IN THIS ISSUE

Three of this issue's research articles deal with issues of war, within or between states; four with the behavior of democratic citizens; two with corruption in democracies; and one with the effect of quotas on representation. The issue concludes with an enlightening forum, in which Jane Mansbridge and Andrew Rehfeld exchange, qualify, and modify their views about concepts of representation, a conversation inspired by Rehfeld's earlier article in these pages, “Representation Rethought” (APSR 2009).

Conflict within and between states. In our lead article, “The Cost of Empty Threats: A Penny, Not a Pound,” Jack Snyder and Erica D. Borghard confront the theory of audience costs (notably Fearon's), and experimental work on that subject (e.g., Tomz), with the historical evidence of post-1945 diplomatic crises, particularly ones in which threats that turned out to be empty were actually made, or partially executed (notable case: Suez). They see little evidence that leaders who made such empty threats, or who backed down, suffered a loss of popularity or power. Domestic audiences appear to care more about the substance of policy in a crisis (e.g., they may prefer a “hawkish” or a “dovish” policy) and much less about whether empty threats were issued.

But what of leaders who actually initiate wars, and then lose them? Such “culpable” leaders are indeed punished by domestic audiences, Sarah E. Croco finds in “The Decider's Dilemma: Leader Culpability, War Outcomes, and Domestic Punishment,” as indeed are ones who take over leadership in the middle of a war, share a political affiliation with the leader who initiated it, and then preside over a defeat. Nonculpable leaders—ones who did not initiate a war or continue an aggression—remain unpunished by their domestic audiences. This finding, based on a new data set, suggests that leaders have less incentive to initiate a war, unless they are certain of winning it.

Turning to civil war, Lars-Erik Cederman, Nils B. Weidmann, and Kristian Skrede Gleditsch find new evidence for “grievance,” as opposed to “greed,” poverty, or terrain, as a cause of domestic conflict. Exploiting a new data set on the spatial location and wealth of ethnic groups (and not merely of individuals) throughout the world, they find—in “Horizontal Inequalities and Ethnonationalist Civil War: A Global Comparison”— that ethnically unequal societies are indeed more liable to civil war; But, surprisingly, both the wealthiest and the poorest ethnic groups in such unequal societies are likelier to initiate conflict. While their empirical result appears to be robust, they offer no explanation for the pattern, leaving it as a mystery for further theoretical work.

Democratic citizenship and voting. One concern about democratic citizenship pervades much recent work on public opinion and voting, namely that most citizens willingly follow partisan elites' “cues” on issues, to the exclusion of information they might independently acquire. This result, John G. Bullock establishes in “Elite Influence on Public Opinion in an Informed Electorate,” holds only when citizens lack independent information. Two experiments show that when voters have even a modicum of information on a subject, what they know influences their attitudes at least as much as the cues they receive. Neither do party cues seem to inhibit citizens' propensity to think about issues. All of this suggests reasons for a bit more optimism about (and for easier dissemination of information among) democratic citizens.

In a similar vein, formal theories of voting have been divided on whether full sharing of otherwise private information and preferences conduces to better collective decisions; and, if it does, what voting rule is likelier to elicit such information. In contrast to some recent work, Jérôme Mathis demonstrates in “Deliberation with Evidence,” that our usual intuition—that a unanimity rule is most conducive to full revelation of preferences and private information—is correct, at least to the extent that individual voters can (even if they need not) provide verifiable evidence of their private information. As in the previous article, more information is better; but unanimity rule is likelier to provide more information.

We know a lot about the preferences of U.S. representatives, senators, and justices (not to mention similar national institutions outside the Unites States), and indeed can map their ideologies over time and across chambers. Perhaps surprisingly, much less is known about the ideologies of state legislators in the United States, even though comparisons among states have proven highly fruitful in other fields of inquiry. In “The Ideological Mapping of American Legislatures,” Boris Shor and Nolan McCarty go far toward greater understanding in this area, not least because they are the first to have collated all roll-call data for every state legislature since the mid-1990s and to have melded those data with a recurring survey of state legislative candidates. They have thus obtained, at least for this period, a complete ideological mapping of state legislatures (including individual chambers) that can be compared with similar studies of the U.S. Congress.

So better information makes for better voting and deliberation, and we will now have more information about legislators' ideologies. But do democracies choose better leaders? By one standard they clearly do, argue Timothy Besley and Marta Reynal-Querol in “Do Democracies Select More Educated Leaders?” Exploiting a new data set on world leaders between 1848 and 2004, and controlling for virtually all other possible influences, Besley and Reynal-Querol find that, even within countries, democracies (periods of democratic rule) select leaders with more formal education. To those of us who recall that Abraham Lincoln had less than a year of formal schooling, this is not necessarily a good thing; but the finding appears to be highly robust.

Democratic corruption. We have long assumed that political patronage helps parties and “machines” to control their electorates, and thus to maintain power, but solid evidence has been lacking and problems of potential endogeneity abound (e.g., are parties with stable majorities likelier to gain control of patronage?). In “Patronage and Elections in U.S. States,” Olle Folke, Shigeo Hirano, and James M. Snyder Jr. find strong evidence that control of patronage by state-level parties in the United States considerably raised the probability that those parties would win control of their state legislatures and their state-level executive offices. Exploiting the wide variation in the timing of civil service reform among the states, and employing a differences-in-differences design, Folke, Hirano, and Snyder establish that this result is robust to alternative explanations, and that, with regard to patronage, “practice made perfect.” The longer a party remained in control, the more effectively it could use patronage.

A recurring question in many countries, not least of all the United States, and not only in the present, is the extent of “voter fraud,” i.e., casting of ballots by voters not legally qualified to do so.Footnote ¹ In Japan, it has long been alleged that in local elections, and particularly in tight races, parties exploit somewhat elastic regulations on residency to “ship in” just before voting day adherents who actually reside elsewhere. In “Making Outsiders' Votes Count: Detecting Electoral Fraud through a Natural Experiment,” Kentaro Fukumoto and Yusaku Horiuchi exploit a seemingly randomized assignment of voting dates for municipal elections to establish both the existence and the magnitude of this kind of voter fraud in Japan. Through a quirk in the relevant Japanese electoral law, a seemingly randomized subsample of municipalities was required to elect municipal officers “out of sync” in April 2003. Precisely in those municipalities, and particularly where the election seemed likely to be close, new “residents” enrolled in much higher numbers just before the election than in municipalities that had no such election. The presumably fraudulent voters were numerous enough to have determined the outcome in many close municipal elections. Fukumoto and Horiuchi go on to consider the broader implications of their natural experiment for historical and present-day investigations of voter fraud, and for the design of electoral registration requirements.

Effect of quotas. Continuing a conversation that began in these pages with Rikhil Bhavnani's article (APSR 2009) on the lasting effect of quotas for female candidates in Mumbai, Melanie M. Hughes undertakes—in “Intersectionality, Quotas, and Minority Women's Representation Worldwide”—an examination of the effects of quotas at the national level (whether imposed by law or by the statutes of individual parties), for the representation of minorities, of women, or of both. Hughes pursues in particular the question of whether the two kinds of quotas might be at odds: An ethnic quota may actually reduce the representation of women (encouraging male minority candidates to run), or a quota for women may equally reduce the representation of minorities (favoring particularly female candidates of the dominant ethnicity). Her evidence, drawn from a large new data set that she has constructed on national parliaments, parties, and ethnicities, suggests that the former holds with more force than the latter. Quotas rarely challenge majority men's domination of the process, and only in the rare case of what she calls “tandem quotas,” requiring the election both of minorities and of women, does any great change in representation seem to occur.

Forum on Representation. Finally, in our forum on representation, Jane Mansbridge and Andrew Rehfeld engage in a thoughtful, civil, and (not least for those reasons) extremely fruitful conversation about Rehfeld's earlier Review article, which had as one of its main points a critique and refinement of Mansbridge's concept of “gyroscopic” representation. Mansbridge had defined gyroscopic representation as the deliberate choice by principals (voters) of an agent (representative) whose preferences were so fixed and well-known that principals could exert less effort in monitoring. (An obvious parallel, from political economy, is Kenneth Rogoff's argument for choosing a central banker known to have a strong anti-inflation bias, indeed one stronger than the median principal who appoints her). To Rehfeld, such a concept conflated two criteria that needed to be kept distinct: the agent's fixed pursuit of her own established principles, in doubtful cases exercising her own judgment; and the agent's indifference to possible sanctions (e.g., non-reelection if principals' preferences change—in the political economy case, principals might suddenly decide that they want more inflation after all). Here, in “Clarifying the Concept of Representation,” Mansbridge augments “gyroscopic” representation by recourse to the “selection model,” an alternative in principal-agent work to the “sanction model,” and denies that we gain much purchase by separating what Rehfeld calls the “judgment” and “responsiveness” parts of a selection model of representation. What might be the “off-diagonal” cells of such a 2×2 matrix, she argues, are empirically (and probably logically) empty: Principals would not trust the independent judgment of an agent who yielded readily to sanctions, nor would an agent who was indifferent to sanctions fail to exercise independent judgment. Rehfeld, while conceding that such combinations may be empirically rare, denies that they are logically empty, and indeed presents hypothetical cases of both in his forum response,“The Concepts of Representation.”

Generalizing from this case, Mansbridge argues strongly for (and has consistently practiced in her own work) a melding of theoretical and empirical work on representation. Indeed, we found it impossible to read this exchange without thinking immediately of the current (as we write) conflict in Washington over raising the debt ceiling. Are the “tea party” Republicans really “gyroscopes,” chosen by their constituents precisely because they were known to be unwavering on such an issue, and indifferent to whether such a principled stand endangers their reelection? Or are they sacrificing their own judgment (which might push toward compromise, or at least avoidance of default) to the threat that, in the next primary, compromisers may be faced with an even more intransigent tea-party challenger? What empirical tests could distinguish between the two motivations, or the relative weights that a given representative might assign them?Footnote ² Extending the argument even further, what do we make of the empirical finding (albeit contested) that representatives are often ideologically more extreme than their constituents (Bafumi and Herron 2010, in the APSR).

To us, this important exchange again buttresses the importance of conversations that extend across political science, in this case, exactly as Mansbridge and Rehfeld argue, engaging both normative and empirical students of representation and deliberation. Much the same “boundary crossing” characterizes other articles in this issue: Qualitative or quantitative evidence (respectively, Snyder and Borghard; Croco) is brought to bear on theoretical and experimental work; experimental evidence appears to contradict survey findings (Bullock); and our taken-for-granted intuitions are tested theoretically (Mathis), historically (Folke, Hirano, and Snyder; Besley and Reynal-Querol), and empirically (Fukumoto and Horiuchi; Hughes). Such endeavors represent, in our view, the virtue and continuing indispensability of a general and widely read journal in political science. Readers (or, at the very least, authors) appear to agree. The APSR, we are happy to report, has resumed its position as the highest-ranked journal in terms of “impact factor” in the latest Thomson-Reuters ratings—released earlier this year—and indeed, at 3.278, has equaled its previous peak, achieved in 2005.

INSTRUCTIONS TO CONTRIBUTORS

The American Political Science Review (APSR) publishes scholarly research of exceptional merit, focusing on important issues and demonstrating the highest standards of excellence in conceptualization, exposition, methodology, and craftsmanship. A significant advance in understanding of politics—whether empirical, interpretive, or theoretical—is the criterion for publication in the Review. Because the APSR reaches a diverse audience, authors must demonstrate how their analysis illuminates or answers an important research question of general interest in political science. For the same reason, authors must strive to be understandable to as many scholars as possible, consistent with the nature of their material.

The APSR publishes original work. Submissions should not include tables, figures, or substantial amounts of text that already have been published or are forthcoming in other places. In many cases, re-publication of such material would violate the copyright of the other publisher. Neither does the APSR consider submissions that are currently under review at other journals or that duplicate or overlap with parts of larger manuscripts submitted to other publishers (whether of books, printed periodicals, or online journals). If you have any questions about whether these policies apply in your case, you should address the issues in a cover letter to the editors or as part of the author comments section during online submission. You should also notify the editors of any related submissions to other publishers, whether for book or periodical publication, during the pendency of your submission's review at the APSR—regardless of whether they have yet been accepted. The editors may request copies of related publications.

The APSR uses a double-blind review process. You should follow the guidelines for preparing an anonymous submission in the “Specific Procedures” section that follows.

If your manuscript contains quantitative evidence and analysis, you should describe your procedures in sufficient detail to permit reviewers to understand and evaluate what has been done and—in the event the article is accepted for publication—to permit other scholars to replicate your results and to carry out similar analyses on other data sets. With surveys, for example, provide sampling procedures, response rates, and question wordings; calculate response rates according to one of the standard formulas given by the American Association for Public Opinion Research, Standard Definitions: Final Dispositions of Case Codes and Outcome Rates for Surveys (Lenexa, KS: AAPOR, 2006).Footnote ³ For experiments, provide full descriptions of experimental protocols, methods of subject recruitment and selection, payments to subjects, debriefing procedures, and so on. In any case involving human subjects, the editors may require certification of appropriate institutional review and/or conformity with generally accepted norms.Footnote ⁴

The strength of evidence necessary for publication of quantitative empirical findings cannot be captured by any single criterion, such as the conventional .05 level of statistical significance. The journal's co-editors—following the evolving disciplinary standard among reviewers—will evaluate the strength of findings on a range of criteria beyond statistical significance, including substantive significance, theoretical aptness, the importance of the problem under study, and the feasibility of obtaining additional evidence.

In addition, authors of quantitative or experimental articles are expected to address the issue of data availability. You must normally indicate both where (online) you will deposit the information that is necessary to reproduce the numerical results and when that information will be posted (such as “on publication” or “by [definite date]”). You should be prepared, when posting, to provide not only the data used in the analysis but also the syntax files, specialized software, and any other information necessary to reproduce the numerical results in the manuscript. Where an exception is claimed, you should clearly explain why the data or other critical materials used in the manuscript cannot be shared, or why they must be embargoed for a limited period beyond publication.

Similarly, authors of qualitative, observational, or textual articles, or of articles that combine such methods with quantitative analysis, should indicate their sources fully and clearly enough to permit ready verification by other scholars, including precise page references to any published material cited and clear specification (e.g., file number) of any archival sources. Wherever possible, use of interactive citations is encouraged. Where field or observational research is involved, anonymity of participants will always be respected; but the texts of interviews, group discussions, observers' notes, etc., should be made available on the same basis (and subject to the same exceptions) as with quantitative data.

Articles should be self-contained; you should not simply refer readers to other publications for descriptions of these basic research procedures.

Please indicate variables included in statistical analyses by italicizing the entire name of the variable—the first time it is mentioned in the text—and by capitalizing its first letter in all uses. You should also use the same names for variables in text, tables, and figures. Do not use acronyms or computational abbreviations when discussing variables in the text. All variables that appear in tables or figures should have been mentioned in the text, standard summary statistics (n, mean, median, standard deviation, range, etc.) provided, and the reason for their inclusion discussed. However, tables and figures should also be comprehensible without reference to the text (e.g., in any figures, axes should be clearly labeled). Please bear in mind also that neither the published or online versions of the Review normally can provide figures in color; be sure that a grayscale version will be comprehensible to referees and readers.

You may be asked to submit additional documentation if procedures are not sufficiently clear. If you advise readers that additional information is available on request, you should submit equally anonymous copies of that information with your manuscript as “supplemental materials.” If this additional information is extensive, please inquire about alternate procedures.

Manuscripts that, in the judgment of the co-editors, are largely or entirely critiques of, or commentaries on, articles previously published in the Review will be reviewed for possible inclusion in a forum section, using the same general procedures as for other manuscripts. Well before any publication, however, such manuscripts will also be sent to the scholar(s) whose work is being addressed. The author(s) of the previously published article will be invited to comment to the editors and to submit a rejoinder, which also will be peer-reviewed. We do not publish rejoinders to rejoinders.

The APSR accepts only electronic submissions (at www.editorialmanager.com/apsr). The web site provides detailed information about how to submit, what formatting is required, and what type of digital files may be uploaded. Please direct any questions to the journal's editorial offices at apsr@polisci.ucla.edu

ELECTRONIC ACCESS TO THE APSR

Back issues of the APSR are available in several electronic formats and through several vendors. Except for the last three years (as an annually “moving wall”), back issues of the APSR beginning with Volume 1, Number 1 (November 1906), are available on-line through JSTOR (http://www.jstor.org/). At present, JSTOR's complete journal collection is available only via institutional subscription, e.g., through many college and university libraries. For APSA members who do not have access to an institutional subscription to JSTOR, individual subscriptions to its APSR content are available. Please contact Member Services at APSA for further information, including annual subscription fees.

Individual members of the American Political Science Association can access recent issues of the APSR and PS through the APSA website (www.apsanet.org) with their username and password. Individual nonmember access to the online edition will also be available, but only through institutions that hold either a print-plus-electronic subscription or an electronic-only subscription, provided the institution has registered and activated its online subscription.

Full text access to current issues of both the APSR and PS is also available on-line by library subscription from a number of database vendors. Currently, these include University Microfilms Inc. (UMI) (via its CD-ROMs General Periodicals Online and Social Science Index and the on-line database ProQuest Direct), Online Computer Library Center (OCLC) (through its on-line database First Search as well as on CD-ROMs and magnetic tape), and the Information Access Company (IAC) (through its products Expanded Academic Index, InfoTrac, and several on-line services [see below]). Others may be added from time to time.

The APSR is also available on databases through six online services: Datastar (Datastar), Business Library (Dow Jones), Cognito (IAC), Encarta Online Library (IAC), IAC Business (Dialog), and Newsearch (Dialog).

The editorial office of the APSR is not involved in the subscription process to either JSTOR for back issues or the other vendors for current issues. Please contact APSA, your reference librarian, or the database vendor for further information about availability.

OTHER CORRESPONDENCE

The American Political Science Association's address, telephone, and fax are 1527 New Hampshire Avenue, N.W., Washington, D.C. 20036, (202) 483-2512 (voice), and (202) 483-2657 (fax). E-mail: apsa@apsanet.org. Please direct correspondence as follows.

Information, including news and notes, for PS:

• Dr. Robert J-P. Hauck, Editor, PS

• E-mail: rhauck@apsanet.org

Circulation and subscription correspondence (domestic claims for nonreceipt of issues must be made within four months of the month of publication; overseas claims, within eight months):

• Sean Twombly,

• Director of Member Services

• E-mail: membership@apsanet.org

• Reprint permissions:

  • E-mail: Rights@cambridge.org

• information and rates:

  • Advertising Coordinator,

  • Cambridge University Press

  • E-mail: journals_advertising@cambridge.org

EXPEDITING REQUESTS FOR COPYING APSR AND PS ARTICLES FOR CLASS USE AND OTHER PURPOSES

INDEXING

Articles appearing in the APSR before June 1953 were indexed in The Reader's Guide to Periodical Literature. Current issues are indexed in ABC Pol Sci; America, History and Life 1954–; Book Review Index; Current Contents: Social and Behavioral Sciences; EconLit; Energy Information Abstracts; Environmental Abstracts; Historical Abstracts; Index of Economic Articles; Information Service Bulletin; International Bibliography of Book Reviews of Scholarly Literature in the Humanities and Social Sciences; International Bibliography of Periodical Literature in the Humanities and Social Sciences; International Index; International Political Science Abstracts; the Journal of Economic Literature; Periodical Abstracts; Public Affairs; Public Affairs Information Service International Recently Published Articles; Reference Sources; Social Sciences and Humanities Index; Social Sciences Index; Social Work Research and Abstracts; and Writings on American History. Some of these sources may be available in electronic form through local public or educational libraries. Microfilm of the APSR, beginning with Volume 1, and the index of the APSR through 1969 are available through University Microfilms Inc., 300 North Zeeb Road, Ann Arbor, MI 48106 (www.umi.com). The Cumulative Index to the American Political Science Review, Volumes 63 to 89: 1969–95, is available through the APSA.