Political science as architecture

This article compares political science to another discipline, with which it has much in common. That discipline is architecture. The political-science-as-architecture analogy has a long history in political thought. It also has important implications for the ends, means, and uses of political science. It follows from the political-science-as-architecture analogy that political science is necessarily a heterogeneous and pluralistic discipline. It also follows that political scientists have a common purpose, which is to conceive of institutional structures that allow humans to live together in societies, just as the purpose of architecture is to conceive of physical structures in which humans can live together.


The political-science-as-architecture analogy
Architecture is the art and science of building houses and other physical structures that are meant to be used by humans. Academic and professional architects disagree among themselves on the precise meaning of the term architecture and often add a great deal of nuance to this simple, everyday definition. But a simple definition will do for now. 1 The inclusion of physical structures in general, and not only individual buildings, is important for this article since I will sometimes refer to metaphors of the state or political system as a building and at other times to metaphors of the state or political system as a city. This is standard usage: ever since ancient times, architects have designed both buildings and cities. Indeed, textbooks on architecture typically note that it's a mistake to distinguish between building houses and planning cities, for buildings are what they are because of their place in the city, and cities are made up of buildings.
There are many ways for politics and architecture to meet. Perhaps most obviously, political decisions are made, and power resides, in public spaces and in designated buildings such as parliaments, government ministries, and high courts. 2 That's why many of the words we use to analyze politics have spatial, architectural origins, including the terms "left" and "right," which go back to the seating arrangements in the French National Assembly in the early days of the Great Revolution. "Forum" is another example of a term with architectural origins. Political scientists use that word figuratively, but for the Romans, a forum was a combined marketplace and town square.
Another way for political science and architecture to meet is the study of urban politics. A large proportion of the world's population lives in cities, and all cities are shaped by political decisions about construction, infrastructure, and zoning. The political lives of cities are consequently an important concern for political scientists and architects alike. 3 The relationship between cities and their surroundings has equally important political consequences: recent work in political geography has demonstrated the powerful political effects, especially in the United States, of the flow of middle-class and upper-middle-class voters from cities to single-family homes in the suburbs in the second half of the twentieth century (Baum-Snow 2007;Nall 2018).
But this is not an article about how politics and architecture meet in practice; it's an article about political science as an intellectual endeavor. My argument is that the political-science-as-architecture analogy can help political scientists think more clearly about our own discipline. The main idea behind the analogy is that both political science and architecture are concerned with the organization of social space: how structures shape the way humans live together. 4 There are of course many differences between the two disciplines: my argument is one of analogy, not of identity. Unlike architects, political scientists study institutional structures, not tangible physical ones, and we typically address the problem of how humans might live together at a larger scale than architects do: that of entire countries, or even the whole world. But the similarities are striking enough to justify the comparison.
The similarities between political science and architecture become especially apparent when one considers modern theories of architecture that examine the effects of buildings and cities on social space. The theory developed by Hillier and Hanson in their book The Social Logic of Space (1984) and by Hillier in his Space is the Machine (1996) is one example. Hillier (1996, 4) argues directly against the view that architecture is "building plus art" and instead defends the view that buildings "constitute the social organisation of everyday life as the spatial configurations of space in which we live and move," and "represent social organisation as physical configurations." It is no coincidence that The Social Logic of Space ends with an overtly political analysis of the societal consequences of modern building and city planning (Hillier and Hanson 1984, 256-261).
There is a curious historical fact that supports the idea behind the political-science-as-architecture analogy. In Book 1, Part VIII of Politics, Aristotle (1908, 76 [c350 BC]) identifies a man called Hippodamus of Miletus as "the first person not a statesman who made inquiries about the best form of government." Aristotle wasn't particularly impressed with Hippodamus's theory of government and does not seem to have thought very highly of him as a person (he notes that many people found him "affected" since he would "wear flowing hair and expensive ornaments"). But he does give him credit for being first. Today, however, Hippodamus is best remembered for his day job: he was an architect, who became famous for the new town plan of Piraeus, the port of Athens, in the fifth century BC. 5 It can plausibly be argued, then, that the first political scientist, avant la lettre, was also an architect.
But there are many other reasons to find the political-science-as-architecture analogy apt. Most importantly, a building's purpose is to be useful to those who live, work, and convene within it, even if those people are very different from each other. Similarly, the purpose of a well-ordered political system is to make life easier and better for those who live in it, while recognizing that people differ from each other and have different goals and ideals in life.

Buildings and cities as metaphors
Buildings-or combinations of buildings and other structures, such as cities-have long been used as metaphors for the state or the political system. It is not the only such metaphor; there are numerous others. One is the Leviathan: the state as a synthetic person: an "Artificial Man" or a "Mortall God" (Hobbes 2009, Introduction andChapter XVII [1651]). Another is the ship of state, with the ruler as a captain who steers a vessel through treacherous waters (Plato 2000, Book VI [c380 BC]). But the metaphor of the political system as a building or a city seems more appropriate in a modern, pluralistic, changing society. The problem with the ship-of-state metaphor, in my view, is the notion that the state is going somewhere, as if on a pilgrimage or voyage of discovery. The problem with the state-as-person metaphor is that political society isn't a single, corporate organism: it's made up of numerous diverse people, or peoples.
The origins of the very word "politics" are architectural: the Greek politika derives from the word for city, polis, a root that is shared by words such as "policy," "polity," and "police." As noted by Deutsch (1970), the word "government," by contrast, derives from the Greek kubernan, which means "to steer or pilot a ship." In other words, whereas the word "politics" is associated with the concept of a city (and, grammatically, with a noun), the word "government" is associated with the ship-of-state metaphor and its very different historical and normative connotations (and, grammatically, with a verb).
As the etymology of the word "politics" suggests, there was a close connection between politics and place in the ancient world. Classical philosophers such as Plato and Aristotle, therefore, didn't treat buildings and cities as metaphors for the state or the political system: they were deeply interested in the construction of actual buildings and the planning of actual cities. Think, for example, of the discussions of city planning in Book V of Plato's Laws (2016 [c360 BC]) and Book 7, Part IX of Aristotle's Politics (1908 [c350 BC]).
In St. Augustine's The City of God (2004 [426]), however-written after the fall of the world's greatest city, Rome-we find a clear example of the city as metaphor. Throughout The City of God, Augustine juxtaposes two cities that represent contrasting ideas about the nature and purpose of human existence: the City of God and the earthly city. Augustine's ideas were influential throughout the medieval period. That was also an era in which political and religious beliefs were manifested in architectural practice, most famously in the construction of Europe's great cathedrals.
For an early Renaissance example of how a city was used as a metaphor for the political system, consider Ambrogio Lorenzetti's fourteenth-century frescoes "Good and Bad Government" in Siena's Palazzo Pubblico, where the Good City becomes an allegory of good government and the Bad City becomes an allegory of bad government (Skinner 2002, Ch. 3). When studying Lorenzetti's frescoes with these allegories in mind, one notices the ongoing construction work in the Good City, where buildings and structures are repaired and extended, and the decay and rot that corrupt the buildings in the Bad City.
Architectural metaphors for states and political systems were especially common in the eighteenth century. The reason is most likely that there's a natural affinity between architectural metaphors and the idea that political society should be governed by rules and laws that are deliberately constructed. Here, for example, is James Madison, arguing in Federalist 38 that it would be unreasonable for Americans to hold on to the older Articles of Confederation just because the new Constitution had some faults of its own: No man would refuse to quit a shattered and tottering habitation for a firm and commodious building, because the latter had not a porch to it, or because some of the rooms might be a little larger or smaller, or the ceilings a little higher or lower than his fancy would have planned them. (Hamilton et al. 2003, Federalist 38 [1787-1788 A few years earlier, Kant had used another building metaphor in his essay "Idea for a Universal History with a Cosmopolitan Purpose." According to Kant, the problem of how to design a "lawful civic constitution" is "the most difficult and the last to be solved by mankind," and he explained why in this famous sentence: This task is therefore the hardest of all; indeed, its complete solution is impossible, for from such crooked wood as man is made of, nothing perfectly straight can be built. (1991, sixth thesis [1784]) I will come back to Kant's essay in the next section. But let us first consider a few other examples of building metaphors in political thought. Two decades before Kant, Rousseau used one in The Social Contract (1920, Book II, Chapter 8 [1762) when describing the ideal legislator: As, before putting up a large building, the architect surveys and sounds the site to see if it will bear the weight, the wise legislator does not begin by laying down laws good in themselves, but by investigating the fitness of the people, for which they are destined, to receive them.
Toward the end of the eighteenth century, Burke used several building metaphors in Reflections on the Revolution in France (1790), notably in his defense of the English constitution (355): I would not exclude alteration either, but even when I changed, it should be to preserve. … In what I did, I should follow the example of our ancestors. I would make the reparation as nearly as possible in the style of the building. 6 In the next century, Karl Marx (1859Marx ( , preface [1977) also used a building metaphor to describe the political system, although he had something a little different in mind: The totality of these relations of production constitutes the economic structure of society, the real foundation, on which arises a legal and political superstructure and to which correspond definite forms of social consciousness.
And then there's this famous quote, from a speech Abraham Lincoln gave in Illinois in 1858 (Lincoln and Douglas 1860, 1): A house divided against itself, cannot stand. I believe this government cannot endure permanently half slave and half free. I do not expect the Union to be dissolved-I do not expect the house to fall-but I do expect it will cease to be divided. It will become all one thing or all the other.
There are twentieth-century examples too-think, for instance, of Foucault's discussion of the Panopticon as a metaphor for modern political rationality in Discipline and Punish (1975)-but let's end the historical exposé here and move on to the implications of the political-science-as-architecture analogy for political science as a discipline.

Architecture and the ends of political science
One of the main implications of the political-science-as-architecture analogy is that political science is and must be a discipline that's both normative and empirical. Political scientists study institutional structures, architects study physical structures, but both political scientists and architects want to understand how to build a world in which humans can live together. Solving that problem requires both normative ideas and empirical knowledge. On the one hand, as Gerring and Yesnowitz (2006, 101) note, "If political science is to matter to policymakers or citizens, as most political scientists believe it should, authors must be clear about how their subject ties into some broader telos that others might share." On the other hand, political science mustn't become idealistic, for the construction of institutions and political systems, like the construction of buildings and cities, is a practical matter. Normative thinking in political science needs to be tempered by the realism that comes with the empirical study of political systems.
Let us return to one of the classics of political thought that we discussed in the previous section: Kant's "Idea for a Universal History." In that essay, Kant wrote that the greatest problem mankind has ever faced and will ever face is "the achievement of a universal civic society which administers law among men." The driving force in human history, according to Kant, is mankind's "unsocial sociability": our "propensity to enter into society, bound together with a mutual opposition which constantly threatens to break up the society" (Kant 1991, fourth thesis [1784]). The purpose of political institutions is to help humans live together, even if they have competitive instincts that drive them apart. The solution is a "lawful civic constitution," and it's the problem of how to design such a constitution that is "the most difficult and the last to be solved by mankind." I don't think it's a coincidence that Kant used the building metaphor of "crooked wood" in his discussion of this problem. What is perhaps even more interesting about this section of "Idea for a Universal History," however, is Kant's next observation, which is that the design of a lawful constitution requires several different forms of knowledge.
That it is the last problem to be solved follows also from this: it requires that there be a correct conception of a possible constitution, great experience gained in many paths of life, and-far beyond these-a good will ready to accept such a constitution. (Kant 1991(Kant , sixth thesis [1784 In contemporary terms, the implication is that designing a lawful constitution requires an understanding of both normative ideas ("a correct conception," "good will") and empirical facts ("experience gained in many paths of life").
Architects have known since ancient times that constructing a building or planning a city involves many different types of considerations. Most famously, the Roman architect Vitruvius (1999 [c15BC]) argued two thousand years ago that a good building is characterized by three values: firmitas (strength), utilitas (utility), and venustas (beauty). That's also a pretty good summary of what we want from a constitution: we want it to be robust; we want it to be useful to those who live under it; but we also want it to reflect our values and our best aspirations-to be, in that sense, "beautiful." If we think of political science as a form of architecture, it follows that political scientists must answer several different types of questions to do their job well. The question of what makes a constitution "robust" can and should be studied with empirical methods, just as it is possible to study empirically what makes beams and walls robust. The question of what makes a constitution "useful" is empirical in part, but also calls for a normative investigation (useful in what sense and for whom?). The question of what makes a constitution "beautiful" is clearly normative.
But the political-science-as-architecture analogy doesn't merely suggest that both facts and values matter: it also suggests that questions facts and questions about values are intertwined. Vitruvius's point in On Architecture was that strength, utility, and beauty cannot be thought of in isolation. For example, a building's strength matters little if it isn't useful and beautiful and its beauty matters little if it isn't robust and useful. Similarly, we want more from a political system than mere resiliencerealism without ideals is cynicism-but a constitution that reflects high ideals but cannot be sustained and doesn't work in practice is a dangerous thing. As Mark Lilla (2001) argues in The Reckless Mind, there is something deeply irresponsible about forms of political thought that don't consider the realities of politics.
Political science is an alliance of those who study politics as it is, those who study politics as it should be, and those who do both. That alliance is sometimes seen as an uneasy one, but the political-science-as-architecture analogy suggests it's a defining quality of the discipline.

Architecture and the means of political science
The political-science-as-architecture analogy also suggests ways of thinking more clearly about the nature of political inquiry and the methods political scientists use and should use. One of the defining characteristics of architecture is that it combines art and science, with the term "art" being used in the sense of "skills acquired through experience" and not only in the sense of "producing aesthetic objects." That goes for political science too. Some of the questions political scientists ask are possible to answer with quantitative scientific methods. But there are other, equally important questions that aren't possible to answer with such methods; they call for the sorts of research methods that are commonly known as qualitative.
Incidentally, since political science is both art and science, it would perhaps be best to call it something else. My own preference is "politics." Again, the politicalscience-as-architecture analogy is helpful. The term "architecture" refers both to an academic discipline and to that which it studies (as in the sentence "this building is a good example of Baroque architecture"). So could "politics." To see how the political-science-as-architecture analogy works for methodology, consider the different types of problems architects need to solve when they design buildings. Some of those problems require quantitative evidence and experimental or statistical methods. Think, for instance, of questions about the strength of a building's construction materials, the flow of air through its ventilation systems, or the energy needed to power its heating apparatus. Knowledge of such applied-science problems is essential for any architect. Similarly, political scientists ask many questions that are best answered using quantitative evidence and experimental or statistical methods. Think of the effects of the electoral system, which is one of the main building elements in any democratic constitution. It is difficult to see how one could begin to understand those effects without learning from statistical analyses of the origins of party systems and electoral behavior under different electoral rules. When Kant wrote of the "great experience gained in many paths of life" that's required to build a political society from the crooked wood of humanity, scholars of politics had just begun to use statistics and statistical methods-the term "statistics" originated in eighteenth-century statecraft-but in today's social sciences, experiments and statistical methods are essential tools.
However, just as architects need other, less technical forms of knowledge to do their work well-designing a building is not only an engineering problem-so do political scientists. The questions that can be answered with experimental and statistical methods are but elements in the larger set of questions that scholars of politics need to answer. This is why the term "constitutional engineering" seems ill-chosen to me (Sartori 1997). Designing a constitution cannot be treated as an optimization problem (for a discussion, see especially Elster 1987). "Constitutional architecture" is a better term. Only a foolish architect would ignore the knowledge of engineers who can calculate the strength of frames and beams, and it would be equally foolish for a political scientist to ignore the "great experience" that has been gained through quantitative empirical research on politics (or indeed the "conceptions" of "possible constitutions" that can be derived from formal, mathematical models of institutions). But an architect who only paid attention to engineering problems, and who didn't make judgments about the needs of those who would use the building and the building's appearance and style, would be equally foolish.
The critical question, then, isn't whether both quantitative and qualitative methods belong in political science-they clearly do, and the political-science-as-architecture analogy explains why. The critical question is how quantitative and qualitative methods should be combined. The political-science-as-architecture analogy is helpful in at least two ways when thinking through this important question. First of all, the analogy suggests that fitting methods to research questions is a concrete, practical problem, not an abstract, theoretical one: it is ultimately a matter of what knowledge is required to inform ongoing discussions about the reform of political regimes, institutions, and systems. Second, the analogy accentuates the perils of excessive methodological specialization. Since the problems political scientists study cannot be solved using one single type of method, it is essential for a political scientist to understand and engage with scholarly work that relies on many different types of methods.

Architecture and the uses of political science
So far, we have discussed the implications of the political-science-as-architecture analogy for the questions political scientists ask (the "ends" of political science) and the methods we rely on when we answer those questions (the "means" of political science). This section discusses the broader implications of the political-science-asarchitecture analogy for the role of political scientists in society and how politicalscience knowledge is put to use.
If political science is society's architecture, the purpose of political science is to develop ideas about how to construct political institutions and systems, just as the purpose of architecture is to develop ideas about how to construct buildings and cities. But to whom are those ideas supposed to be directed, and whose interests should they serve?
Starting with the first question, in a democratic society, the construction of the political system must be a joint endeavor. This means that political scientists ought to direct their ideas to the public and not whisper in the prince's ear how he should rule the state. Architects don't work for themselves; they work for those who will eventually dwell and work in the buildings they design and the cities they plan. Similarly, the job of political scientists is not to impose their ideas on others, but to contribute to a public debate, among their fellow citizens, about institutions, government, and politics. In their contribution to the recent book Political Theory and Architecture (Bell and Zacka 2020), Lindstrom and Malpas (2020) advocate a practical, modest, and inclusive architecture that is responsive to those who live and work in the affected buildings, neighborhoods, and cities. That's also how political scientists ought to think about their role in society.
This brings us to the second question: whose interests political scientists ought to serve. Again, the political-science-as-architecture analogy suggests a way of thinking about the problem, for the history of architecture is replete with examples of how architectural designs, construction projects, and urban renewal have been used to serve the interests of the powerful-whether they are authoritarian governments in non-democratic states or powerful individuals and interest groups in democracies.
One doesn't have to turn to extreme cases such as Adolf Hitler's architect, Albert Speer, to find examples of how architecture has served the powerful. Consider the role played by the Prefect of the Department of Seine, Georges-Eugène Haussmann, in Second-Empire Paris in the 1850s and 1860s. Haussmann was commissioned by France's authoritarian ruler, Napoleon III, to put in place a program for urban renewal in the capital. It was a program that changed the layout of Paris forever. Some of Haussmann's motives were progressive-including improvements in public health and sanitation and more efficient urban transport-but the program was also designed to facilitate political repression. In the first half of the nineteenth century, the narrow streets and alleyways of old Paris had been used effectively by revolutionaries and rebels, but Haussmann's new, wide boulevards allowed the military to deploy large numbers of troops throughout Paris, using its manpower and technology to suppress urban unrest and defeat rebellions (on the defeat of the Paris Commune in the spring of 1871, see Merriman 2014).
As this example shows, architecture is not innocent-indeed, as discussed by Aslam (2020ab), architecture, government, and power are inextricably linked. And architects sometimes serve the interests of the rich and powerful against those of the poor and powerless. This is a temptation that political scientists can also face. Consider, for example, the practice of partisan gerrymandering in systems with firstpast-the-post elections: the redrawing of electoral district boundaries to increase the likelihood that the incumbent party wins. A political scientist who offers advice on how to keep the opposition out of power is changing the institutional environment to protect the powerful, much like Haussmann protected the powerful by changing the physical environment of nineteenth-century Paris.
There is another sense in which architecture is not innocent. Even if architecture isn't used to control, defeat, and segregate-even if the motives aren't sinisterthere are almost always individuals and groups who are harmed by great building projects and urban-renewal projects. In his biography of New York City's powerful city planner Robert Moses, who incidentally had a doctorate in political science from Columbia (Moses 1914), Robert Caro (1974 documents the many ways in which urban renewal in New York City served the interests of the powerful and was used to silence, relocate, and keep out the powerless. He also demonstrates that the big development projects led to great suffering as thousands and thousands of families were forced to leave their homes. Big political reforms have similarly disruptive consequences, and those consequences need to be accounted for when we analyze institutional change and policy change.

Conclusion
When I was younger, I did not worry much about the purpose of political science: it just seemed like a scholarly field among others where one could do interesting work on important topics. As I grow older, I find myself thinking more about what political science is and why it is worthwhile. In this article, I have tried to answer those two questions through analogical reasoning, by comparing political science to architecture.
As I noted at the outset, the political-science-as-architecture analogy isn't meant to exclude other ideas of what political science is for. On the contrary, I hope this article will encourage others to propose alternative answers. I have argued that the political-science-as-architecture analogy is apt and has important implications for the ends, means, and uses of political science, but there are no doubt other analogies that have other thought-provoking implications.
I would also like to note that the political-science-as-architecture analogy doesn't suggest the domain of political science should be restricted to a narrow set of questions about institutions. Architects have always put the built environment in context by examining its psychological effects on individuals and groups, analyzing its role in economic and social life, and studying the history of architecture and architectural styles. Similarly, political scientists have every reason to put political institutions and political systems in their social-psychological, economic, sociological, and historical contexts, and in other contexts besides. This article's discussion of the political-science-as-architecture analogy is related to an ongoing discussion about the ends, means, and uses of economics, which is a scholarly discipline that is closely related to ours. Interestingly, there are several examples from the last couple of decades of leading economists who have compared their own discipline with other disciplines. I am thinking of Esther Duflo's recent argument that economists are best thought of as plumbers (2017), which is itself a response to Alvin Roth's argument that the role of the economist is comparable to that of an engineer (2002; see also Mankiw 2006 on "the macroeconomist as scientist and engineer"). For Roth, the key point of the economics-as-engineering analogy is that the task of designing markets comes with a "responsibility for detail" (2002,1342), which means that economists need to go beyond microeconomic theory and use computational and experimental methods. In Duflo's view, modern economics is an even more hands-on, practical discipline than the engineering analogy suggests: The economist-plumber stands on the shoulder of scientists and engineers but does not have the safety net of a bounded set of assumptions. She is more concerned about 'how' to do things than about 'what' to do. In the pursuit of good implementation of public policy, she is willing to tinker. Field experimentation is her tool of choice. (Duflo 2017, 3) Both the economist-as-engineer analogy and the economist-as-plumber analogy start with the subject matter and purpose of economics as a discipline and then draw lessons about how research questions are asked, how methods are developed, and how economic knowledge is used. That's what this article has sought to do for political science.
It is striking that Roth and Duflo compare economics to two intellectual and practical endeavors that are themselves closely related to architecture, the intellectual and practical pursuit for which political science is compared in this essay. If these analogies are apt-as I think they are-they suggest a division of labor between political science and its sister discipline, economics, that fits well with Fritz Scharpf's discussion of the relationship between political science and economics in his Games Real Actors Play (1997). For Scharpf, the distinction between political science and economics is that economics studies the effects of policies (how policies result in intended and non-intended outcomes), whereas political science studies the context in which policies come about in the first place (how policies are made).
Is the architecture analogy uniquely apt for political science, or is it also relevant across the social sciences? We have already seen that economists think of their discipline as comparable to other pursuits-such as engineering and plumbing-but what about other disciplines, such as sociology and anthropology? My own view is that the architecture analogy is most directly relevant to political science. All the social sciences study how humans live together in societies, of course, but political scientists are uniquely interested in how social life is shaped by institutions that we humans design and build ourselves: public policies, constitutions, laws, administrative organizations, and international regimes.
The idea of political science as a form of societal architecture is ambitious in one respect and unambitious in another. It is ambitious since it asks political scientists that we integrate art and science and combine insights and ideas from different intellectual traditions into a coherent vision of how political systems and institutions can and ought to be organized. But in another important sense, it is an unambitious idea, for good architects prioritize the needs of those who would dwell, work, or convene in the buildings and cities they design over their own ideas and goals. The job of an architect is not to realize some grand plan for how other people should live their lives; it is to help people ensure that the structures in which they live their lives are robust, useful, and beautiful.
Loosely translated into English, this reads, An open city, not a fortified one, we build together. Its lights flare up toward the desolate heavens.
Funding Open access funding provided by University of Gothenburg.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.