problem with unanimity requirements is conformity and the pernicious effects it can have on group decision making. Ideally, agreement is the result of rational and reflexiv where agreement becomes the point, rather than the byproduct, of deliberation (Janis and Mann 1 977) In these cases, a minority that disagrees with the prevailing majority
151 psychology has shown that when saddled with a unanimity requirement, even absent any apparent coerci on, people tend to agree with the majority even if they know it is incorrect. For instance, Asch (1956) conducted a now classic social psychology experiment to see how people would react to unanimous majorities who are obviously and verifiably incorrect. T o test this, a subject was placed in a group that was instructed to compare the length of various lines and choose the two of the same length. 10 Unknowingly, the other group members were confederates instructed ahead of time to choose two lines of obviously unequal length and insist that they were identical. Despite knowing the group decision to be wrong, the subject agreed with it almost three quarters dur ing collective deci sion making (Harvey 1974) Many or all group members consent to a collective choice despite personally disagreeing with it, because they believe their preferences conflict with the preferences of the larger group. What encourages conformity in matters of shared academic governance? One reason is the preference for efficiency. If unanimity is required and a conclusive decision is in sight, many may choose not to vocalize opposing reasons or viewpoints that could prevent or delay making it. One professor put the matter this way: [Many] would be happy to give their reasons and know that a timely decision will be made and they may be outvoted. But they will remain silent rather than give their opposing reasons if they think this will lead to endless discussion in pursuit of unanimity. One department meeting I was in, with unanimity as a goal, ended only when the two people holding the opposing point of view had either left the meeting (to deal with family 10 one card is a single line. On the other card are three lines of different lengths. The subject is then asked which line on the second card is identical to the line on the first. In the Asch experiment the sizes of lines on the second card were drastically different in length such that the correct choice was obvious.
152 issues) [and the other] had literally lain her head down on the table in fatigue, and chose not to oppose the motion. simple meeting into a wa r of attrition, in which the final decision is not necessarily the best but the one that was argued the longest and most forcefully. In other cases, conformity can be the product of coercion rather than genuine agreement. For instance, I spoke with one pro fessor whose institution had recently hired a corporate minded vice president for academic affairs from outside academe who was draining academic programs. At this pa rticular school, this vice president (a non faculty position) is a non voting ex officio member of the faculty senate. This became changes were the very issues being discussed Among the faculty, it was generally intimidate and silence dissenters. Unsurprisingly, the result was tense, awkward, silent, and often unproductive meetings. As this prof essor put it, the provost effectively created minds. In cases like this, many people disagree but very little dissent is vocalized or presented. ion requires that everyone agree, a dissenting minority even just one person has unequal power relative to the group (Dahl 1989) This disproportionately favors the status quo by giving a single individual the power to reject any possible changes. This sce ne plays out in shared institutional governance
153 practices where faculty and administration have low levels of trust. For instance, Duderstadt (2009) finds that when faculty believe academic freedom is under attack, they are more likely to play the role of using the faculty senate to stymie or kill proposals from senior administrators. This distrust of policy and policymakers, he argues, leads them act as a sort of guardian of the status quo. In these cases, the ne ed to reach a unified opinion is used against the 11 The problem of conformity is particularly pressing when unanimity determines a nity. In the scientific as well as academic communities, for instance, disagreement is often understood as undermining empirical validity and thus the power of the community as a whole. Beatty and Moore (2010: 200), e.g., find that here the requirement of unanimity is pernicious. It has a detrimental effect on scientific deliberation, encouraging agreement where there is none in order to protect the authority of the group. It encourages misleading reports of the state of scientific agreement to the public. It undermines the epistemic equality of the deliberators. And it unfairly privileges the status quo with regard to any decisions that hinge on the outcome of the contest. In some cases, the unanimity requirement may lead to coercion by the majority (Beat ty 2006, Longino 2002) In others, it may cause a minority to conform in the name of agreement is given such weight that it becomes the point or end of deliberation rather than t he byproduct. 11 Far from being an academic problem, these concerns have even manifested significant changes in the rules of jury deliberation. In 1967, the United Kingdom abolished the unanimity requirement for juries in favor of a majority rule, citing the ease with whi ch coercing just one juror yielded acquittal (Hans and Vidmar 1986). Similarly, in 1972 the US Supreme Court ruled in Apodaca v. Oregon that states may convict with non unanimous juries.
154 Compounding these matters are issues that prevent informed decision, such as the elements of wickedness discussed in Chapter s 2 and 3. For instance, the problem of underdetermination occurs when evidence supports numerous theories but none c onclusively. Suppose we want to find out whether person X, Y, or Z took the last cookie from the cookie jar. If persons X and Y have short brown hair, and we find a short brown hair at the scene of the crime, this evidence supports two theories: that X too k it, and that Y took it. As we saw, diversity theorists endorse the unanimity rule because they assume a decision making body has enough evidence to know with certainty. But in the case of underdetermination, there is not enough evidence to eliminate all but one possibility. As the problems of higher education governance become more and more wicked, underdetermination promises to intensify. Indeed as we pointed out in Chapter 2, decision making is increasingly done under conditions of uncertainty. One cont ributing factor is a lack of evidence to suggest, not just what might be causing a problem, but what solution will help solve it. Epis temic O bjections There are also specifically epistemic objections to the unanimity requirement that rely (in part) on thes e non epistemic objections. In some cases, unanimity does not On Liberty argues challenges, and meaningful alternatives. His defense of freedom of thought and opinion the contrary opinion, mankind would be no more justified in silencing that one person,
155 expression of liberal values, Mill believes that censoring opinions diminishes the possibility of collective intelligence; it is an epistemic injustice: the peculiar evil of silencing the expression of opinion is, that it is robbing the human race; posterity as well as the existing generation; those who dissent from the opinion, still more than those who hold it. If the opinion is right, they are deprived of the opportunity to exchange error for truth: if wrong, they lose, what is almost as great a benefit, the clearer perception and livelier impression of truth, produced by its collision with error ( 1859/ 1993: 85). Incorrect opinions are nonetheless valuable because they redouble the epistemic value with blind certainty that our own is correct. But confidence in our own views should both encourage and rely on critique and tests of fallibility. That i s, if we really are correct, the truth of our opinion should have enough merit to withstand scrutiny and efforts to prove it wrong or false. Thus, he avers, truth is the product of contestation rather than a priori cting and disproving our opinion is the very condition which justifies us in assuming its truth for purposes of action; and on no other ( 1859/ hen, we have reason to believe the best decisions are reached in the dialogue between disparate voices, not necessarily in the overcoming of disagreement. This argument suggests a unanimous decision indicates accuracy or validity only in the presence of me aningful alternative viewpoints and opinions. It is one thing to be proven right because all other options were obviously erroneous; it is another when they are equally likely to be true or valid. While a dissenting minority of epistemic peers is not itsel f an indicator of truthfulness or even quality of decision, it does indicate the
156 presence of meaningful alternatives or a lack of conforming pressures. 12 It shows that the conditions were right for a choice based on merit or rational argument. Further, it s uggests, at least theoretically, that the problems of conformity have been obviated because by definition there is a small group that chooses not to abide the majority e confidence in the outcome of a democratic decision if there was a minority that voted Individual Judgments of Collective Solutions: The Problem with the Diversity Argument We can now consider the reasons intrinsic to diversity theory that render the unanimity requirement problematic. As we saw earlier, one precondition for diversity theory is sufficient difficulty: a problem must be complex or difficult enough that one person is incapable of finding the optimal solu tion (otherwise there is no real problem, just a lack of applied expertise) (Page 2007: 159). We can imagine that any individual problem solver experiences a problem of this magnitude as something s/he only partially understands and can only partially solv e. Thus perspectival limitations prevent each individual from understanding a difficult problem in its totality. Because of this difficulty, the best solution is often imperceptible except through a sufficient pooling of diverse perspectives and knowledge. And yet, each contributor is expected to realize the optimality of the best solution to that problem. This paradoxically requires each individual deliberator to understand the solution in its totality despite partial understanding of the problem it solves If we acknowledge the condition of sufficient difficulty and the independent nature of the oracle solution, then we cannot assume that 12 Again, this assumes unitary conditions, where everyone shares mutua l interests.
157 a collectively intelligent and better solution would appear as such to each and every individual problem solver. In oth er words, we cannot assume collective solutions will be individually self evident. I suggest the ultimate shortcoming with the epistemic defense of unanimity is that it is too cognitively demanding of individual problem solvers. Because of reasons and cond itions immanent to diversity theory, individual problem solvers are required to realize the optimality of solutions, despite the fact that both the problems and solutions are so difficult they can only be conceived of collectively. Take, for instance, Pag sufficient difficulty condition). Imagine a research team comprised of oncologists, biologists, chemists, biochemists, and other medically related research professionals. Suppose they share evaluation co nsensus on a more efficient delivery of a cancer could add insight about the treatment aspect that the others do not have. Assuming the oracle condition holds, this grou p should deliberate until it finds the solution that will produce its preferred outcome. But should we expect this pharmacist to understand the solution as well as the oncologist? Should we assume that her ability to contribute to the solution is tantamoun t to her understanding of that solution? Is it even possible to assume a self evident oracle solution in situations like this one, where each problem specialized expertise? It is perfectly likely that her knowledge perspective, heuristics, and input were of use to the group but that she does not completely understand why or how they were While rational understanding and subsequent unanimous agreement are certainly a viable deliberative possibility,
158 there is no gua rantee that every person will recognize the superiority of the best answer. Moreover, epistemically speaking there is nothing to be gained from the as optimal 13 It could be the case that she would never fully come to comprehend how she had contributed to finding a better solution unless she herself became an oncologist. But this would potentially undermine the claims of diversity theory (by reducing cognitive div ersity) and it would impose an unrealistically stringent requirement on consensus. The problem, I want to suggest here, is not with the oracle condition but with assumption that unanimity is the litmus test for finding an optimal solution. We would not wan t to deny that the group found a better solution merely because one or some of its members did not fully comprehend it. be reason to continue deliberation, especially if it might lead to compromises resulting in suboptimality. But this is exactly what a unanimous stopping rule might do in this scenario, potentially undermining the epistemic value of the group deliberative solution. It might be argued that an imperfect optimality is a sufficient basis for unanimity. 14 Suppose, for instance, the oncologist discovers an optimal solution and uses a series of metaphors, examples, or other heuristic devices to prove to the group that it i s the best. They need not understand specifics in order to get the gist and thus agree. But if this is the case, then unanimity becomes the product of persuasion rather than reason, and diversity theorists have 13 14 special thanks to Hlne Landemore, for suggesting and debating this point.
159 unwittingly omitted their own oracle conditio n. That is, if we (i) assume that optimal solutions are self evident once revealed, but (ii) rely on persuasion as a unifying force, then (iii) the self evident requirement must be false. 15 The optimal solution is imperceptible to some problem solvers becau se of the perspectival limitations assumed by the sufficient difficulty condition. The oncologist ( qua persuader) is not trying to introduce reasons in a more palatable way, because, by definition, those reasons cannot be understood by her interlocutors. I n this case, persuasion is used to yield unanimity through acquiescence rather than understanding. 16 This returns us to the problem of unanimity in existing epistemic deliberative models: If the optimal solution exists, then the fundamental deliberative tas k is finding that solution. But because knowledge is unequally distributed among deliberators, it is perfectly likely that some will realize it before others, and that some may not see it as optimal at all (despite contributing to its discovery). If a majo rity of decision makers agree the best solution has been located and are actually correct 17 then theoretically it would be best to end deliberation with a majority rule. Continuing in the hopes of unanimity would likely produce one of four conclusions: the minority comes around and a unanimous and optimal decision is reached; compromises are made until a unanimous but suboptimal decision is reached; the problem solving attempt is killed and the status quo remains; an external constraint (e.g. a time limit) is met and the decision is postponed. Note that only one of these four conclusions sounds like the scenario the 15 This us e of persuasion differs from the Carol Moseley Braun example given in Chapter 4. There impassioned and persuasive speech conveyed reasons and valid points that would have been ignored otherwise. It was a vehicle for her reasons. Here, persuasion substitut es for reasons. 16 In other words 17
160 epistemic diversity theorists foresee. In the other three, neither unanimity nor an optimal solution is found. What the unanimity requirement do es, then, is force deliberators into an all or nothing situation: By assuming the best solution will be equally self evident, they are led to discount anything less than full rational Habermasian agreement as suboptimal. My point is not to say that diversi ty theory cannot work, or that we should not be (Landemore and Page 2015: 9). At the very l east, epistemic theorists of democracy who rely on diversity theory must better explain why the best solution will also be self evident. In problems where the solution immediately solves the problem, this may be a trivial concern: the right answer balances the ledger, unlocks the lock, or satisfies the Sphinx who then allows us to pass. But in the case of wicked problems, what is there to guarantee that a perfectly rational solution will transcend the perspectival limitations and distortions that prevent u s from wholly understanding the problem in the first place (i.e. that contribute to making the problem sufficiently difficult)? If solutions are not self evident, then unanimity is both an unlikely and unnecessarily demanding stopping rule. A better decisi on rule would rely on the collective evaluation of collective solutions. Epistemic Acceptance and the Rule of Non opposition We see, then, that the argument in favor of unanimity loses ground upon closer s. In this section, I offer an epistemic defense of majority rule and suggest it should substitute for this overly stringent ity rule, they argue, should elicit minority
161 consent if that minority was fairly heard in deliberation. In doing so, however, their argument relies on a normative justification, viz that majority rule is acceptable to the extent that deliberation meets su bstantive and procedural standards. Here, I substitute epistemic value for these procedural standards. Beatty and Moore introduce the concept of deliberative acceptance by observing that unanimity juries tend to deliberate longer and rate the quality of de liberation higher than their non unanimity counterparts. One explanation for both trends is that this type simply tends to find the right answer. But this does not explain a third observation: even in unanimity trials some jurors go along with a verdict de is that [these jurors] are freely reporting a position with which they somehow identify, even though it is not their own personal position. With whom or what do they identify, if not themselves? They identify with their group. They are reporting what they take to be the position of the jury. This is not just the majority position, because they do not identify with the position (2010: 206). The explanation, as they have it, is that jurors are not asking themselves the answered that, each dissenti ng juror then decides whether or not that verdict was reached through sound reasoning and good deliberation. If so, they choose to go along with the group despite their personal opinion to the contrary. Beatty and Moore call this p if and only if the individual members, based on the quality of their deliberation, have openly agreed to let p
162 As a decision mechanism, deliberative acceptance wor ks in the following way 18 : once (1) full and fair deliberation has ended, someone volunteers what s/he takes to be a fair summary of the decision the group has yielded thus far and (2) an informal vote is conducted to establish the overall support for the d ecision as it stands. This voting mechanism is designed only to identify objections, and can be as simple as raising a hand. Here, silence is taken as consent. 19 From here three outcomes are possible. If no objections are raised, (3a) the decision stands an d an external rule is applied to conclude deliberation and oblige the group to act on its decision. If objections are raised, then (3b) the group hears them out by beginning another deliberative session (1) to address them. If no conclusion can be reached, (4) the session ends and the decision is postponed until a later session. 18 The summary and illustration are my own interpretations of the process. 19 Because the vote is used only as a way to signal dissent or disagreement, those in agreement are not resolved or may simply wish to withdraw them. Thus w e cannot derive from this process an explicit count of opinions, nor can we consider this vote to be a decision mechanism. Rather, it is simply a sort of queue for voicing opposition.
163 Figure 5 1. Chart of t he deliberative acceptance decision making process. decision are procedural concerns like fairness and equality and the substance of the debate produced therefrom: What better way to inspire confidence in a deliberative outcome t han to show that 1) the position in question had been tested against a worthy alternative; 2) the minority felt that they had been heard, that they had been treated as deliberative equals; and 3) having been heard, even the minority agreed to let the posit 209). Thus, a decision is acceptable if it was reached fairly and everyone was given equal chance to affect the outcome. I suggest, however, a modified version intended to substitute epistemic criteria and their 1. Meta consensus and shared interests: the group must agree on the nature of the problem and the preferred solution, and members
164 2. Satisfaction of deliberative requirements: all deliberators must acknowledge that deliberation has met epistemic and democratic requirements reason giving, decision oriented debate for and against a set of propositions conducted by relevantly diverse problem solvers following democratic rules (see Chapter 4 ). 3. Majority consent: a majority of deliberators must believe that group deliberation has yielded an optimal decision or s olution; i.e. steps 1 through 3 have been reached and it is obvious to the group that those who oppose it are few in number. 4. Potential for epistemic harm: the majority must believe that further deliberation would produce a suboptimal solution. Call this m odified version epistemic acceptance 20 Under these conditions, a dissenting minority of deliberators who disagree with the majority supported solution can nonetheless agree to let the group position stand as their own because they accept that that decision was reached according to the deliberative standards that tend to produce epistemic value, and because they trust the group is promoting their interests. The legitimating factor of this approach is the exhaustion of reasons 21 A traditional majority rule might consider a decision taken when it is established that one choice enjoys more support than its alternatives. But epistemic acceptance differs in that deliberation continues as long as good objections and reasons are raised This means a majority opinion is insufficient to end deliberation if valid objections remain. In order to satisfy the second condition (deliberative requirements), all reasons must be produced and heard from everyone willing and able to produce them. In step 3, any and all objections are considered by returning to deliberation. This recognizes the dissenting minority as epistemic peers. Once reasons have been exhausted (i.e. once the demands of epistemic justice have been satisfied) any remaining minority should be 20 21
165 able to admit that they were fairly heard. Though objections may still remain, they are unlikely to produce any good reasons for further deliberation. 22 At this point once we can say all reasons have been exhausted and a decision must be reached epistemic acceptance is invoked. The first condition (meta consensus and shared interests) justifies dissenters bearing the cost of consensus viz. living with a decision despite disagreement because all members know the group is committed to achieving its shared interests. Thus, I may fail to see why solution X is better than Y, but nonetheless consent to go along with solution X because I trust the group has deliberated according to meta consensus and with our mutually shared preferences in mind. The seco nd condition (deliberative requirements) gives the minority reason to believe that, despite personal disagreement or cognitive error, the solution located by the group and enjoying majority consent is likely to be the right or best one. Recall from Chapter 4 that reason giving, decision orientation, and relevant diversity of problem solvers invoking democratic rules tends to optimize the deliberative process. Along these lines, and following the basic premises of diversity theory, I argued that group delibe ration tends to outperform elite or expertocratic decision making groups. While no group can be expected to get every answer right every time, diverse deliberation is the safest and most hedged bet. But we also have reason to believe that the uneven distri bution of knowledge and the sufficient difficulty of problems being solved means that the best solution may not appeal to everyone, and so unanimity is not always an appealing decision rule. Epistemic acceptance acknowledges the real likelihood of 22 In other words, because deliberation requires reasons for and against propositions, we can say that any objections outside this definition are unreasonable. This is not to say they ar e irrational, but only that they cannot help reach a decision. This is grounds for the group to collectively refuse an objection. For more on the validity of objections, see Urfalino 2014: 334 7.
166 disagree unanimity rules. This solution, I contend, is more logically coherent with the basic claims underlying diversity theory because it relies on collective (rather than individual) ev aluation of collective solutions. The epistemic acceptance decision mechanism does this by invoking what makers answer th yielding an exact accounting of opinions or preferences This facilitates voting, which operates on the accumulation of th ese opinions. Thus in a majority vote, the proposal enjoying the most approval wins. Within a logic of consent, 330). Here, decision makers either object to a proposal o r they remain silent. Explicit approval plays no role in decision 2015, Moore and O'Doherty 2014). Because consent is shown through silence, deciding through a voting mechanism is unnecessary. 23 T here is no definitive distribution of voices accompanying each decision. Rather, there is the appearance of unanimity, which is actually a lack of objection. Because epistemic value is the product of good deliberation, the best guarantee that a solution is optimal (absent full unanimous agreement) is a majority agreement. Given that disagreement is to be expected in a properly diverse situation, a dissenting 23 To be clear, while approval is not required, it is also no t verboten. During deliberation (step 1), support for the best decision will likely come out as participants increasingly support it. My point is only to say that in the proposal phase (step 2), deliberators may speak in support of the proposed decision bu t are not required to as they are in a typical voting situation.
167 minority, despite its disagreement, should be wil l ing to consent given that the decision enjoying majority a pproval was made according to standards that tend to produce the based on the [solution finding] quality of your deliberation, even if there is some residual certain ty from individual cognition to deliberative process. Procedurally, judgments are still made at the individual level. However, the subject of those judgments is whether or not the group has yielded a majority decision as a result of following the rules and concern with finding an optimal solution. It merely denies the self evident quality of optimal solutions and consequently relocates the indicator of optimality to somet hing more easily and collectively identifiable. Non opposition as Realistic Stopping Rule Though a decision mechanism is sufficient for ending deliberation, it cannot impose a duty on the group to put that decision into action. Deliberation, as it was defi ned in Chapter 3, is only a way of evaluating information. As Urfalino (2010: 123) shows, it is incapable of bringing itself a close, because collective decisions do not themselves oblige the group or others to act on it: [W]e cannot say that a collective deliberation culminates in a collective intention to perform action A in the same way that intra personal deciding requires a collective stopping rule that produces an obligation, preci sely because a group cannot produce an intention for the future without such a stop.
168 Consider, for example, the difference between participating in presidential polls versus casting a ballot for president. In both cases, participants indicate their prefer ence and may even perform the same act (e.g. checking a box, raising a hand, punching a chad). But the ballots cast on election day determine the next president because we have determined that this particular round of voting will end with an official elect ion winner. We assign to it a majority stopping rule. This rule obliges the group to act on its decision, but we would not say that the decision was itself the source of that obligation. Without this external rule, a presidential vote is merely another pol l. The problem solving process is thus comprised of three parts: a deliberative weighing of information and reasons that ends in a decision ( viz which solution is best), a mechanism for yielding that decision, and a rule that tells us what to do with or how to act on that decision. In the previous section I proposed a new decision mechanism. Since we have also rejected unanimity as a stopping rule, what remains is to replace it with a non numerical majority rule that meets our specific needs. 24 Ideally, a decision mechanism will be sensitive to the background conditions that inform the nature of decision have predominant ly common or conflicting interests on matters about which the group participant weig interests are shared, as they are in our situation, promotion of those interests becomes 24 For a thorough accounting of numerical and non numerical rules, see Krick 2015.
169 the more important goal (Mansbridge 1980: 5 6). So our concern is not equal voting weight but e qual opportunity to contribute to the problem solving process. This is paired equality of participation to the process coexists with a recognition of the legitimacy of Likewise, stopping rules should be sensitive to the decision mechanisms that precede them. For example, because presidential elections take place under adversary ystem equally protects conflicting interests. A numerical superiority rule (e.g. plurality, majority, or supermajority) would then be the most appropriate rule, as it acknowledges disagreement and enacts the option with the most approval. The background co nditions necessitate a voting mechanism, which itself necessitates a majority rule. It would be absurd to implement a unanimity rule that states a decision is not considered taken until all voters agree on the same candidate. (We would never elect another president again!) Because epistemic acceptance follows a logic of consent, it requires what numerical e, during a moment of pending acutely felt by all, an absence : the absence of any overt opposition 25 It is non numerical because veto, but not an 25 This moment may not be reached until more than one round of deliberation, proposal, and objection (steps 1 through 3 in Figure 5.1).
170 that a decision faces approval, only a lack of explicit disapproval. T making process is significantly accelerated. While deliberation can and should be exhaustive, the concluding steps require less time because they involve only mentioning objections. In contrast to something like a vote, where each member or participant must voice his/her opinion and be counted, a non opposition rule takes into account only what needs to be said to reach an optimal decision. 26 Second, and more importantly, a decision by non Furthermore, because it avoids explicitly counting opinions, individuals do not have to declare themselves part of the minority if, for whatever reason, they do not want to (2014: 328). Conclusion Unanimity is often difficult, if not impossible, to observe in practice, even among those sharing meta consensus and mutual interests. While it retains ap peal as the ideal deliberative outcome, unanimity is so unlikely in everyday practice that we should be willing to find other rules for making binding decisions. A better alternative would be sensitive to this normative appeal but also recognize the exigen cies of real world decision people can disagree, and if this disagreement does not diminish the epistemic value of 26 E.g., a non opposition stopping rule of this type (or variants of) can be found in the European Council, council of the European Union, World Trade Organization, International Monetary Fund, World Bank, United Nations subcommitt ees, and the British cabinet (Krick 2015).
171 deliberation, then we have good reason to relax the unanimity condi tion and allow for less stringent but still effective decision making like a majority rule. find optimal solutions to difficult epistemic problems. They find that unanim ous agreement is the ideal stopping rule when problem solvers share meta consensus on an optimal solution and are deliberating over the best way to achieve it. Here, full unanimous agreement reflects that the group has found the ideal solution. While sympa thetic to this argument, I identified two problems with the unanimity requirement within diversity theory: (i) that perspectival limitations may prevent problem solvers from realizing the optimality of a solution (i.e. that oracles may not be self evident) ; and (ii) that despite these limitations, consent is based on individual realization of the optimal solution. I argued that when problems are so complex and difficult that they can only be conceived of collectively, we cannot require that individual probl em solvers immanently recognize the optimality of the best solution. Thus, the unanimity rule proves too cognitively demanding to act as a workable stopping rule for deliberation. of these problems. In shifting the basis of consent from the individual to the process, it joins the necessary background conditions consensus (meta consensus, shared interests, epistemic problems) with the reality that disagreement will likely accompany a division of cognitive labor. In doing so, this alternative promises to improve the existing epistemic models by making decision mechanisms sensitive to the optimality of a group decision or solution rather than to personal recognition and approval of it.
172 The Habermasian assumption holds that the better argument will appear or be received as such to each and every deliberator, as if there is a rationality at play that cannot help but transcend the cognitive limitations that determine the scope and range of our individual perspective s But when solving difficult epistemic problems that contain elements of wickedness, this intuition may reach its limit. In reality, there are lots of reasons that we may not be able to realize the best solution, even if we clea rly understand the problem and what it would take to solve it. However, we can overcome this problem if we can adopt a system of non unanimity consensus in which group members, even if they disagree, accept the outcome of a deliberative problem solving ses sion. Put another way, if we can devise a process that tends to produce the right or best answers, solutions, decisions, or choices, then a dissenting minority has good y or disagree that it is the best. This relies on a meta consensus of background conditions and shared interests between deliberators such that each knows the group represents their interests. This also requires a process that does in fact come to find the epistemically warranted decision more often than not. In doing so, the minority also has reason to believe procedural justice was served, because, referencing Chapter 4, reaching a decision requires that all deliberators are heard and that all information and arguments are weighed. Thus, in consenting, the minority both acknowledges that epistemic requirements are fulfilled and that procedural justice has been met.
173 CHAPTER 6 CONCLUSION From Talk to Work 1 My dissertation addresses whether higher education governance should be an elite or collective responsibility. The scope of this project has been relatively narrow: (i) to show that higher education is a venue best served by diverse and collective decision making, (ii) to show the conditions unde r which groups are (or can be) collectively intelligent problem solvers capable of addressing complex problems and (iii) to propose a theory of deliberation consistent with these conditions. To this end, I have focused on achieving two objectives: (i) show ing that collective intelligence is a tangible, if often unmet, goal in shared governance and (ii) suggesting that sensitivity to epistemic or knowledge related concerns (in addition to procedural/normative concerns) offers a way to improve and adapt share d governance. In the practice and study of democracy and democratic theory, collective intelligence is widely accepted as a problem solving asset. Indeed it helps explain not just the increasing popularity but also the accuracy of practices such as crowds ourced policy making or deliberative polling for addressing collective problems. In the debate over shared academic governance, however, group decision making is often perceived as an unrealistic and outdated, though admirable, institution one that sounds good in principle but rarely if ever produces good results. This opinion is emphasized most by those who prefer the elite and technocratic decision making regime I call the expertocracy. I have argued that while their diagnoses about the failures of shared governance are well grounded, their prescription (a system of divided governance with 1 This phrase is borrowed from Levine (2013: 39).
174 voluntary consultation) is misguided. In applying the theory of collective intelligence to shared governance, this dissertation has provided good reason to refute the ex pertocratic claim that one or a few elite administrators can outperform a diverse, deliberative problem solving body. In doing so, it has defended the claim that higher education governance should be a collective responsibility. But deliberation is not a panacea appropriate for all situations. In order to maintain an appropriately narrow scope, I have imposed a set of restrictions on the province of shared governance: 1. Problem solving: its fundamental task should be locating epistemically warranted solutions. 2. Sufficient difficulty: it must address problems so difficult, complex, and unprecedented (i.e. wicked) that no single expert, body of knowledge, or past solution is adequate to so lve them. In other words, they must require original and/or contextually specific solutions that are most likely produced by integrating diverse perspectives. 3. Epistemic benefits: Deliberation should be used to the extent that it produces, aggregates, and synthesizes relevant information. Chapter 1 noted that deliberation is an effective problem solving device when it comes to planning irreversible actions 2 but that executing action is often best done by skilled experts. Next, Chapters 2 and 3 showed that a procedure independent standard of correctness, however it is defined, is best met by deliberating over judgments about how to achieve it. Under these conditions, it is possible for collective problem solving groups to locate the best or right solutions t o concrete problems that, if not completely wicked, contain elements of wickedness, and increasingly so. This should cast doubt on the expertocratic claim that collective decision making has no place in institutions like 2
175 higher education. Chapter 4 propose d a theory of deliberation consis tent with this epistemic focus. It argued the best way to locate optimal solutions to these problems is to engage in decision oriented, reason giving debate by a sufficiently large and relevantly diverse group of problem so lvers. Finally, Chapter 5 found that a majority rule is both an efficient and epistemically beneficial stopping rule for ending deliberation that follows these deliberative requirements. Institutional Design: Solving Problems, Implementing Solutions David Hume famously said we cannot derive an ought from an is 3 Similarly, we cannot derive a will be from an ought 4 Firm values do not suffice for good predictions. This project has merely laid out the potential of collectively intelligent decision making. Th ere is no telling whether or not the institution of shared governance will improve or even continue to exist as higher education becomes increasingly technological and managerial in character. Thus while I have argued that academic governance can and shoul d be a collective responsibility, I also believe the institution must adapt and improve to accommodate the changing nature of the university. In these concluding remarks, I will more fully describe what an improved model might look like in practice. As we saw in Chapter 3, groups excel at problem solving and predicting future outcomes. When compared to skilled individuals, and even small groups of them, individuals of average intelligence in larger diverse groups tend to find better solutions and make more accurate predictions. When it comes to executing those tasks, however, groups often perform worse than skilled individuals. Combining the work from Chapter s 3 4
176 3 and 4, I make two observations: (i) epistemically, the task of moving from a point of uncertaint y (what is the nature of the problem?) to a point of risk (what is the best way to produce our desired outcome?) is best achieved by large groups using deliberative and democratic problem solving mechanisms; (ii) large groups are less effective at dealing with risky situations by implementing solutions than elite expertocratic groups. What would shared governance look like after taking into account the deliberative theory I propose here? Ideally, shared governance would fuse the theoretical defense of epis temic democracy with critiques of its current inefficiencies to produce a three step process: solving problems, writing policy, and executing policy I suggest the best institutional design is a sequential, mixed decision making regime that includes (i) a large deliberative problem solving body, (ii) a body that translates optimal solutions into workable policy, and (iii) an executive body tasked with putting the policy into practice. Identifying solutions should be the province of intelligent and capable problem solvers So the first task would be concerned with eliminating uncertainty by identifying the nature of the problems facing the school and then discerning an optimal solution. This step would be the province of a large, deliberative body that follo ws the requirements established in Chapter 4. Problem solving should include the input, expertise, judgments, and justified beliefs of all people who are likely to help identify the solution. This means any internal or external stakeholders could plausibly have something to contribute. As Chapter 4 shows, the number of deliberative participants would theoretically be large enough to produce cognitive diversity but small enough to avoid the impracticalities of large group discussions. While I believe the sit uation at hand will determine this floor and ceiling, a plausible range would be from twenty or
177 thirty people into the hundreds. They might meet together or in divided groups (as in a committee system), and might work continuously (at one time) or continua lly (iterated over a period of time with interruptions) depending on the situation. But from the claim that collective problem solving produces epistemic benefits, it does not follow that these groups should also design or execute policy, or that they woul d be the best vessel for doing so. Rather, diverse and deliberative bodies are useful to the extent they provide the necessary cognitive and epistemic means to plan for irreversible actions like policy making. It is helpful to remember that we are concern ed with nearly wicked problems that are multifaceted and that may require simultaneous attempts on each facet to find workable solutions. For this reason I think the overall size and inclusion of participants should be open to debate and determined accordi ng to the scope of the problem. Successful governance might require multiple information gathering sessions in open forums, as we saw in the case of Macalester College. There, the number of deliberators was both large possibly numbering in the hundreds and included faculty members, current and prospective students, and administrators. Crowd size was incidental and determined by its decision orientation; the priority was finding an answer to the er cases, a committee approach may be more appropriate, as we saw at the University of Maryland. Here the president and provost met in small but numerous meetings with faculty members to determine the best criteria for making very necessary budget cuts. Af ter hearing from hundreds of faculty (a few at a time) and gathering a sufficient amount of decision
178 oriented information, a committee system was implemented to turn these criteria into a workable policy. Because the goal is collectively intelligent decis ion making, deliberation should end once we locate the solutions for achieving mutually preferred outcomes. The stopping rule that brings deliberation to a close signifies (though probably imperfectly) a transition from uncertainty a bout what is to be done, to the risk of whether a solution will work. This signals the limit of deliberation. This stopping rule should signify the transition from the democratic step to the expertocratic one. Put anot her way, achieving solutions is exclu ded from deliberation. Collective involvement would end once we have figured out what is to be done. The second step of writing and executing policies would remain the province of elite administrative positions like presidents or provosts. For instance, in the case of Macalester College it was sufficient for the open deliberative forums to determine what was really the cause of lagging enrollment in the Russian Studies department. These forums allowed students to air their opinions and preferences, and to e xplain what deterred them from joining the program. But we would not then expect this faculty group, comprised of perhaps one hundred people, to use their findings to draft policy that corrected the problem. 5 This transition from problem solving to decisio n making would be indicated by which all sides of a debate and the comparative advantages of all possible solutions have weighed, resulting in at least a majority favo ring one decision over its alternatives. 5 organizers recommended to policymakers that an additional tenure line be opened (a recommendation that was ult imately adopted).
179 It might be objected that deliberators have no incentive to put forth a full effort if they know ahead of time that their decisions will not directly yield policy or think they will not be taken seriously. However, research suggests that on the contrary, deliberative experiments tend to increase knowledge and inform preferences even when they are dissociated from any formal decision making authority ( Barabas 2004, Elstub 2008, Farrar et al 2010, Fishkin 2011, Fishki n and Farrar 2005, Goodin and Dryzek 2006, Knobloch and Gastil 2015, List et al 2013, Niemeyer 2011) deliberative experiments found that participants were less susceptible to emotional appeals, more responsive to evidence and reaso n and included a wider array of considerations when they deliberated Despite the lack of real world authority, Niemeyer This sentiment is echoed in the many deliberative experiments that result in more informed voters. For instance, in post deliberative polls, Farrar et al (2010) and List et al (2013) find that subjects in deliberative experiments tend to improve the ordinal ranking of their prefe (a consistent ordering of preferences) than in their pre deliberative polls. This happens despite being in an experimental setting detached from any decision making authority. This reinforces the claims made by long time deliberative advocates like Fishkin, whose deliberative polls routinely find that deliberation produces information gains that yield better understanding of issues, more informed perspectives, and ultimately more knowledgeable voters. Ultimately, research suggests that participants will engage in deliberation for the sake of epistemic concerns like increased knowledge.
180 Once problems and solutions a re collectively identified i.e. once we know what to do experts often prove more effective at getting things done. Thus the third step would be accomplished by an executive body that can operate with expertocratic virtues like nimbleness and adaptivity The relationship between them would be one of proven trust, because ultimately all sides operate with shared interests and common concerns. Because there is no telling ahead of time what kind of policies or action will solve a problem, the executive body should be comprised of what higher education sc holar ad hoc task forces comprised of faculty or other relevant actors that convene to achieve a specific outcome and dissolve upon completion. Like the problem solvers, the people comprising th ese groups will vary from one project to the next, and while individuals might be reused, the task forces themselves are not permanent. This compositional fluidity would help is all to say that once we overcome uncertainty and know how to solve a problem, then the risk of achieving that result should be approached with expertocratic virtues. Assessing the Advantages of Deliberative Problem Solving This institutional design fin ds middle ground between on the one hand the epistemic and procedural virtues of inclusive deliberation, and on the other the need for schools to adapt to the quickening pace of academic decision making. The primary advantage this model has over the exper tocracy is that it helps prevent cognitive myopia and the outcome that results from it: premature consensus on a suboptimal solution. The expertocratic model would have leaders consult with other
181 governance units or external experts only to the extent 6 tha t they find such measures necessary to make their decision. Epistemically, this is problematic because a myopic (or worse, ideological) understanding of the problem at hand could set the agenda and constrain problem solving. Consider an example: for decade s expensive faculty salaries, the protection of academic freedom, and insensitivity to market principles were considered primary contributing factors to the rising costs of college attendance (Gardner et al 1983) lleges and universities in the United States began decreasing labor costs by replacing tenure line academic positions with non faculty now comprise over half of all academic positions and are responsible for more than 70 percent of teaching in the United States (Kezar and Maxey 2013) However, college costs are still higher than ever, 7 and it has now become clear that the increasing size of college administrations and student services (epi (Campos 2015, Ginsberg 2011, Greene, Kisida, and Mills 2010, Woodhouse 2015) The contingent labor solution was formulated by an inaccurate und erstanding of the real causes of rapidly increasing tuition, and perhaps more generally by an inability (such as a cognitive or perception bias) or unwillingness (such as scapegoating) to look within college administration. We now see through hindsight tha t increasing expensive student services and diminished funding have most likely been the motors driving up tuition costs. Consequently, the exponential increase in contingent labor might therefore 6 7 From 2001 prices for undergraduate tuition, room, and board at public institutions rose 40 percent, and prices at priva te nonprof it institutions rose 28 percent. See NCES 2013, 2016.
182 be considered a suboptimal solution. In contrast, in the go vernance model proposed in this dissertation, understanding or determining the nature of the problems would be the province of group deliberation and a crucial first step in the problem solving process. Knowing what problem to solve is equally important to solving it. As diversity theory shows, the reason that even a small number of average intelligence problem solvers tend to perform so well is because there are numerous perspectives that can identify incorrect or inaccurate information. This helps prevent the biases or scapegoating that might otherwise impede the effective detection of problems. On these grounds, it would be harder to set an ill advised policy agenda in my model simply because the result is whatever survives analysis by many differing view points and perspectives. It would thus prevent or make very unlikely the problem of cognitive homogeneity that leads to inaccurate agenda setting and premature consensus. However, my model sets expertocratic priorities in the subsequent executive stage, w here vision and action realize solutions. Here, the execution of action is best performed by skilled experts who know how to accomplish goals and produce outcomes. The primary difference is that in my model these goals would be set by diverse groups of cam pus stakeholders each contributing to a mutually preferred outcome. A second advantage is that by using governance as a venue for democratic co creation, this model would help schools fulfill their citizen building mission. As we saw in Chapter 1, the driv e for democratic civic engagement on campus is perhaps at an all time high. Both colleges and the broader community recognize the need to teach students how to be citizens. To this end, service learning programs have proliferated, as
183 have improvements in c ampus community relations. Why not also instill democratic and civic virtues into the fabric of academic decision referred not just to ruling collectively but also to collectively wielding the power to shape and determine the public sphere (Ober 2008) Similarly, the emerging field of civic (Boyte 2011, Dzur 2008, 2010) Deli berative shared governance is a mechanism for democratic thinking of this type at the institutional level. An expertocratic or technocratic model, by contrast, teaches students that a limited number of professions and people are worthy of making the kind s of major decisions that stand to affect the whole. By introducing epistemic democracy into governance I am thus suggesting that colleges and universities should not just teach democratic values but should also practice what they preach and operate by them. Finally, democratic inclusion can build social and civic capital within the university. As we saw in Chapter 2, shared governance often suffers from a lack of cohesion, trust, or reciprocity across governance units. Studies show that people are more likel y to comply with decisions when they feel included, heard, or constitutive in the process. In fact, Elinor Ostrom won the Nobel Prize in economics for showing that when it comes to protecting common follow a set of rules and to monitor and enforce rule conformance themselves has been (1996: 23) This same basic intuition has been shown to exist in other venues, such as highe r education: smart and effective campus policies will often be ignored if there is
184 sufficient distrust or antagonism between governance units (Johnson and Pajares 1996, Yankelovich 1999). Furthermore, studies show that faculty who feel powerless outside th eir own department are less likely to abide campus initiatives (Tierney and Minor 2004) Simply put, with sufficient distrust faculty tend to resist change purely because they believe it will disadvantage them. However, meaningful shared governance can obv iate this problem. By making democratic inclusion a cornerstone of the process, my model would use collective decision making as the means by which the campus builds its own community, putting governance units into what Leighninger (2006) calls an a fact that social and civic capital is just as necessary as perfect information and nimble actors. Ultimately, this model allows for both campus wide inclusion and quick exper t action. In a sense, it isolates what is admirable on both sides of the shared governance debate and finds ways for them to complement each other to the advantage of the broader institution. Expertocrats quite rightly identify a lack of epistemic concern as the fundamental problem with current governance practices. However, their prescription for simply faster action seems reactionary, as if to say the opposite of what does not work must work. Further, it ignores rather than confronts the very idea that de mocratic principles are messy by design because governance can be too efficient that makes it possible to make good decisions more quickly also makes it possible to ot be that decisions are made too slowly because of the drag of consultation, but that they are
185 olutions time to bubble to the surface and be proven better than alternatives. There are both epistemic and procedural/normative benefits to diverse and inclusive problem solving that is lacking in cognitively homogenous alternatives The debate over the future of shared academic governance concerns much more than institutional decision making. American higher education has traditionally carried the civic mission of transforming young adults into citizens who can cooperate, work together across differences and democratically co create their world. The lectures and lessons we teach students are meant to inform and enrich their democratic experiences, allowing them to better understand the value of such efforts. The expertocracy in contrast, would reserve authority for a knowledge elite. This kind of epistemological not ignore the very real need for responsiveness and action. But neither can we ignore that higher education is a citizen building enclave. Rather than give in to technocratic world making at the institutional level.
186 LIST OF REFERENCES AAUP. 2015. "Statement on Government of Colleges and Universities." American Association of University Professors Accessed March 2015. http://www.aaup.org/report/statement government colleges and universities AAUP. 2016. "Sanctioned Institutions." American Association of University Professors Accessed February 2015. http://www.aaup.org/our work/shared governance/sanctioned institutions Ackerman, Bruce, and James S. Fishkin. 2004. Deliberation Day New Haven: Yale University Press. Allen, Danielle S. 2004. Talking to Strangers : Anxieties of Citizenship Since Brown v. Board of Education Chicago: The University of Chicago Press. Anderson, Elizabeth. 2006. "The Epistemology of Democracy." Episteme 3 (1 2):8 22. doi: 10.1353/epi.0.0000. Andler, Daniel. 2012. "What Has Collective Wisdom to Do with Wisdom?" In Collective Wisdom: Principles and Mechanisms edited by H l ne Landemore and Jon Elster. Cambridge: Cambridge University Press. Ansell, Chris, and Alison Gash. 2007. "Collaborative Governance in Theory and Practice." Journal of Publi c Administration Research and Theory 18 (4):543 571. doi: 10.1093/jopart/mum032. Aristotle. 1996. The Politics Edited by Stephen Everson. Cambridge: Cambridge University Press. Arneson, Richard. 2013. "Democracy Is Not Intrinsically Just." In Political Philosophy in the Twenty First Century edited by Steven M Cahn and Robert B Talisse, 181 198. Boulder: Westview Press. Arrow, Kenneth. 1951. Social Choice and Individual Values New York: Wiley. Asch, Solomon. 1956. "Studies of Independence and Conformity : I. A Minority of One Against a Unanimous Majority." Psychological Monographs: General and Applied 70 (9):1 70. Austen Smith, David, and Timothy J Feddersen. 2006. "Deliberation, Preference Uncertainty, and Voting Rules." American Political Science Revi ew 100 (2):209 217. Bchtiger, Andre, Simon Niemeyer, Michael Neblo, Marco Steenbergen, and Jurg Steiner. 2010. "Disentangling Diversity in Deliberative Democracy: Competing Theories, Their Blind Spots and Complementarities." Journal of Political Philosop hy 12:32 63. doi: 10.1111/j.1467 9760.2009.00342.x.
187 Bahls, Steven C. 2014. Shared Governance In Times of Change: A Practical Guide for Universities and Colleges Washington, DC: AGB Press. Bahrani, Bahador, Karsten Olsen, Peter Latham, Andreas Roepstorff, Geraint Rees, and Chris Frith. 2010. "Optimally Interacting Minds." Science 329 (5995):1081 1085. American Political Science Review 98 (4):687 701. Barber, Benjamin. 1984. Strong Democracy: Participatory Politics for a New Age Berkeley: University of California Press. Barber, Benjamin, and Richard Battistoni. 1993. "A Season of Service: Introducing Service Learning into the Liberal Arts Curriculum." PS: Political Science & Politics 26 (2): 235 240. Barber, Michael, Katelyn Donnelly, Saad Rizvi, Avalanche Is Coming: Higher Education and the Revolution A head. Institute for Public Policy Research. Barden, Dennis, a nd Janel Curry. 2013. "Faculty M e mbers Can Lead, But Will They?" Chronicle of Higher Education Accessed February 19. http://chronicle.com/article/Faculty Members Can Lead but/138343/. Battistoni, Richard. 1997. "Service Learning and Democratic Citizenship." Theory Into Practice 36 (3):150 156. Beatty, John. 2006. "Masking Disagreement Among Experts." Episteme 3 (1):52 67. Beatty, John, and Alfred Moore. 2010. "Should We Aim for Consensus?" Episteme 7 (3):198 214. Ben Porath, Sigal. 2013. "Education for Shared Fate Citizenship." In Education, Justice, and Democracy edited by Danielle S. Allen and Rob Reich, 80 100. Chicago: University of Chicago Press. Benhabib, Seyla. 1996a. Democracy and Difference: Contesting th e Boundaries of the Political Princeton: Princeton University Press. Benhabib, Seyla. 1996b. "Toward a Deliberative Model of Democratic Legitimacy." In Democracy and Difference: Contesting the Boundaries of the Political edited by Seyla Benhabib, 67 94. Princeton: Princeton University Press. Benson, Lee, Ira Harkavy, and Matthew Hartley. 2005. "Integrating a Commitment to the Public Good Into the Institutional Fabric." In Higher Education for the Public Good: Emerging Voices from a National Movement edit ed by Adrianna J Kezar, Anthony Chambers and John Burkhardt, 185 216. San Francisco: Jossey Boss.
188 Benson, Lee, Ira Richard Harkavy, and John L Puckett. 2007. Dewey's Dream: Universities and Democracies in an Age of Education Reform Philadelphia: Temple Un iversity Press. Birnbaum, Robert. 1988. How Colleges Work: The Cybernetics of Academic Organization and Leadership San Francisco: Jossey Bass. Birnbaum, Robert. 2000. Management Fads in Higher Education: Where They Come From, What They Do, Why They Fail San Francisco: Jossey Bass. Birnbaum, Robert. 2004. "The End of Shared Governance: Looking Back or Looking Ahead." New Directions for Higher Education (127):5 22. Bohman, James. 1996. Public Deliberation: Pluralism, Complexity, and Democracy Cambridge: M IT Press. Bohman, James. 1998. "Survey Article: The Coming of Age of Deliberative Democracy." Journal of Political Philosophy 6 (4):400 425. doi: 10.1111/1467 9760.00061. Bohman, James. 2003. "Deliberative Toleration." Political Theory 31 (6):757 779. do i: 10.1177/0090591703252379. Bohman, James. 2006. "Deliberative Democracy and the Epistemic Benefits of Diversity." Episteme 3 (03):175 191. Bohman, James. 2007. "Political Communication and the Epistemic Value of Diversity: Deliberation and Legitimation in Media Societies." Communication Theory 17 (4):348 355. Bohman, James. 2009. "Epistemic Value and Deliberative Democracy." The Good Society 18 (2):28 34. Bohman, James, and William Rehg. 1997. Deliberative Democracy: Essays on Reason and Politics Camb ridge: MIT Press. Bok, Derek. 2003. Universities I n the Marketplace: The Commercialization of Higher E ducation Princeton, N.J.: Princeton University Press. Bowen, William G., and Eugene M. Tobin. 2015. Locus of Authority: The Evolution of Faculty Roles in the Governance of Higher E ducation Princeton: Princeton University Press. Boyte, Harry C. 2011. "Constructive Politics as Public Work: Organizing the Literature." Political Theory 39 (5):630 660. doi: 10.1177/0090591711413747. Boyte, Harry C, and Eric F retz. 2011. "Civic Professionalism." In "To Serve a Larger Purpose": Engagement for Democracy and the Transformation of Higher
189 Education edited by John A Saltmarsh and Matthew Hartley, 82 101. Philadephia: Temple University Press. Boyte, Harry C. 2008. "Against the Current: Developing a Civic Agency of Students." Change 40 (3):8 15. Boyte, Harry C. 2015. Democracy's Education: Public Work, Citizenship, & the Future of Higher Education Nashville: Vanderbilt University Press. Boyte, Harry C., and Elizabeth Hollander. 1999. "Wingspread Declaration on Renewing the Civic Mission of the American Research University." http://digitalcommons.unomaha.edu/slceciviceng/8 Bringle, Robert, Richard Games, and Edward Malloy. 1999. Colleges and Universities as Citizens Massachusetts: Allyn and Bacon. Brown, Wendy. 2015. Undoing the Demos : Neoliberalism's Stealth Revolution Cambridge: MIT Press. Buchanan, Allen. 2002. "Politic al Legitimacy and Democracy." Ethics 112 (4):689 719. Buchanan, Richard. 1992. "Wicked Problems in Design Thinking." Design Issues 8 (2):5 21. Butin, Dan W. 2010. Service learning in Theory and Practice: The Future of Community Engagement in Higher Educa tion : Palgrave Macmillan. Campos, Paul F. 2015. "The Real Reason College Tuition Costs So Much." New York Times http://www.nytimes.com/2015 /04/05/opinion/sunday/the real reason college tuition costs so much.html Chambers, Simone. 2003. "Deliberative Democratic Theory." Annual Review of Political Science 6 (1):307 326. Chambers, Simone. 2009. "Rhetoric and the Public Sphere: Has Deliberativ e Democracy Abandoned Mass Democracy?" Political Theory 37 (3):323 350. doi: 10.1177/0090591709332336. Chappell Zsuzsanna. 2011. "Justifying Deliberative Democracy: Are Two Heads Always Wiser Than O ne?" Contemporary Political Theory 10 (1):78 101. doi: 10.1057/cpt.2010.8. Checkoway, Barry. 2001. "Renewing the Civic Mission of the American Research University." The Journal of Higher Education 72 (2):125 147. Christiano, Thomas. 1996. The Rule of Many Boulder: Westview Press.
190 Christiano, Thomas. 2004. "T he Authority of Democracy." Journal of Political Philosophy 12 (3):266 290. Cohen, Joshua. 1986. "An Epistemic Conception of Democracy." Ethics 97 (1):26 38. Cohen, Joshua. 1989. "Deliberation and Democratic Legitimacy." In The Good Polity edited by Alan Hamlin and Philip Pettit. Oxford: Basil Blackwell. Cohen, Joshua. 1997. "Deliberation and Democratic Legitimacy." In Deliberative Democracy: Essays on Reason and Politics edited by James Bohman and William Rehg. Cambridge: MIT Press. Cohen, Joshua. 2 007. "Deliberative Democracy." In Deliberation, Participation, and Democracy: Can the People Govern? edited by Shawn W Rosenberg, 219 236. London: Palgrave Macmillan. Cohen, Joshua, and Joel Rogers. 2003. "Power and Reason." In Deepening Democracy: Instit utional Innovations in Empowered Participatory Governance edited by Archon Fung and Erik Olin Wright, 237 255. London: Verso. Colby, Anne, Elizabeth Beaumont, Thomas Ehrlich, and Josh Corngold. 2010. Educating for Democracy: Preparing Undergraduates for R esponsible Political Engagement San Francisco: Jossey Boss. Collis, David. 2004. "The Paradox of Scope: A Challenge to the Governance of Higher E ducation." In Competing Conceptions of Academic Governance edited by William G. Tierney. Baltimore: The Johns Hopkins University Press. Coughlan, Peter. 2000. "In Defense of Unanimous Jury Verdicts: Mistrials, Communication, and Strategic Voting." American Political Science Review 94:375 93. Dahl, Robert. 1989. Democracy and Its Critics New Haven: Yale Universi ty Press. Darden, Mary Landon. 2009. Beyond 2020: Envisioning the Future of U niversities in America Lanham: Rowman & Littlefield. Delli Carpini, Michael X, Fay Lomax Cook, and Lawrence R Jacobs. 2004. "Public Deliberations, Discursive Participation and Ci tizen Engagement: A Review of the Empirical Literature." Annual Review of Political Science 7 (1):315 344. Dryzek, John S 2000. Deliberative Democracy and Beyond: Liberals, Critics, and Contestations Oxford: Oxford University Press. Dryze k, John S 2001 "Legitimacy and E conomy in Deliberative D emocracy." Political Theory 29 (5):651 669.
191 Dryzek, John S 2010. "Rhetoric in Democracy: A Systemic Appreciation." Political Theory 38 (3):319 339. doi: 10.1177/0090591709359596. Dryzek John S. and Simon Niem eyer. 2006. "Reconciling Pluralism and Consensus as Political Ideals." American Journal of Political Science 50 (3):634 649. Duderstadt, James J. 2004. "Governing the Twenty first Century University: A View from the Bridge." In Competing Conceptions of Ac ademic Governance edited by William G. Tierney. Baltimore: The Johns Hopkins University Press. Duderstadt, James J. 2009. The View from the Helm: Leading the American University During an Era of C hange Ann Arbor: University of Michigan Press. Duderstadt, James J, and Farris W Womack. 2004. The Future of the Public U ni versity in America: Beyond the C rossroads Baltimore: Johns Hopkins University Press. Dzur, Albert W. 2008. Democratic Professionalism: Citizen Participation and the Reconstruction of Professional Ethics, Identity, and Practice University Park: Pennsylvania State University Press. Dzur, Albert W. 2010. "Democratizing Academic Professionalism Inside and Out." New Directions for Higher Education 2010 (152):75 82. doi: 10.1002/he.415 Eckel, Peter D. 2000. The Review of Higher Education 24 (1):15 39. Eisenstadt, Todd, A Carl LeVan, and Tofigh Maboudi. 2015. "When Talk Trumps Text: The Democratizing Effects of Deliberation during Constituion Making, 1974 2011." American Political Sci ence Review 109 (3):592 612. Elster, Jon. 1997. "The Market and the Forum: Three Varieties of Political Theory." In Deliberative Democracy: Essays on Reason and Politics edited by James Bohman and William Rehg, 3 33. Cambridge: MIT Press. Elster, Jon. 199 8. Deliberative Democracy Cambridge: Cambridge University Press. Elstub, Stephen. 2008. Towards a Deliberative and Associational Democracy Edinburgh: University of Edinburgh Press. Estlund, David. 1994. "Opinion Leaders, Independence, and Condorcet's Jur y Theorem." Theory and Decision 36:131 162. Estlund, David. 1997. "Beyond Fairness and Deliberation: The Epistemic Dimension of Democratic Authority." In Deliberative Democracy: Essays on Reason and Politics edited by James Bohman and William Rehg, 173 2 04. Cambridge: MIT Press. Estlund, David. 1999. "Arguing and Bargaining in Constituent Assemblies." University of Pennsylvania Journal of Constitutional Law 2 (2):345 421.
192 Estlund, David. 2008. Democratic Authority: A Philosophical Framework Princeton: P rinceton University Press. Farrar, Cynthia, James Fishkin, Donald Green, Christian List, Robert Luskin, and British Journ al of Political Science 40 (2): 333 47. Feldman, Richard, and Ted A Warfield. 2010. Disagreement Oxford: Oxford University Press. Finkelstein, Martin, Ming Ju, and William K Cummings. 2011. Changing Governance and Management in Higher Education: The Perspectives of the A cademy Vol. 2: Springer Science & Business Media. Fish, Stanley. 2007. "Shared Governance: Democracy is Not an Educational Idea." Change: The Magazine of Higher Learning (March April). Fishkin, James S. 1991. Democracy and Deliberation: New Directions for Democ ratic Reform New Haven: Yale University Press. Fishkin, James S. 1995. The Voice of the People: Public Opinion and Democracy New Haven: Yale University Press. Fishkin, James S. 2009. When the People Speak: Deliberative Democracy and Public Consultation New York: Oxford University Press. Fishkin, James S. 2010. "Consulting the Public Thoughtfully: Prospects for Deliberative Democracy." In Deliberative Democracy in Practice edited by David Kahane, Daniel Weinstock, Dominique Leydet and Melissa WIlliams, 1 94 208. Vancouver: UCB Press. Fishkin, James S Critical Review 23 (3): 393 403. Fishkin, James S The Deliberat ive Democracy Handbook: Strategies for Effective Civic Engagement in the 21st Century edited by J ohn Gastil and Peter Levine, 68 79. San Francisco: Jossey Bass. Fuerstein, Michael. 2008. "Epistemic Democracy and the Social Character of Knowledge." Episte me 5 (1):74 93. Fuerstein, Michael. 2014. "Democratic Consensus as an Essential Byproduct." Journal of Political Philosophy 22 (3):282 301. doi: 10.1111/jopp.12040. Fung, Archon. 2003. "Recipes for Public Spheres: Eight Institutional Design Choices and Their Consequences." Journal of Political Philosophy 11 (3):338 367.
193 Fung, Archon. 2006. "Varieties of Participation in Complex Governance." Public Administration Rev iew (December). Fung, Archon. 2015. "Putting the Public Back into Governance: The Challenges of Citizen Participation and Its Future." Public Administration Review doi: 10.1111/puar.12361. Fung, Archon, and Erik Olin Wright. 2003. Deepening Democracy: Institutional Innovations in Empowered Participatory Governance London: Verso. Gaff, Jerry G. 2007. "What If the Faculty Really Do Assume Responsibility for the Educational Program?" Liberal Education 93 (4):6 13. Ganz, Marshall. 2004. "Why David Sometim es Wins: Strategic Capacity in Social Movements." In Rethinking Social Movements : Structure, Meaning, and Emotion edited by Jeff Goodwin and James M. Jasper, 177 198. Lanham: Rowman & Littlefield. Gardner, David P, Y Larson, W Baker, A Campbell, and E Cro sby. 1983. A Nation at Risk: The Imperative for Education Reform. United States Department of Education. Gaus, Gerald. 1997. "Reason, Justification, and Consensus: Why Democracy Can't Have It All." In Deliberative Democracy: Essays on Reason and Politics edited by James Bohman and William Rehg, 205 242. Cambridge: MIT Press. Generation Citize n 2015. "Returning to our Roots: Educating for Democracy (White Paper)." Accessed April 2016. http://www.generationcitizen.org/wp content/uploads/2015/11/FINAL Educating for Democracy 11.16.15.pdf Gerber, Larry G. 2014. The Rise and Decline of Faculty Governance: Professionalization and the Modern American University Baltimore: John Hopkins University Press. Gerber, Larry G. 2015. "College and University Governance: How the AAUP Has Established Widely Accepted Norms of Shared Governance." Academe 2015 (Jan Feb). Gideon, Gavan. 2012. "The Challenge of 'Shared Governance'." Yale Daily News Accessed May 2016. http://yaledailynews.com/blog/2012/04/12/up close the challenge of shared governance/ Ginsberg, Benjamin. 2011. The Fall of the Faculty: The Rise of the All Administrative University and Why it M atters New York: Oxford University Press. Giroux, Henry A. 2004. The Terror of Neoliberalism: Authoritarianism and the Eclipse of Democracy Virgi nia: Paradigm Publishers.
194 Goldman, Alvin. 1999. Knowledge in a Social World Oxford: Clarendon Press. Goodin, Robert E. 2003. Reflective Democracy Oxford: Oxford University Press. Goodin, Robert E. 2005. "Seque ncing Political Moments." Acta Politica 40 (2 ):182 196. Goodin, Robert E. 2008. Innovating Democracy Oxford: Oxford University Press. Goodin, Robert E., and John Political Uptake of Mini Politics and Society 34 (2): 219 44. Greene, Jay, Brian Kisida, and Jonathan Mills. 2010. Administrative Bloat at American Universities: The Real Reason for High Costs in Higher Education. Phoenix, AZ: Goldwater Institute. Greenhalgh, Trisha. 2015. "Higher Education Governance as Language Games: A Wittgenstei nian Case Study of the Breakdown of Governance at the London School of Economics 2004 2011." Higher Education Quarterly 69 (2):193 213. doi: 10.1111/hequ.12064. Grofman, Bernard, and Scott Feld. 1988. "Rousseau's General Will: A Condorcetian Perspective." American Political Science Review 82 (2):567 576. Guinier, Lani. 2015. The Tyranny of the Meritocracy: Democratizing Higher Education in America Boston: Beacon Press. Gutmann, Amy, and Dennis Thompson. 1996. Democracy and Disagreement Cambridge: Harvard University Press. Gutmann, Amy, and Dennis Thompson. 2002. "Deliberative Democracy Beyond Process." Journal of Political Philosophy 10 (2):153 174. Gutmann, Amy, and Dennis Thompson. 2004. Why Deliberative Democracy? Princeton: Princeton University Press Haber, Jonathan. 2014. MOOCs Cambridge: MIT Univeristy press. Habermas, Jurgen. 1975. Legitimation Crisis Boston: Beacon Press. Habermas, Jurgen. 1996. Between Facts and Norms Translated by William Rehg. Cambridge: MIT Press. Hans, Valerie, and Neil Vidmar. 1986. Judging the Jury New York: Plenum Press. Harvey, Jerry B. 1974. "The Abilene Paradox: The Management of Agreement." Organizational Dynamics 3 (1):63 80. Hergovich, Andreas, Reinhard Schott, and Christoph Burger. 2010 "Biased Evaluation of Abstracts Depending on Topic and Conclusion: Further Evidence of a
195 Confirmation Bias Within Scientific Psychology." Current Psychology 29 (3):188 209. doi: 10.1007/s12144 010 9087 5. Hess, Diana E. 2009. Controversy in the Classroo m: The Democratic Power of Discussion New York: Routledge. Hirsch, Werner Z, and Luc E Weber. 2001. Governance in Higher Education: The University in a State of Flux London: Economica. Hirt, Edward R, and Keith D Markman. 1995. "Multiple Explanation: A C onsider an Alternative Strategy for Dibiasing Judgments." Journal of Personality and Social Psychology 69 (6):1069 1086. Hobbes, Thomas. 1642/1983. De Cive Edited by Howard Warrender. Oxford: Clarendon Press. Hobson, Kersty, and Simon Niemeyer. 2011. "Pu blic Responses to Climate Change: The Role of Deliberation in Building Capacity for Adaptive Action." Global Environmental Change 21 (3):957 971. Holzinger, Katharina. 2004. "Bargaining Through Arguing: An Empirical Analysis Based on Speech Act Theory." Political Communication 21 (2):195 222. doi: 10.1080/10584600490443886. Hong, Lu, and Scott E. Page. 2001. "Problem Solving by Heterogenous Agents." Journal of Economic Theory 97 (1):123 63. Hong, Lu, and Scott E. Page. 2004. "Groups of Diverse Problem So lvers Can Outperform Groups of High Ability Problem Solvers." Proceedings of the National Academy of Sciences of the United States of America 101 (46):16385 89. Hong, Lu, and Scott E. Page. 2012. "Some Microfoundations of Collective Wisdom." In Collective Wisdom: Principles and Mechanisms edited by H l ne Landemore and Jon Elster, 56 71. Cambridge: Cambridge University Press. Hunter, Susan, and Richard Brisbin Jr. 2000. "The Impact of Service Learning on Democratic and Civic Values." PS: Political Scie nce & Politics 33 (3):623 626. Ingham, Sean. 2012. "Disagreement and Epistemic Arguments for Democracy." Politics, Philosophy & Economics :1 20. doi: 10.1177/1470594X12460642. Janis, Irving, and Leon Mann. 1977. Decision Making: A Psychological Analysis of Conflict, Choice, and Commitment New York: Free Press. Johnson, Margaret J, and Frank Pajares. 1996. "When Shared Decision Making Works: A 3 year Longitudinal Study." American Educational Research Journal 33 (3):599 627.
196 Kant, Immanuel. 1788/1997. Criti que of Pure Reason Translated by Mary J Gregor. Cambridge: Cambridge University Press. Kaplan, Gabriel E. 2004a. "Do Governance Structures Matter?" In New Directions for Higher Education edited by William G Tierney and Vicente M Lechuga, 25 34. Kaplan, G abriel E. 2004b. "How Academic Ships Actually Navigate." In Governing Academia edited by Ronald Ehrenberg, 165 208. Ithaca: Cornell University Press. Keller, George. 2004. "A Growing Quaintness: Traditional Governance in the Markedy New Realm of US Higher Education." In Competing Conceptions of Academic Governance edited by William G. Tierney. Baltimore: The Johns Hopkins University Press. Kellogg, Alex P. 2001. "Faculty Senate at Notre Dame, Angry Over Lack of Clout, Votes to Dissolve." The Chronicle of Higher Education Accessed October 2015. http://chronicle.com/article/Faculty Senate at Notre Dame/11235 Kezar, Adrianna J. 2004. "What is More Important to Effective Governa nce: Relationships, Trust, and Leadership, or Structures and Formal Processes?" In New Directions for Higher Education edited by William G Tierney and Vicente M Lechuga, 35 46. Kezar, Adrianna J, Anthony Chambers, and John Burkhardt. 2005. Higher Education for the Public Good: Emerging Voices from a National Movement San Francisco: Jossey Boss. Kezar, Adrianna J, and Peter D Eckel. 2004. "Meeting Today's Governance Challenges: A Synthesis of the Literature and Examination of a Future Agenda for Sc holarship." The Journal of Higher Education 75 (4):371 399. Kezar, Adrianna J, and Daniel Maxey. 2013. "The Changing Academic Workforce." Trusteeship King, Jonathan Alan, Ruth Perry, and Frederick Salvucci. 2014. "Commercial Intrustion into Academic Spac e." American Association of University Professors Accessed August 2016. https://www.aaup.org/article/commercial intrusion academic space Kitcher, Philip. 1990. "The Divisio n of Cognitive Labor." The Journal of Philosophy 87 (1):5 22. Klein, Joe. 2012. "Russian Studies Back on Track Post vote." The Mac Weekly Accessed October 2015. http ://themacweekly.com/2012/09/russian studies back on track post vote/ Knight, Jack, and James Johnson. 1994. "Aggregation and Deliberation: On the Possibility of Democratic Legitimacy." Political Theory 22 (2):277 296.
197 Knobloch, Katherine R, and John Gas til. 2015. "Civic (Re)socialisation: The Educative Effects of Deliberative Participation." Politics 35 (2):183 200. Koehler, Jonathan. 1993. "The Influence of Prior Beliefs on Scientific Judgments of Evidence Quality." Organizational Behavior and Human D ecision Processes 56 (1):28 55. Kolowich, Steve. 2013. "Why Professors at San Jose State Won't Use a Harvard Professor's MOOC." Chronicle of Higher Education. Accessed May 2016. http://chronicle.com/article/Why Professors at San Jose/138941/ Koriat, Asher. 2012. "When Are Two Heads Better Than One and Why?" Science 336 (6079):360 2. Krick, Eva. 2015. "Consensual Decision Making Without Voting". In ARENA Centre for European Studies Working Paper Accessed November 2015. http://www.sv.uio.no/arena/english/research/publications/arena publicatio ns/workingpapers/working papers2015/wp3 15.pdf Landemore, Hlne. 2012. "Why the Many Are Smarter than the Few and Why It Matters." Journal of Public Deliberation 8 (1). Landemore, Hlne. 2013. Democratic Reason: Politics, Collective Intelligence, and the Rule of the M any Princeton: Princeton University Press. Landemore, Hlne. 2014. "Democracy as Heuristic: The Ecological Rationality of Political Equality." The Good Society 23 (2):160 174. Landemore, H l ne, and Jon Elster. 2012. Collective Wisdom : Principles and Mechanisms Cambridge: Cambridge University Press. Landemore, Hlne, and Scott E. Page. 2015. "Deliberation and Disagreement: Problem Solving, Prediction, and Positive Dissensus." Politics, Philosophy & Economics 14 (3):229 254. doi: 10. 1177/1470594x14544284. Leighninger, Matthew. 2006. The Next Form of Democracy: How Expert Rule is Giving Way to Shared G overnance and Why Politics Will Never Be the S ame Nashville: Vanderbilt University Press. Lerner, Jennifer, and Philip Tetlock. 1999. Accounting for the Effects of Accountability." Psychological Bulletin 125 (2):255 275. Levine, Peter. 2013. We Are the Ones We Have Been Waiting For: The Promise of Civic Renewal in America New York: Oxford University Press. List, Christian. 2007. "Delib eration and Agreement." In Deliberation, Participation, and Democracy: Can the People Govern? edited by Shawn Rosenberg, 64 81. London: Palgrave Macmillan.
198 List, Christian, and Robert E. Goodin. 2001. "Epistemic Democracy: Generalizin g the Condorcet Jury Theorem." Journal of Political Philosophy 9 (3):277 306. List, Christian, Robert C. Luskin, James S. Fishkin, and Iain McLean. 2013. "Deliberation, Single Peakedness, and the Possibility of Meaningful Democracy: Evidence from Deliberative Polls." Journal of Politics 75 (1):80 95. doi: 10.1017/s0022381612000886. Lo, Alex, Kim Alexander, Wendy Proctor, and Anthony Ryan. 2013. "Reciprocity as Deliberative Capacity: Lessons from a Citizen's Deliberation on Carbon Pricing Mechanisms in Australia." Environment and Planning C: Government and Policy 31 (3):444 459. Longino, Helen. 2002. The Fate of Knowledge Princeton: Princeton University Press. Mallory, Bruce L. 2010. "Practicing what we preach: Democratic practices in institutional governance." New Directions for Higher Education 2010 (152):91 97. doi: 10.1002/he.417. Manin, Bernard. 1987. "On Legitimacy and Political Deliberation." Political Theory 15 (3):338 368. Manin, Bernard. 2005. "Democratic Deliberation: Why We Should Promote Debate Rather Than Discussion." Program in Ethics and Public Affairs Seminar, Princeton University. Mansbridge, Jane. 1980. Beyond Adversary Democracy Chicago: University of Chicago Press. Mansbridge, Jane. 1999. "Everyday Talk in the Deliberative System." In Deliberative Politics: Essays on Democracy and Disagreement edited by Stephen Macedo, 211 239. Oxford: Oxford University Press. Mansbridge, Jane. 2007a. ""Deliberative Democracy" or "Democratic Deliberation?"." In Deliberation, Participation, and Democracy: Can the Pe ople Govern? edited by Shawn Rosenberg, 251 271. New York: Palgrave. Mansbridge, Jane. 2007b. "Self Interest in Deliberation." The Kettering Review (Winter):62 72. Mansbridge, Jane. 2015. "A Minimalist Definition of Deliberation." In Democracy and Develo pment: Rethinking the Role of Voice and Collective Action in Unequal Societies edited by Patrick Heller and Vijayendra Rao, 27 50. Washington, DC: World Bank Group. Mansbridge, Jane, James Bohman, Simone Chambers, Thomas Christiano, Archon Fung, John Park inson, Dennis F Thompson, and Mark E Warren. 2012. "A Systemic Approach to Deliberative Democracy." In Deliberative Systems edited
199 by John Parkinson and Jane Mansbridge, 1 26. New York: Cambridge University Press. Mansbridge, Jane, James Bohman, Simone Ch ambers, David Estlund, Andreas Fllesdal, Archon Fung, Cristina Lafont, Bernard Manin, and Jos Luis Mart. 2010. "The Place of Self Interest and the Role of Power in Deliberative Democracy." Journal of Political Philosophy 18 (1):64 100. doi: 10.1111/j.1 467 9760.2009.00344.x. Mart, Jos Luis. 2006. "The Epistemic Conception of Deliberative Democracy Defended: Reasons, Rightness and Equal Political." In Democracy and Its Discontents: National and Post national Challenges edited by Samantha Besson and Jos Luis Mart, 27 56. Burlington: Ashgate. Mathews, David. 2005. "Listening to the Public: A New Agenda for Higher Education?" In Higher Education for the Public Good: Emerging Voices from a National Movement edited by Adrianna J Kezar, Anthony Chambers an d John Burkhardt, 71 86. San Francisco: Jossey Boss. Mathews, David. 2015. "Har Megiddo: A Battle for the Soul of Higher Education." In Democracy's Education: Public Work, Citizenship, and The Future of Colleges and Universities edited by Harry C. Boyte. Nashville: Vanderbilt University Press. Mendelberg, Tali. 2002. "The Deliberative Citizen: Theory and Evidence." Political Decision Making, Deliberation and Participation 6 (1):151 193. Mercier, Hugo, and H l ne Landemore. 2012. "Reason Is for Arguing: Understanding the Successes and Failures of Deliberation." Political Psychology 33 (2):243 258. Mercier, Hugo, and Dan Sperber. 2011. "Why Do Humans Reason? Arguments for an Argumentative Theory." Behavioral and Brain Sciences 34 (2):57 74. Mercier, Hugo and Dan Sperber. 2012. ""Two Heads Are Better" Stands to Reason." Science 336 (6084):979. Merriam Webster. 2015. Deliberation. In Merriam Webster Dictionary : http://www.merriam webst er.com/dictionary/deliberation Mill, John Stuart. 1859/1993. On Liberty : Everyman. Miller, Boaz. 2013. "When Is Consensus Knowledge Based? Distinguishing Shared Knowledge From Mere Disagreement." Synthese 190:1293 1316. Minor, James T. 2004. "Understanding Faculty Senates: Moving from Mystery to Models." The Review of Higher Education 27 (3):343 363.
200 Moore, Alfred. 2014. "Democratic Reason: Book Review." Contemporary Political Theory 13:e12 e15. doi: 10.1057/cpt.2013.26 Moore, Alfred, and Kieran O'Doherty. 2014. "Deliberative Voting: Clarifying Consent in a Consensus Process." Journal of Political Philosophy 22 (3):302 319. doi: 10.1111/jopp.12028. Mutz, Diana C. 2006. Hearing the Other Side: Deliberative versus Partic ipatory Democracy Cambridge: Cambridge University Press. National Task Force on Civic Learning and Democratic Engagement 2012. "A Crucible Moment: College Learning and Democracy's Future." Association of American Colleges and Universities. Accessed June 2015. http://www.aacu.org/sites/default/files/files/crucible/Crucible_508F.pdf NCES. 2013. "Tuition costs of colleges and universities." US Department of Education Accessed March 2016 http://nces.ed.gov/FastFacts/display.asp?id=76 NCES. 2016. "Tuition Costs of Colleges and Universities." US Department of Education. Accessed June 2016. http://nces.ed.gov/FastFacts/display.asp?id=76 Nelson, William. 2008. "The Epistemic Value of the Democratic Process." Episteme 5 (1):19 32. doi: 10.3366/E174236000800021X. Nickerson, Raymond S. 1998. "Confi rmation Bias: A Ubiquitous Phenomenon in Many Guises." Review of General Psychology 2 (2):175 220. Niemeyer, Simon. 2011. "The Emancipatory Effect of Deliberation: Empirical Lessons from Mini Publics." Politics & Society 39 (1):103 140. Niemeyer, Simon, and John S. Dryzek. 2007. "The Ends of Deliberation: Meta consensus and Inter subjective Rationality as Ideal Outcomes." Swiss Political Science Review 13 (4):497 526. Nino, Carlos. 1996. The Constitution of Deliberative Democracy New Haven: Yale University Press. Nussbaum, Martha. 2010. Not For Profit: Why Democracy Needs the Humanities Princeton: Princeton University Press. ot Majority Rule." Constella tions 15 (1):3 9. Ober, Josiah. 2009. Epistemic Democracy in Classical Athens: Sophistication, Diversity, and Innovation ." Princeton/Stanford Working Papers in Classics. Ober, Josiah. 2010. Democracy and Knowledge: Innovation and Learning in Classical At hens Princeton: Princeton University Press.
201 Ober, Josiah. 2012. Relevant Expertise Aggregation: An Aristotelian Middle Way for Epistemic Democracy ." Princeton/Stanford Working Papers in Classics. Ober, Josiah. 2013. "Democracy's Wisdom: An Aristotelian Middle Way for Collective Judgment." American Political Science Review 107 (1):19p. doi: 10.1017/s0003055412000627. Ostrom, Elinor. 1996. "Covenants, Collective Action, and Common Pool Resources." In The Constitution of Good Societies edited by Karol Edw ard Soltan and Stephen L Elkin, 23 38. University Park: Pennsylvania State University Press. Oxford. 2015. Deliberation. In Oxford English Dictionary : http://www. oxforddictionaries.com/us/definition/american_english/deliberation Page, Scott E. 2007. The Difference: How the Power of Diversity Creates Better Firms, Schools, Groups, and Societies Princeton: Princeton University Press. Peter, Fabienne. 2007. "Democr atic Legitimacy and Proceduralist Social Epistemology." Politics, Philosophy & Economics 6 (3):329 252. Peter, Fabienne. 2009. Democratic Legitimacy New York: Routledge. Peter, Fabienne. 2013. "The Procedural Epistemic Value of Deliberation." Synthese 1 90:1253 1266. Pierce, Susan Resneck. 2014. Governance Reconsidered: How Boards, Presidents, Administrators, and Faculty Can Help Their Colleges T hrive San Francisco: Jossey Bass. Plato. 1945. The Republic of Plato Edited by Francis M Cornford. Oxford: Oxford University Press. Pope, Myron. 2004. "A Conceptual Framework of Faculty Trust and Participation in Governance." In New Directions for Higher Education edited by William G Tierney and Vicente M Lechuga, 75 84. Rawls, John. 1971. A Theory of Justice Cambridge: Belknap Press of Harvard University Press. Rawls, John. 1993. Political Liberalism New York: Columbia University Press. Rawls, John. 2001. Justice as Fairness: A Restatement Cambridge: Harvard University Press. Rawls, John. 2002. "The Idea of Public Reason Revisited." In The Law of Peoples Cambridge: Harvard University Press.
202 Raz, Joseph. 1990. "Facing Diversity: The Case of Epistemic Abstinence." Philosophy & Public Affairs 19 (1):3 46. Rees, Jonathan. 2014. "More than MOOCs." Academe (May June). Richardson, Henry S. 2010. "Public Opinion and Popular Will." In Deliberative Democracy in Practice edited by David Kahane, Daniel Weinstock, Dominique Leydet and Melissa Williams, 177 193. Vancouver: UCB Press. Riker, William H. 1982. Liberalism Against Populism: A Confrontation Between the Theory of Democracy and the Theory of Social Choice San Francisco: W.H. Freeman. Rittel, Horst, and Melvin Webber. 1973. "Dilemmas in a General Theory of Planning." Policy Sciences 4:155 169. Rodrigues, Adria n. 2014. "Faculty Governance: A Power Within." Yale Daily News Accessed May 2016. http://yaledailynews.com/blog/2014/04/22/faculty governance a power within/ Rosen berg, Brian. 2014. "Essay on a New Approach to Shared Governance in Higher E ducation." Inside Higher Ed Accessed February 19 2015 h ttps://www.insidehighered.com/views/2014/07/29/essay new approach shared governance higher education Roth, Michael. 2015. "Essay on the state of governance and faculty administration relations." Accessed February 17 2015 https://www.insidehighered.com/views/2015/02/05/essay state governance and faculty administration relations Rousseau, Jean Jacques. 1762/1993. The Social Contract London: Everyman. Ruiz, Diego. 2012. "Russian Studies to Stay." The Mac Weekly Accessed October 2015. http://themacweekly.com/2012/03/russian studies to stay/ Saffon, Maria Paula, and Nadia Urbinati. 2013. "Procedural Democracy, the Bulwark of Equal Liberty." Political Theory 41 (3):41p. doi: 10.1177/0090591713476872. Saltmarsh, John A, and Matthew Hartley. 2011. "To Serve a Larger Purpose": Engageme nt for Democracy and the Transformation of Higher Education Philadelphia: Temple University Press. Saltmarsh, John A, Emily M Janke, and Patti H Clayton. 2015. "Transforming Higher Education Through and For Democratic Civic Engagement: A Model for Change. Michigan Journal of Community Service Learning Fall:122 127. Sanders, Lynn. 1997. "Against Deliberation." Political Theory 25 (3):347 376.
203 Schauer, Frederick. 1999. "Talking as a Decision Procedure." In Deliberative Politics: Essays on Democracy and Di sagreement edited by Stephen Macedo, 17 27. Oxford: Oxford University Press. Schmidt, Benno. 2014. "Governance for a New Era: A Blueprint for Higher Education T rustees." American Council of Trustees and Alumni Accessed February 26 2015 http://www.goacta.org/images/download/governance_for_a_new_era.pdf Schram, Sanford. 2014. "The Future of Higher Education and American Democracy: Introduction." New Political Science 36 ( 4):425 437. Schrecker, Ellen. 2010. The Lost Soul of Higher Education: Corporatization, the Assault on Academic Freedom, and the End of the American university New York: The New Press. Schulz Hardt, Stefan, Felix C Brodbeck, Andreas Mojzisch, Rudolf Kersc hreiter, and Dieter Frey. 2006. "Group Decision Making in Hidden Profile Situations: Dissent as a Facilitator for Decision Quality." Journal of Personality and Social Psychology 91 (6):1080 1093. Schwartzberg, Melissa. 2015. "Epistemic Democracy and its C hallenges." Annual Review of Political Science 18 (May):187 203. Shumar, Wesley. 1997. College for Sale: A Critique of the Commodification of Higher Education London: Falmer Press. Silver, Nate. 2015. "Donald Trump Is Winning The Polls And Losing The N omination." The New York Times Accessed October 2015. http://fivethirtyeight.com/datalab/donald trump is winning the polls and losing the nomination/ Simha, OR. 2011. "MIT 2030: Concerns for the Future." MIT Faculty Newsletter Smith, Graham, and Matt Ryan. 2014. "Defining Mini Publics." In Deliberative Min i Publics: Involving Citizens in the Democratic Process edited by Kimmo Gronlund, Andre Bachtiger and Maija Setala. ECPR Press. Sperber, Dan, Fabrice Clment, Christophe Heintz, Olivier Mascaro, Hugo Mercier, Gloria Origgi, and Deirdre Wilson. 2010. "Epis temic Vigilance." Mind and Language 25 (4):359 393. Sperber, Dan, and Hugo Mercier. 2012. "Reasoning as a Social Competence." In Collective Wisdom: Principles and Mechanics edited by H l ne Landemore and Jon Elster, 368 386. Cambridge: Cambridge Univer sity Press. Stains, Robert R. 2014. "Repairing the Breach: The Power of Dialogue to Heal Relationship and Communities." Journal of Public Deliberation 10 (1).
204 Stasser, Garold, and William Titus. 1985. "Pooling of Unshared Information in Group Decision Mak ing: Biased Information Sampling During Discussion." Journal of Personality and Social Psychology 48 (6):1467 1478. Stensaker, Bjrn, and Agnete Vab. 2013. "Re inventing Shared Governance: Implications for Organisational Culture and Institutional Leaders hip." Higher Education Quarterly 67 (3):256 274. doi: 10.1111/hequ.12019. Straumsheim, Carl. 2014. "U. of Florida political science department declines to build a fully online program." Inside Higher Ed Accessed June 2016. https://www.insidehighered.com/news/2014/11/10/u florida political science department declines build fully online degree Sunstein, Cass R. 2002. "Th e Law of Group Polarization." Journal of Political Philosophy 10 (2):175 195. doi: Doi 10.1111/1467 9760.00148. Surowiecki, James. 2005. The Wisdom of Crowds New York: Doubelday Press. Tan, Chorh Chuan. 2013. "The Changing Nature and Character of Researc h Universities." In Preparing Universities for an E ra of C hange edited by Luc E Weber and James J Duderstadt. London: Economica. Tetlock, Philip. 1983. "Accountability and Complexity of Thought." Journal of Personality and Social P sychology 45 (1):74 83. Tetlock, Philip. 1985. "Accountability: The Neglected Social Context of Judgment and Choice." Research in Organizational Behavior 7:297 332. Theoria 63 (6):35 52. Thoenig, Jean Claude, and Catherine Paradeise. 2014. "Organizational Governance and the Production of Academic Quality: Lessons from Two Top U.S. Research Universities." Minerva 52 (4):381 417. doi: 10.1007/s11024 014 9261 2. Thomas, Nancy L., and Peter Levine. 2011. "Deliberative Democracy and Higher Education: Higher Education's Democratic Mission." In "To Serve a Larger Purpose": Engagement for Democracy and the Transformation of Higher Education edited by John A. Saltmarsh and Ma tthew Hartley, 154 176. Philadelphia: Temple University Press. Thompson, Dennis. 2008. "Deliberative Democratic Theory and Empirical Political Science." Annual Review of Political Science 11:497 520. Thompson, Paul. 2005. "University Governance and the Ac countability of Academic Administrators." Journal of Academic Ethics 2 (3):187 197. doi: 10.1007/s10805 005 2983 x.
205 Tierney, Wi lliam G. 1988. "Organizational Culture in Higher E ducation: Def ining the E ssentials." The Journal of Higher Education :2 21. Tie rney, William G. 2004. Competing Conceptions of Academic Governance: Negotiating the Perfect S torm Baltimore: Johns Hopkins University Press. Tierney, William G. 2006. Trust and the Public Good: Examining the Cultural Conditions of Academic W ork New York : Peter Lang. Tierney, William G., and James T. Minor. 2004. "A Cultural Perspective on Communication and Governance." In New Directions for Higher Education edited by William G Tierney and Vicente M Lechuga, 85 94. Tisch College for Citizenship and Publi c Service 2014. White House Summit on Civic Learning and National Service. Tufts University. UF. 2016. "Graphic Representation of an Example of Shared Governance." University of Florida. Accessed June 2015. http://www.senate.ufl.edu/Data/Sites/20/media/aboutus/sharedgovernancegraphi crepresentation.pdf Urfalino, Philippe. 2010. "Deciding as Bringing Deliberation to a Close." Social Science Info rmation 49 (1):111 140. Urfalino, Philippe. 2012. "Reasons and Preferences in Medicine Evaluation Committees." In Collective Wisdom: Principles and Mechanics edited by H l ne Landemore and Jon Elster, 173 202. Cambridge: Cambridge University Press. Urfalino, Philippe. 2014. "The Rule of Non Opposition: Opening Up Decision Making by Consensus." Journal of Political Philosophy 22 (3):320 341. Van de Ven, Andrew. 2008. Engaged Scholarship: A Guide for Organizational and Social Research New York: Oxfor d University Press. van Hees, Martin. 2007. "The Limits of Epistemic D emocracy." Social Choice & Welfare 28 (4):649 666. doi: 10.1007/s00355 006 0185 0. Wallace, R. Jay. 2014. "Practical Reason." In The Stanford Encyclopedia of Philosophy (Summer 2014 Edi tion) edited by Edward N Zalta. http://plato.stanford.edu/archives/sum2014/entries/practical reason/ Walsh, Katherine Cramer. 2007. "The Democratic Potential of Civic D ialogue." In Deliberation, Participation, and Democracy: Can the People Govern? edited by Shawn Rosenberg, 45 63. London: Palgrave Macmillan. Warren, Mark E., Jane Mansbridge, Andre Bchtiger, Maxwell Cameron, Simone Chambers, John Ferejohn, Alan Jacobs, Jack Knight, Daniel Naurin, Melissa Schwartzberg, Yael Tamir, Dennis Thompson, and Melissa Williams. 2013.
206 "Deliberative Negotiation." In Negotiating Agreement in Politics: Report on the Task Force on Negotiating Agreement in Politics edited by Jane Mansb ridge and Cathy Jo Martin, 86 120. American Political Science Association. Warren, M ark E., and Hilary Pearse. 2008 Designing Deliberative Democracy: T he British Columbia Citizens' Assembly Theories of Institutional D esign New York: Cambridge University Press. Wattenberg, Martin P. 1991. The Rise of Candidate Centered Politics: Presidential Elections of the 1980s Cambridge: Harvard Univ Press. Wattenberg, Martin P. 2004. "Elections: Personal Popularity in US Presidential Elections." Presidential Studie s Quarterly 34 (1):143 155. Weber, Luc E, and James J Duderstadt. 2004. Reinventing the Research University London: Economica. Weber, Luc E, and James J Duderstadt. 2013. Preparing Universities For an Era of Change London: Economica. Weisberg, Michael, a nd Ryan Muldoon. 2009. "Epistemic Landscapes and the Division of Cognitive Labor." Philosophy of Science 76 (2):225 252. Westheimer, Joel, and Joseph Kahne. 2004. "What Kind of Citizen? The Politics of Educating for Democracy." American Educational Research Journal 41 (2):237 269. Wittgenstein, Ludwig. 1958. Philosophical Investigations : Basil Blackwell. Wood, Graeme. 2014. "The Future of College?" The Atlantic Monthly 314(2): 50 60. Woodhouse, Kellie. 2015. "Who's to Blame for Rising Tuition?" Inside Higher Ed Accessed November 2016. https://www.insidehighered.com/news/2015/05/05/report says administ rative bloat construction booms not largely responsible tuition Yankelovich, Daniel. 1999. The Magic of Dialogue: Transforming Conflict into C ooperation New York: Simon & Schuster. Young, Iris Marion. 1996. "Communication and the Other: Beyond Deliberat ive Democracy." In Democracy and Difference: Contesting the Boundaries of the Political edited by Seyla Benhabib, 120 135. Princeton: Princeton University Press. Young, Iris Marion. 2000. Inclusion and Democracy Oxford: Oxford University Press.
207 BIOGRAPHICAL SKETCH Christopher Manick graduated from the University of Florida He earned a Ph.D. in political science. His work specializes in political theory with a focus on contemporary theories of democracy and citizenship.