Week 2: What it means to represent someone

In plain words

Week 2 of the My Vote Forward seminar asks what the word “represent” actually means in politics. Students read Hanna Pitkin’s classic analysis, Jane Mansbridge’s 2003 article identifying four forms of representation, and Andrew Rehfeld’s general theory, and they test each against the idea of collecting public priorities between elections.

People say a representative “speaks for” them, “looks like” them, “stands for” them, or “acts for” them. These are different ideas. A representative can look like a district and vote against its wishes, or vote with its wishes and share nothing else with it. This week sorts out the meanings, so students can say precisely which kind of representation a proposal would improve.

Defined terms

• Formalistic representation: representation defined by the arrangements that authorize a representative and hold them to account, such as elections.
• Descriptive representation: a representative “standing for” a group by sharing its characteristics, such as background or experience.
• Symbolic representation: a representative “standing for” a group because the group accepts the representative as a symbol of it.
• Substantive representation (“acting for”): a representative acting in the interest of, and in some way responsive to, the represented.
• Promissory representation: Mansbridge’s name for the traditional model in which a representative makes promises in a campaign and is judged on keeping them.
• See also the glossary entries for representation, sortition, and citizens’ assembly.

Learning objectives

By the end of week 2, a participant should be able to:

1. Name and distinguish Pitkin’s formalistic, descriptive, symbolic, and “acting for” views of representation.
2. Explain the mandate and independence controversy and connect it to Burke (week 1).
3. Describe Mansbridge’s four forms of representation and her claim that the newer forms call for systemic, deliberative, and plural standards.
4. State Rehfeld’s argument that representation is created by an audience accepting someone as a representative, and why that makes representation “not a democratic phenomenon at all.”
5. For any proposed feedback channel, identify which form of representation it is meant to improve and which it might weaken.

Core readings

1. Hanna F. Pitkin, The Concept of Representation (University of California Press; first published 1967, paperback 1972). Introduction and the chapters on formalistic views, “standing for,” and the mandate and independence controversy.⁠¹
2. Jane Mansbridge, “Rethinking Representation,” American Political Science Review 97(4), 2003, pp. 515 to 528.⁠²
3. Andrew Rehfeld, “Towards a General Theory of Political Representation,” The Journal of Politics 68(1), 2006, pp. 1 to 21.⁠³
4. Edmund Burke, Speech to the Electors of Bristol (1774), reread in light of Pitkin’s chapter on Burke.⁠⁴

Lecture notes

Why define representation at all

Representation is the word every proposal in this course leans on, and week 2 slows down to define it. Reformers say a new channel will make government “more representative.” Critics say it will make representation worse. Both sides can be right, because they are often talking about different relationships. Hanna Pitkin’s The Concept of Representation, first published in 1967, is the standard starting point because it treats representation as a concept with several distinct meanings rather than a single thing that institutions have more or less of. The book describes itself as “primarily a conceptual analysis, not a historical study of the way in which representative government has evolved, nor yet an empirical investigation” of how representatives behave.⁠¹ That is exactly the tool this seminar needs before it evaluates any design.

The site’s chapter on representation states how the My Vote Forward framework positions itself. This week supplies the vocabulary to check whether that position is coherent.

Pitkin’s views

Pitkin’s table of contents is itself a map of the argument: formalistic views of representation; “standing for” as descriptive representation; “standing for” as symbolic representation; representing as “acting for”; the mandate and independence controversy; Burke’s representation of “unattached interests”; and liberal representation of “people who have interests”.⁠¹ The lecture summarizes each view as the seminar will use it. These summaries are the seminar’s paraphrase, and students should check them against the text.

The formalistic views define representation by its rules. On an authorization view, a representative is anyone authorized to act for others; on an accountability view, a representative is anyone who will later have to answer to them. Pitkin’s objection to both is that they say nothing about what the representative does in between. A formally authorized official who ignored everyone would still count as a representative.

“Standing for” views define representation by resemblance or symbolism. Descriptive representation asks whether a legislature mirrors the population, which is the intuition behind proposals for random selection, discussed in week 8. Symbolic representation asks whether people accept a figure as standing for them, whatever that figure does. Neither view, Pitkin argues, captures the activity of representing.

“Acting for” is the substantive view: representing means acting in the interest of the represented, in a manner responsive to them. This is where the mandate and independence controversy lives. Must the representative do what constituents say (mandate), or what the representative judges to be in their interest (independence)? Pitkin treats the controversy as a sign that representation contains a built-in tension: the represented must be in some sense present in the action, yet the representative must actually act.

Burke revisited

Burke appears in Pitkin’s account as the representative of “unattached interests,” interests that exist independent of anyone’s wishes and can be discerned by judgment. Burke’s line from week 1, that a representative “betrays, instead of serving you, if he sacrifices it to your opinion,” makes sense on this reading.⁠⁴ If interests are objective and judgment can discover them, then constituents’ opinions are evidence about interests, not instructions. The liberal view Pitkin contrasts with Burke holds that interests belong to people who have them, so the people themselves are the main authority on what those interests are.

The distinction matters for this course because any between-election channel collects what people say. On the liberal view, that is close to the thing representation is supposed to track. On the Burkean view, it is one input among several, and possibly a misleading one.

Mansbridge: four forms

Jane Mansbridge’s 2003 article updates the debate with forms of representation that empirical political scientists had observed. Her abstract names them: the traditional “promissory” form, and three others she calls “anticipatory,” “gyroscopic,” and “surrogate.” She argues that none of the newer forms “meets the criteria for democratic accountability developed for promissory representation,” yet each generates its own normative criteria, and that these criteria are “systemic, in contrast to the dyadic criteria appropriate for promissory representation,” “deliberative rather than aggregative,” and “plural rather than singular”.⁠²

The seminar’s working paraphrase of the four forms is as follows. Promissory representation is the familiar model: campaign promises, then judgment at the next election on whether they were kept. Anticipatory representation describes representatives who focus on what voters will reward at the next election, which can differ from what voters wanted at the last one; on this form, a representative has reason to educate and persuade, not only to follow. Gyroscopic representation describes representatives who act from their own internal principles, which voters chose when they elected them; voters influence policy by selection rather than by instruction. Surrogate representation describes representatives who speak for people outside their own district, such as those who share a cause or identity but have no representative of their own.

The point that carries into the rest of the course is Mansbridge’s claim that the standards are systemic. Under promissory representation, you can judge one representative by one district’s promises. Under the other forms, you have to judge the system: whether, across all legislators, communication between the public and representatives is good, whether the views that matter get voiced somewhere, and whether deliberation happens. That reframing helps a reader see what a new feedback channel could plausibly improve. It is unlikely to change gyroscopic representation, because gyroscopic representatives are chosen for their compass, not their responsiveness. It could improve anticipatory representation, by giving representatives and the public a shared, early view of how opinion is moving. It could support surrogate representation by making visible the views of people whose own representative does not share them.

Rehfeld: representation without democracy

Andrew Rehfeld pushes the concept further. He observes that nondemocratic actors are called representatives every day: dictatorships send “representatives” to the United Nations, and nongovernmental organizations claim to represent prisoners of war. Standard accounts tie representation to elections and to proper conduct such as deliberation and accountability, and so, he argues, cannot explain how these actors are representatives at all. His alternative explains representation “simply by reference to a relevant audience accepting a person as such.” When audiences use democratic rules of recognition, familiar democratic cases arise; when they use nondemocratic rules, nondemocratic cases arise. His conclusion is that “political representation, per se, is not a democratic phenomenon at all”.⁠³

Rehfeld’s account is useful precisely because it is deflationary. It separates two questions people often run together: whether someone is a representative, and whether they are a good democratic representative. Any new institution that publishes “what the public prioritizes” is making a representative claim about the public. Rehfeld’s theory asks who the relevant audience is that would accept that claim, and by what rules. A published priority list accepted by officials, journalists, and the public as “the public’s view” would function as a representative claim whether or not it deserved to.

Representation and the problems this site examines

The site’s problem chapters can be read through these lenses. The representation gap chapter concerns substantive representation: whether outcomes track what people want. Unequal participation concerns who gets to be present in the representative relationship at all. Gerrymandering concerns the formal arrangements that authorize representatives. Polarization affects anticipatory and gyroscopic representation, because it changes both what representatives expect voters to reward and which compasses voters select. A single reform rarely addresses all of these. Week 2 trains students to name which one a proposal targets.

What this means for priorities, opinions, and votes

The My Vote Forward framework separates three things: priorities, which subjects matter most; opinions, positions a person chooses to share; and the private vote. Pitkin’s and Mansbridge’s categories suggest why the separation might matter. A priority signal speaks most naturally to anticipatory representation: it tells representatives what voters are likely to judge them on. An opinion shared publicly is a form of instruction, closer to the mandate side of the old controversy, and it raises the question the site examines in what if someone changes position?. A secret vote is the formal, authorizing act. Students should ask whether keeping these apart preserves the strengths of each form or simply relabels the old tension.

A worked example: one vote, four readings

Hypothetical example: a city council member campaigned on expanding bus service, then votes for a budget that cuts two bus routes to fund road repair. How should constituents judge the vote? On a promissory reading, the member broke a campaign promise and voters should hold that against them at the next election. On an anticipatory reading, the member may be betting that by election day voters will care more about the repaired roads, and the question is whether the member explains that bet openly enough for voters to judge it. On a gyroscopic reading, the question is whether cutting routes is consistent with the principles voters knew the member held when they chose them. On a surrogate reading, the question is who outside the district relies on those routes and whether anyone on the council spoke for them.

Each reading points to different evidence. The promissory reading needs the campaign record. The anticipatory reading needs to know what voters will value later, which is exactly what a between-election priorities signal claims to offer. The gyroscopic reading needs the member’s stated principles. The surrogate reading needs to know who uses the routes. A single satisfaction score could not distinguish among these, which is why the course treats “is this official representative?” as several questions rather than one.

Discussion questions

1. Pitkin argues that formal authorization alone cannot define representation. Is there any official in your life whose only claim to represent you is formal? Does that matter?
2. Which of Mansbridge’s forms best describes how you would want your own representative to behave? Would you want the same form for a school board member and a national legislator?
3. If standards are systemic rather than dyadic, can a legislature be representative even if no single member represents your views? Is that acceptable?
4. Rehfeld says representation is created by an audience’s acceptance. Who would be the audience for a published list of “public priorities,” and what rules of recognition should it apply?
5. Does descriptive representation require elections at all? What do randomly selected bodies (week 8) offer and lose on this view?
6. Is a shared opinion a form of instruction to a representative? Should it be?

Assignment

Representation map (1,200 to 1,500 words). Choose one proposed or existing reform from the site’s chapters, for example participatory budgeting, a citizens’ assembly, ranked-choice voting, or the My Vote Forward priorities proposal. Using Pitkin’s views and Mansbridge’s four forms, state which kind of representation the reform aims to improve, which kind it leaves unchanged, and which kind it could weaken. Then use Rehfeld’s account to identify the audience whose acceptance would make the reform’s outputs function as representative claims. End with the strongest objection to your own analysis.

Evidence

This week’s material is conceptual. Descriptions of the readings are drawn from the publisher’s description and table of contents for Pitkin and from the published abstracts of Mansbridge and Rehfeld, all opened on the review date. The seminar’s paraphrases of Mansbridge’s four forms go beyond the abstract and are labeled as paraphrase; students should verify them against the article.

Competing interpretations

Pitkin’s analysis has been read as favoring substantive representation over descriptive representation; later theorists defend descriptive representation more strongly, especially for groups historically excluded from legislatures. Mansbridge’s systemic standards are welcomed by some as realistic and criticized by others as making accountability too diffuse to enforce, since no single representative can be held to a system-level standard. Rehfeld’s deflationary theory is praised for clarity and criticized for separating representation from the democratic values that make it worth caring about. The seminar does not choose among them.

Application to this framework

My Vote Forward claims that collecting stated priorities between elections can inform representatives without binding them. In Mansbridge’s terms, that is an attempt to strengthen anticipatory and system-level communication while leaving promissory and gyroscopic representation intact. The framework’s representation chapter should be tested on whether it says which form it targets, and whether it accounts for the risk Rehfeld’s theory highlights: that a published list could be accepted as speaking for “the public” when it speaks only for participants.

Unresolved questions

• Can systemic standards of representation be measured, or only argued about?
• Would a widely accepted priority signal turn anticipatory representation into something closer to mandate representation?
• Who, if anyone, should be the audience that decides whether a published aggregate counts as speaking for the public?