Constitution-Making and Violence Academic research paper on "Law"

CONSTITUTION-MAKING AND VIOLENCE

Jon Elster1

ABSTRACT

Contrary to a traditional view, constitutions are rarely written in calm and reflective °

moments. Rather, because they tend to be written in period of social unrest, constituent n

moments induce strong emotions and, frequently, violence. The paper examines two a

such cases: the Federal Convention of 1787 and the French Assemblée Constituante of e

1789-1791. These involved state violence as well as popular violence. In the USA, the o

unequal political representation of the backcountry explains both the violent events 3

leading to the Convention and its outcome. In France, the dismissal of the King's Minister p

Necker explains the subsequent urban and rural violence, and ultimately the abolition of l

feudalism and the fall of the monarchy. 0

L'anarchie est un passage effrayant, mais nécessaire, et c'est le seul moment où l'on § peut arriver à un nouvel ordre des choses. Ce n'est pas dans des temps de calme qu'on

prendrait des mesures uniformes. ("Anarchy is a frightening but necessary passage, o

and the only moment when one can establish a new order of things. It is not in a

calm times that one can adopt uniform measures".) (Comte de Clermont-Tonnerre, N

AR 9, 461) 3

1. INTRODUCTION S

This is an essay in macro-historical sociology. I am not an historian, but I read historians and some of their primary sources. On the basis of my readings about the American and French constitution-making processes in the late eighteenth century, I shall try to distill some ideas that may have general application. Specifically, I shall consider the role of violence in the making of the two constitutions: actual violence, threats of violence, warnings of violence, fear of violence, and even hope of violence. In a broader perspective, we should also include acts of resistance or disobedience to authorities.

1 Robert K. Merton Professor of the Social Sciences and Department of Political Science, Columbia University, New York, NY, USA. Email: je70@columbia.edu. I am grateful to Jack Rakove and Keith Dougherty for comments on two earlier drafts and to Henrik Elster for technical assistance.

© The Author 2012. Published by Oxford University Press on behalf of The John M. Olin Center for Law, Economics and Business at Harvard Law School.

This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited. doi:10.1093/jla/las009

I shall also distinguish between visceral or emotional fear of violence and prudential or rational fear. Whereas the former is a genuine emotion, caused by the belief in an imminent danger to the agent, the latter does not amount to more than a simple belief-desire complex (Gordon 1987, 77 and passim). As an example, "I fear that it will rain" means "I believe it will rain and I do not want it to rain". When de Montaigne (1991, 83) wrote that "it is fear I am most afraid of' and FDR said that "the only thing we have to fear is fear itself', they were referring to prudential fear of visceral fear (thanks to Ken Shepsle for this observation).

We should not be surprised that constitution-making goes together with violence. According to a cliche to which I have unfortunately contributed (Elster 1984, ch. II.7), constitutions are typically written in a calm and reflective moment that enables sober and public-spirited framers to design institutions that will prevent the interests and passions of future actors from acting against the general interest. The reality is different. "No liberal demo- I

cratic state has accomplished comprehensive constitutional change outside the a

context of some cataclysmic situation such as revolution, world war, the with- f

drawal of empire, civil war, or the threat of imminent breakup" (Russell 1993, o

106). In these settings, strong passions are inevitable and violence is likely. g

The American and French cases certainly confirm this expectation. Before I O

discuss them, I shall mention a few other cases. The work of the Frankfurt c/

constituent assembly of 1848 was "threatened by the hunt of the crowd for e

unpopular members of the assembly" (Eyck 1968, 312). One member of the Y

Right Center was beaten up, and two were killed. If we compare the two 3

drafts made by the Committee of Constitution of the French constituent i

assembly of 1848, before and after the June insurrection of the Parisian work- S

ers, the second was considerably less radical than the first, by abolishing the O

right to work and substituting proportional for progressive taxation. The >

constitution of the Fifth French Republic was adopted when the parliamen- 3;

tarians of the Fourth Republic granted full powers to the Gaulle under the 2

pressure of events in Algeria. In his inimitable telescoping, "In 1958 I had a 5

problem of conscience. I could just let things take their course: the paratroopers in Paris, the parliamentarians in the Seine, the general strike, the government of the Americans: it was written on the wall. Finally a moment would have arrived when everybody would have come looking for de Gaulle, but at what price? Thus I decided to intervene in time to prevent the drama" (Peyrefitte 1995, 262). It makes sense to assume that some parliamentarians feared for their lives and that their visceral fear affected their decision to abdicate from power.

Among the many differences between the two eighteenth-century processes, some can be traced back to the fact that they represent different stages in the

frequently occurring revolutionary pattern of "two steps forward, one step backward", first observed in seventeenth-century England. In that country, the abolition of the monarchy in 1649 constituted the two steps forward, and the creation of a constitutional monarchy in 1660 and more definitively in 1688 marked one step backward. In 1787, America took one step backward. The process may be seen—and was seen by some contemporaries—as a "counter-revolution against popular democratic ideals" (Bouton 2007, 4). In 1789, France took two steps forward, leaving it to Napoleon and more definitively to Louis XVIII to take the one step backward. This contrast between the counterrevolutionary thrust in America and the revolutionary impulse in o France provides one key to the differences between the two processes. § Needless to say, the vast differences in social and economic structure also d

had repercussions at the political level. America was largely a country of free- r

holders, France a country of peasants under a feudal regime. 3

Popular violence was a driving force in the calling of a constituent assembly : (America) or in the shaping of the document it produced (both countries). a In America, the violence was predominantly rural; in France it was both rural f and urban. In both countries, it was triggered by a potent combination of O objective hardships and subjective beliefs about the malevolent intentions that produced those hardships. In both countries, beliefs about speculators were important, as a direct causal force in America and as a background factor in France.

State violence also shaped the constitution-making in crucial ways. In the USA, the defeat of Shays' Rebellion by the army raised by the governor of Massachusetts was a close thing. Only lack of coordination prevented the rebels from seizing the federal arsenal at Springfield. Had they succeeded, they might have marched on Boston (Richards 2002, 29-30). Some clauses in the 1787 Constitution can be traced back to the desire of the framers to have a more robust repressive machinery at their disposal. In France, state violence was seen as an option on two occasions. In July 1789, the King's failed attempt to repress the National Assembly contributed immeasurably to its radicalization. In August, the Assembly briefly considered repressing the peasant insurrections in the provinces before choosing instead the path of concession.

Emotions were both causes and effects of popular and state violence. Anger, resentment, and fear caused violence; violence caused fear. A debatable question is whether the American framers were subject to visceral fear (as the French certainly were) or to prudential fear only. I certainly cannot claim to resolve this question, but my inclination is to believe that an element of visceral fear was present and that, as Jefferson suggested, it caused the Founders to overreact to Shays' rebellion.

2. AMERICA: FARMERS AND FRAMERS

Since a major line in my argument will address state politics in the 1780s, I should first state the inevitable limitations of my analysis. In a short treatment, even a summary discussion of events in all thirteen states is impossible. I shall focus on Massachusetts and Pennsylvania, with occasional remarks on other states.

My account of the run-up to the Federal Convention is very much influenced by the neo-Progressive analyses of Bouton (1996, 2007) and Holton (2007). Although these are mainly works of social history, they also deal with the eco- o

nomic causes of the class struggles and their political continuations. On the l

economic and especially the financial side, the classical study by Ferguson e

(1961) is invaluable, while Brown (1993) provides a useful state-by-state break- o

down. On the political side, I draw on the equally classical work by Pole (1966) h

on representation in the American colonies and states. /

I shall organize my discussion as follows. I first summarize some aspects of .

the conflicts between the backcountry and the seaboard in the colonies and at o the Convention. Second, I summarize the relevant forms of paper money and debt certificates issued by the American Confederation. Third, I describe the

class struggles arising from the redemption of these instruments and the pol- o

itical responses from the state legislatures. Finally, I link these issues to the t

calling of the Convention and to the document it proposed for ratification. e

2.1 The (Doubly) Neglected Backcountry n

In 1775, probably around 25 percent of Americans lived in what was called "the sS

backcountry" (Jensen 1968, 9). In Pennsylvania and the Carolinas, people living O

in these areas amounted to 40 or 50 percent of the population (ibid.). The western counties were neglected at the time, in the sense that they often faced larger burdens and had weaker rights than those closer to the seaboard. This difference is also neglected—often mentioned but rarely highlighted—in scholarly discussions of the period. Whereas economic qualifications for voting and eligibility are frequently cited as sources of bias in representation, the geographical bias in apportionment has received less attention. With respect to the back-country, this distinction matters. "The property qualification for voting meant little in the backcountry, where land was easy to acquire. Those areas could gain political influence only if they could gain representation in the legislatures, and this most colonial legislatures were unwilling to give them" (id. 26).

In the colonial period, "legislators saw that granting equal representation to newly created communities meant diminishing their own power. To preserve their superiority, some legislatures withheld representation completely or

assigned new regions fewer representatives than the older areas had" (Zagarri 1987, 43). The underrepresentation continued well beyond the colonial period. By 1787, Maryland, both Carolinas and Virginia used regional representation to both houses that, in the last three of these states, entailed very poor representation of the backcountry (Gazell 1970). In 1785, South Carolina blatantly ignored its own constitution when it failed to reapportion the assembly (Zagarri 1987, 48). Georgia, Connecticut, and New Jersey had proportional representation for one house and regional representation for the other, the remaining six states using proportional representation for both houses (Gazell 1970).

Yet even when the state constitutions did not stipulate unequal influence of eastern and western counties, geography and money often conspired to produce the same effect. The case of Massachusetts is emblematic in this regard. Although several members of the Convention that drafted the 1780 Constitution demanded that the state assume all costs of the delegates,

[the] Constitution as drafted and adopted provided that each town x

should pay the expenses of its own representatives incurred in d

attending the session. [...] The point about payment was of very great §

consequence and of greater practical signification, in all probability, |

than the question of the precise basis of representation. Interior towns, g

especially those at more than a day's journey from Boston, very 3

frequently failed to send a member at all. The cost of maintaining a §

representative in the capital through the legislative session as a heavy o

burden to which the frugal farmers saw little reason to subject U

themselves; were it necessary to be represented, in order to put the e

town's view in some dispute, a single member would be cheaper than y

two. For the seaboard towns the capital was relatively accessible. Their g

greater wealth also made it easier for them to maintain representatives. §§

All the normal circumstances of economic and political life therefore 0

tended to give the advantages to the east coast (Pole 1966, 204; emphasis 0

added). 5

In addition, the low quorum—60 out of 228 representatives—ensured that attendance from country districts was not required. "Some fifteen town meetings had the foresight to raise objections against permitting so small a number when attendance from country districts was likely to be dangerously thin [... ]. These interior towns hoped to ensure that some of their members were always present, to counter the danger that a small minority from the eastern towns might push through social interest legislation" (id. 199-200). Some of the measures to be discussed below owed their origin to the low quorum and the location of the assembly: "During Shays' Rebellion a Hampshire County writer

charged that a recent tax relief proposal had been 'spun out and put off till July just at the close of the session, when many of the country members [were] under a necessity of returning to their farms'" (Holton 2007, 169). Below I discuss other, emotionally based reasons for the low attendance of the backcountry.

In Virginia, according to Jefferson (1784, Query XIII), "nineteen thousand men, living below the falls of the rivers, possess half the senate, and want four members only of possessing a majority of the house of delegates; a want more than supplied by the vicinity of their situation to the seat of government, and of course the greater degree of convenience and punctuality with which their o

members may and will attend in the legislature. These nineteen thousand, l

therefore, living in one part of the country, give law to upwards of thirty thou- |

sand, living in another, and appoint all their chief officers executive and judi- £

ciary". Jefferson's "analysis grew constantly more correct as time went on and 3

population moved into the western sections" (Pole 1966, 297).

In Pennsylvania, the 1776 constitution substituted proportional for regional a

representation and thus did away with the underrepresentation of the western f

counties. These areas were, however, disadvantaged in other respects. "Farmers 0

struggling to get from under a load of unpaid debts and taxes could not afford g

to leave the plow to spend months in Philadelphia debating laws" (Bouton 0

2007, 129). To obtain legal title to land on the frontier, poor settlers had to a

make the expensive trip to Philadelphia (id. 121-122). Without the title, they e could not vote.

The state delegations to the Federal Convention were elected by the legisla- 3

tures, and reflected the geographical biases of the latter. The average distance of i

the county of a framer to navigable water was 16 miles, the maximal distance S

being 200 (McGuire 2003, 69). The distance is highly correlated with the votes О

on prohibiting the issuance of paper money by the states—framers from the >

most isolated areas being highly likely to vote against a ban (id. 73). According 3

to McDonald (1982, 37), delegates at the Convention represented thirty-nine 2

out of fifty-five major geographical areas in the nation. Unrepresented were the 5

six major areas in the mountain and transmontane regions of Pennsylvania, Virginia, and South Carolina, as well as the Berkshire areas in Massachusetts and Connecticut.

At the Convention itself, there were many references to the western lands and to future western states. The always outspoken Gouverneur Morris claimed that the experience from Pennsylvania showed that it would be dangerous to let future western states accede to the Union on equal terms with the original ones: "they would not be able to furnish men equally enlightened, to share in the administration of our common interests. The Busy haunts of men not the remote wilderness, was the proper school of political Talents. If the Western

people get the power into their hands they will ruin the Atlantic interests. The Back members are always most averse to the best measures. He mentioned the case of [Pennsylvania] formerly. The lower part of the State had ye. power in the first instance" (Farrand 1966, vol. I, 583). John Rutledge also expressed skepticism towards equal representation of the Western states (id. 534). Nathaniel Gorham, claiming to speak for the committee on representation that he had chaired, asserted that "the Atlantic States having ye. Govt. in their own hands, may take care of their own interest, by dealing out the right of Representation in safe proportions to the Western States" (id. 560). Although Madison and George Mason spoke out in favor of equal, impartial representation of the future Western states (id. 584, 579), we cannot rule out that their arguments, or at least Mason's, were motivated by the belief that slavery would expand in these states (Amar 2005, 90).

It is likely, or at least highly possible, that the Constitution would have been rejected if all districts had been equally represented in the state ratifying / conventions. In South Carolina, "coastal areas [...] overwhelmingly favored .

the Constitution. Up-country areas just as overwhelmingly opposed it. The less C

populated coastal areas, however, had 151 delegates to the up-country's jc

eighty-six" (Roll 1969, 30). In the chain reaction of ratifications, the "psycho- 3

logical effect of the South Carolina ratification on the key state of Virginia o

[which ratified in a close vote of 89 to 79] was all the more important t

because it eliminated the possibility of an attractive alternative [a southern e

confederacy]" (id. 32). In the New York State convention, the K

pro-Federalism majority that emerged after the ratification by Virginia repre- 3

sented a minority of the population (id, 32-33). Overall, the agrarian oppos- 5

ition to the Constitution "tended to lose out in their pursuit of t

'widely-dispersed, strictly-limited powers, located close to the people', largely o

because of unfairly diminished representation at the ratifying conventions" p

(id. 40, citing Benson 1960, 219). 3

2.2 Paper Money and Debts

The "most distinctively Madisonian provisions" (Amar 2005, 123) were the prohibitions in Art I.10: "No state shall [...] coin money; emit bills of credit; make any things but gold and silver a tender in payment of debts; pass any law [. ..] impairing the obligation of contracts". James Wilson and Benjamin Rush both said that if the constitution consisted only of a ban on paper money, it would be worth adopting it (cited in Bouton 2007, 179, 301). Although the Convention at one point voted to strike out a clause authorizing the creation of federal paper money (Farrand 1966, vol. II, 310), the practice is not explicitly banned in the Constitution.

In peacetime, before 1776, many colonies had routinely and successfully used paper money, careful to secure collateral in land or in future taxes (Ferguson 1961, ch. 1; Grubb 2006). During the Revolutionary War, paper money and other instruments issued to fund the war effort depreciated very rapidly. In theory, the states could have levied heavy taxes and, by withdrawing money for tax payment, preserved its value by reducing the amount in circulation. This option was, however, politically impossible: "Having so recently opposed taxation by Parliament, the American people were sensitive on the subject" (Ferguson 1961, 30). Instead, the Continental Congress first let Continental bills depreciate to a few percent of their face value and then turned to other o

instruments, notably federal bonds, certificates issued by the states or the federal §

government as payment for goods, and military certificates issued by the states d

as payment to soldiers. The states, too, issued paper money and bonds. Some of r

these instruments carried interest, others did not. Except for the federal bonds, 3

which were mainly used as an investment, they circulated as a medium of :

exchange and were often used to pay state taxes. Speculators also bought up a

large quantities of certificates at bargain-basement rates from the original f

recipients. O

Once the end of the war was in sight, many holders of these instruments— g

Continental dollars, bonds, and certificates—demanded redemption in specie O

at face value and, when appropriate, with interest. Because the Continental a

Congress could not get enough states to agree on a federal impost (a tariff on e

imports) to fund the redemption, the states had to decide how to meet the Y

demands. Most of them enacted legislation calling for heavy taxes in specie. 3

Although some of these were to come from imposts and excise taxes (consump- i

tion taxes), the bulk of taxation was usually in the form of poll and property S

taxes. As gold and silver were scarce and prices (in specie) therefore subject to O

heavy deflation, farmers found it difficult or impossible to sell their products at >

prices that would allow them to pay their taxes. (On the other hand, deflation 3

brought a windfall gain to state officials such as the governors, who lived on 2

fixed salaries.) These decisions by state assemblies triggered strong reactions in 5

several states. "Attempting to force the common people to pay hard money, the governments of Pennsylvania, South Carolina, Rhode Island, and Massachusetts threatened property sales on a wide scale. Resistance, then retreat, invariably followed [...] Maryland, Virginia, New Hampshire, New Jersey and, and possibly Delaware [also] fit the pressure-resistance retreat model" (Brown 1993, 122). The remaining four states did not attempt to collect taxes in specie (id.).

Because Massachusetts saw the most violent form of resistance, I shall focus on that state. The conservative recovery of power in 1780 led to "an arbitrary state program for consolidating and paying war debts" that was "the most

expensive possible under the circumstances, for the wartime currency was given preferred status" (Ferguson 1961, 245). Between 1780 and 1786, the state legislature enacted nine direct taxes variably payable in Continental currency, specie, Bank of America notes, notes issued by Robert Morris, army certificates, federal indents (certificates of interest on federal bonds) and in beef (Brown 1993, 247). In 1781 and in 1786, the legislature levied heavy taxes in specie. The amount of specie required by the 1786 tax underestimates the actual amount needed, since one-third of the tax was to be paid in indents, which most people had to buy from bondholders with hard money (Holton 2007, 66).

The efficiency of tax collection diminished with the distance of the counties from the coast. In Middlesex, the arrears of taxes for 1780-1782 was 33 percent of the levies, in Berkshire it was 74 percent (Brown 1993, 101). The efficiency also diminished with time. In the state as a whole, arrears increased from 12 percent to 84 percent (id. 102) in the period 1782-1786. At the same time, the scarcity of specie often made it difficult to repay private creditors. / "Many farmers had gone into debt before the war when money was abundant .

and prices were high; now, as money became scarce, prices dropped and farmers C could not sell their crops and livestock for enough to cover their debt" (Bouton jc 2007, 23). Even when farmers could bear one of these charges—taxes and debt 3 repayments—the combined burden was often beyond their means. r

2.3 Resistance, Repression, and Retreat

Although taxes and debts might be equally burdensome in an objective sense, they appeared quite different from a subjective point of view. The relation n between debtor and creditor was not intrinsically hostile. The claims "that procurrency farmers were simply seeking to defraud their private creditors" is a myth (Holton 2007, 61). Farmers and artisans knew quite well that they would need credit later, which would not be forthcoming if they did not service their current loans. In contrast, the relation between taxpayers and bondholders was deeply antagonistic. If bondholders had been mostly war veterans still in possession of their original bonds, taxpayers would have seen the strength of their moral claim. They might still have resisted the demand for tax payment in hard money, but less virulently. This was far from being the case however. "Although a host of farmers and soldiers had held on to their bonds, the majority had not, and by value most of the debt had concentrated in the hands of a few" (Bouton 2007, 37). In Massachusetts, "nearly 80 percent of the state debt made its way into the hands of speculators who lived in or near Boston, and nearly 40 percent into the hands of just thirty-five men" (Richards 2002, 75).

On the one hand, "Americans were acutely aware that most of the tribute that public officials exacted from them went to bondholders" (Bouton 2007,

32) and "resented the sacrifices their assemblymen exacted on behalf of bond speculators" (id. 38). They did not "accept the legislative argument that the chief beneficiaries were 'worthy patriots' who had come to the aid of the state in its time of need" (Richards 2002, 79). Understandably, veterans who had sold their bonds to speculators out of necessity were "especially angry" (id.).

On the other hand, "Bondholders were acutely aware that the value of their investments hinged on the willingness of the state legislatures to impose taxes" (Bouton 2007, 40). As a consequence, "Many bondholders, recognizing that political events determined the value of their investments, made efforts to influence politics" (id. 41). Of "the thirty-five men who held over 40 percent of the state debt, all of them during the 1780s either served in the state house themselves or had a close relative in the state house" (Richards 2002, 78). Although we cannot determine the actual influence of the speculators on the legislators, "in the eyes of their countrymen [their] influence was enormous" (Bouton 2007, 41). In Philadelphia, Pelatiah Webster (1785, 303) proposed that in decisions concerning the public debt, assembly members who were "directly or indirectly possessed, interested, or concerned, otherwise than as an original holder, in any public securities" should not be allowed to vote, any more than "a judge or juryman should sit in judgment in a cause, in the event of which he is personally interested" (id. 302).

To address the issues arising from taxes and debts, the farmers had the choice between political action and private resistance. In April 1786, voters in Rhode Island elected an assembly of "Reliefers" that issued paper money and made it legal tender for all debts, private and public. In Massachusetts, the session that began in May 1786 could, in principle, have enacted pro-relief measures, had not the western counties been underrepresented compared with the eastern ones. The farmers were caught in a classical collective action dilemma. "The very factors which made representation urgent also made it more burdensome. The harder the times, the more inducement to the towns to cut their costs. That any one town's one or two representatives would be able to make an effective impression on the general policies of the [State assembly] or on the condition of the [State] always seemed improbable" (Pole 1966, 234-235). In addition to this rational calculation, the abstention from the assembly also had an emotional root. "Even though numerous New Hampshire and Massachusetts towns defeated their antirelief assemblymen in the in the spring 1786 elections, their action was cancelled out by other towns that expressed their anger at the legislature's harsh fiscal and monetary policies by withdrawing their representatives altogether" (Holton 2007, 134).

Later, the farmers of Massachusetts got their way in the assembly. They did so because they had undertaken widespread private resistance, which escalated into collective violence with Shays' rebellion. Taken as a whole, the states

demonstrated a remarkable variety of forms of resistance. Like their near-contemporaries in France, farmers in Virginia used arson to destroy property records that were needed to execute the law (Holton 2007, 146). In New Jersey, "taxpayers formed groups that purchased the office of excise collector - all with the express purpose of not making anyone pay" (id. 147). In a remarkable analysis, Bouton (2007, 146) explains that in Pennsylvania,

[Ordinary folk] built a series of concentric rings of protection [...] around their communities. [...] Working from the outermost rings to the inner ones, the first was formed by county revenue officials who tried to thwart tax collection. The second ring was composed of county justices of the peace who refused to prosecute delinquent taxpayers and tax collectors. The third ring was formed by juries who acquitted those accused of not paying their taxes. The fourth ring was composed of sheriffs and constables who would not arrest non-paying citizens. The fifth ring involved ordinary folk attempting to stop tax collection and property foreclosures through nonviolent protest. Ring six was people trying to achieve those same goals through violent crowd action. Ring seven as composed of self-directed country militias that refused to follow orders to stop any of this protest.

In Massachusetts, Shays' rebellion produced three effects. First, the governor raised an army from private sources to defeat the rebels. Of the 153 contributors, more than half were speculators (Richards 2002, 78). Although the Continental Congress had requested $530 000 from the states to suppress the rebellion, only Virginia complied. In the eyes of many Nationalists, the failure n to take collective action on this occasion provided the clinching proof of the weakness of the Confederation and of the need for a stronger central government (Dougherty 2001, 128).

Second, the legislature in Massachusetts caved in. In the fall of 1786, "the same legislators who had adopted the punishing taxes that provoked the rebellion - and who had voted down several relief measures during the spring and summer - now granted farmers a broad rang of tax and debt reliefs" (Holton 2007, 155).

Third, in the 1787 elections town meetings replaced two-thirds of the assemblymen and defeated the governor. "What made the April-May 1787 elections different from those held a year earlier was that by that time, anger at the government was such a common topic of conversation that farmers became convinced that like-minded individuals throughout the state were going to show up at their town meetings determined to replace the assembly majority" (id. 156). On this analysis, the collective action problem of the farmers was not (as assumed by Pole 1966, 234-235) a Prisoner's Dilemma, but rather an

Assurance Game. Lack of information rather than lack of solidarity was the main cause of noncooperation in the 1786 elections (see also Bouton 2007, 130 for similar observations on Pennsylvania).

Faced with actual or anticipated violence and resistance, other state legislatures that were not replaced (as they were in Rhode Island and Pennsylvania) also caved in. In South Carolina, the government backed down in 1785 when it seemed "likely that a resort to coercion by low-country leaders would have forfeited their up-country and back-country support, and with it their precarious control of the assembly" (Brown 1993, 80). In Maryland, a "divided government retreated from the brink and eased the pressure" (id. 128). In Virginia, Madison wrote to Jefferson on December 4, 1786, "the specie part of the tax under collection is made payable in [tobacco]. This indulgence to the people as it is called & considered was so warmly wished for out of doors, and so strenuously pressed within that it could not be rejected without danger of exciting some worse project of a popular cast".

2.4 The Convention and the Constitution

The events in Massachusetts and in the other states had almost certainly an important impact both on the calling of the Federal Convention and on the substance of the Constitution. Schematically,

The call for a Convention to revise the constitution issued by the five states that had sent delegates to the meeting at Annapolis in September 1786 might not have gotten very far but for Shays' rebellion, which encouraged merchants, politicians, and generals who desired a stronger national government. The merchant Stephen Higginson who had for many years wished for a crisis that might trigger the establishment of a stronger union, wrote to his friend Henry Knox (Secretary of War) that the rebellion "must be used as a stock upon which the best fruits are to be ingrafted" (Brynner 1993, ch. 5). Knox, in his turn wrote alarmist, exaggerated, and persuasive letters to his friend George Washington (id. ch. 6). Logically enough, some Nationalists hoped that the violence would not be crushed too quickly. "As long as the insurrection was eventually crushed

[General John] Brooks was happy to see it continue. He saw nothing but good coming from it. He even hoped that that the rebels would become more audacious. 'Should the insurgents begin to plunder,' wrote Brooks, 'I think it will have a good effect.' It would provide good propaganda for the cause of a stronger national government" (Richards 2002, 128).

Although there is only indirect evidence that Shays's Rebellion had a decisive impact on the decision of Congress to propose the Federal Convention and on the decision of all the states expect Rhode Island to send delegates, the evidence that Washington would not have attended but for that event seems compelling. It is also plausible—although hard to prove—that the Convention would not o

have managed to keep the vital secrecy of its proceedings if Washington had not l

been President (Rossiter 1987, 167-168), and that the document would not e

have been ratified by the states if he had not lent his prestige to it. o

These issues are secondary, however, to the impact of the Rebellion on the h

debates in the Convention and on the final document. Madison's notes from /

the Convention contain dozens of direct and indirect references to the events in .

Massachusetts, often coupled with denunciations of "the turbulence" of 0 democracy (e.g. Farrand 1966, vol. I, 51 (Randolph), 299 (Hamilton), 430 (Madison)). Here is a representative statement by Elbridge Gerry, a delegate from Massachusetts of "gerrymandering" fame:

The evils we experience flow from the excess of democracy. The people do not want [lack] virtue, but are the dupes of pretended patriots. In Massts. it had been fully confirmed by experience that they are daily misled into the most baneful measures and opinions by the false n

reports circulated by designing men, and which no one on the spot can r

refute. One principal evil arises from the want of due provision for y

those employed in the administration of Governnt. It would seem to §

be a maxim of democracy to starve the public servants. He mentioned i

the popular clamour in Massts. for the reduction of salaries and the ,

attack made on that of the Govr. though secured by the spirit of 1

the Constitution itself. He had he said been too republican heretofore: he was still however republican, but had been taught by experience the danger of the levilling spirit. (Gerry in Farrand, Records I, 48).

It is worth mentioning that the attacks on the salary of the governor was due to the fact, mentioned above, that he had benefited from the general deflation, and not to any "leveling spirit".

The Preamble to the Constitution affirms the intention to "establish domestic justice" and "insure domestic tranquility". These aims can be unpacked, in part at least, as protecting creditors and bondholders and as authorizing the federal government to crack down on Shays-like uprisings. As noted, for some

observers, the most valuable part of the constitution was the prohibition in Art. I.10. For others, it was the authorization in Art. I.8: "Congress shall have power [...] to provide for calling forth the militia to execute the laws of the union, suppress insurrections, and repel invasions", the affirmation in Art. IV.4 that "The United States shall [...] protect [every state] against invasion; and on application of the legislature, or the executive (when the legislature cannot be convened) against domestic violence", or the affirmation in Art. I.9 that "The Privilege of the Writ of Habeas Corpus shall not be suspended, unless when in Cases of Rebellion or Invasion the public Safety may require it". These clauses go far beyond anything found in the Articles of Confederation. o

The intention to protect bondholders in the future must be distinguished l

from the desire to protect existing bondholders and possessors of paper money d

issued during the war. On the latter point, the Convention offered some but not r

full protection. The constitution does not explicitly affirm that the new gov- 3

ernment would redeem federal debts at face value. A proposal to that effect was :

made at the Convention, but rejected as liable to "beget speculations" (Farrand a

1966 II, 413). Instead, the more neutrally phrased Art. VI.1 of the Constitution f

was adopted with the understanding that it would in fact ensure redemption. O

Proposals to give Congress either the power or the duty to redeem state debts g

were rejected. Ellsworth later claimed that Gerry had proposed that O

"Continental money be placed upon the footing with other liquidated securities a

of the United States" (Farrand 1966 III, 171; emphasis added), and that Gerry, e

who "was supposed to be possessed of large quantities of this species of paper", Y

refused to sign the Constitution because the Convention rejected the proposal. 3

Gerry violently rejected this explanation of his behavior (id. 240). i

Gerry also held federal securities, and repeatedly urged for writing an explicit S

duty to redeem them in the constitution. Three other framers (Sherman, King, O

and Ellsworth) also argued for assumption of the very kinds of debt (and only >

those kinds) that they possessed themselves (McDonald 1982, 105-106). 3

Overall, however, and contrary to the "Beard thesis" as commonly—and 2

wrongly—understood, it is unlikely that the decisions of the framers were 5

shaped by their personal economic interests. They may, to be sure, have been shaped by the interests of their constituencies and by the belief that redemption would facilitate ratification.

I believe that the framers overreacted to Shays' rebellion. In a letter to William Smith on November 13, 1787, Jefferson wrote that "Our Convention has been too much impressed by the insurrection of Massachusetts: and in the spur of the moment they are setting up a kite [a hawk] to keep the hen-yard in order. I hope in God this article will be rectified before the new constitution is accepted". I do not know which of the articles citing "rebellion" (Art. I.9), "domestic violence" (Art. IV.4), or "insurrection" (Art. I.8) Jefferson had in

mind. It is clear, however, that he thought the framers had adopted the article in question under a sudden emotional impulse, "in the spur of the moment".

It is impossible to prove, to be sure, that their fear was visceral rather than prudential. My grounds for believing it was in fact visceral can be summarized in the first half of a verse by La Fontaine: "Each believes easily what he fears and what he hopes". The facts do not confirm the claim—made over and over again at the time—that farmers were "levelers" and demanded "agrarian laws". As noted above, the demand for a reduction of the salary of the Massachusetts governor was merely a claim to adjust his salary for deflation. Nor is there any evidence that farmers were trying to "avoid" or "evade" debt repayment. There was no particular reason why the scarcity of specie that led farmers to bankruptcy could not be remedied by the emission of paper money, as it had been before 1776. After the war, the country would not incur the extraordinary expenditures that had led to the emission of unsecured paper money.

The Constitution—or rather the Hamilton plan that it made possible—did /

defuse class warfare. It satisfied the landed as well as the mercantile interests by .

allowing the imposition of federal tariffs and excises and using the proceeds to 0

redeem the state and federal bonds at face value. The Constitution achieved this o

result, however, by overcoming the inefficiency due to the decentralized nature 3 of the Confederation (Dougherty 2001), not the alleged injustice of state r legislation. 3

3. FRANCE: PEASANTS AND DEPUTIES, SOLDIERS °

AND CROWDS i

In France, the link between violence and constitution-making took different O forms. Violence in the countryside or in the towns had no role in the calling of the Estates-General, which later transformed itself into a constituent National Assembly. Louis XVI summoned the Estates-General because the state had run out of money, in part because of his financial assistance to the American revolutionaries, and he needed the nation's agreement to his tax proposals. In contrast, violence in the countryside as well as in Paris had a decisive influence on the constitution itself.

Lefebvre (1973) and Markoff (1996) offer the outstanding syntheses of violence in the countryside in the spring and summer of 1789. (In quoting from the imperfect English edition of the former book I often modify the text.) With regard to state violence and popular violence in Paris over the summer of 1789, Caron (1906-1907) and Godechot (1989) contain perhaps the most complete discussions. Kessel (1969) is the best monograph on the events on August 4, 1789, when the constituent assembly literally abolished feudalism overnight.

I shall discuss these events in the following steps, bypassing chronology for the sake of causal coherence. First, I consider the "Great Fear" of 1789. Second, I discuss the direct causal impact of actions in the provinces inspired by the Great Fear on the decisions the Assembly took on the night of August 4. Third, I retrace the steps of the attempted counterrevolution by Louis XVI and his entourage in late June and early July, how the assembly and the people of Paris thwarted this attempt, and some of the consequences of their actions. The violence in the provinces and in Paris is linked by a common cause, the dismissal of the King's Minister Necker in July 11. Schematically:

3.1 The Great Fear 0

The origin of the Great Fear was the belief of the peasantry that bandits in the |

pay of the nobility were out to ruin their crops and starve them, to create a state s

of anarchy that would prepare the terrain for counterrevolution. This—totally /

unfounded—fear—led them, among other things, to attack castles, burn prop- N

erty records, and to kill nobles. In the following, I flesh out some details of these §

events, with focus on the flow of information from Paris to the provinces. In the o

pinext subsection, I consider the information flow in the opposite direction, as |

well as a subsequent return flow. 0

If the main class conflict in America was between farmers and bondholders, p

the French situation was characterized by the struggle between tenants and their A

feudal lords. The relations between the peasantry and their seigneurs were in- l

tensely personal. De Tocqueville (2011, 38) has a vivid description of the daily p

frustrations and humiliations that the French peasant suffered in his encounters 1

with his noble neighbors:

These neighbors [the agents of the seigneur] arrive to take him away from his field and force him to work elsewhere for no pay. When he tries to protect his seedlings from the animals they hunt, they prevent him; they lie in wait for him at river crossings to exact a toll. At the market there they are again, to make him pay for the right to sell the produce of his land, and when on his return home he wants to use the rest of the grain for his own consumption [...], he has to take it to their mill and have the bread baked in their oven. [...] Whatever he sets out to do, he finds these tiresome neighbors barring his path,

interfering in his pleasures and his work and consuming the produce of his toil.

Over the centuries, this predatory behavior triggered a constant stream of local peasant "emeutes" (Nicolas 2008). At the same time, harsh and arbitrary taxation caused numerous tax rebellions (id.), with an equally local character. Markoff (1996, 261-252) lists the following as the main forms of resistance:

• Seizure or destruction of power-giving documents (the titles of lords, tax rolls, conscription lists).

• Sacking the residence of wrongdoers (the lord, the tax official, the official in charge of food supply, the merchant, the peasant withholding grain from the market).

• The rescue of one's fellow who might have been conscripted or seized in the wake of a resistance movement.

• Redistribution of grain or of money taken from lord or cleric. p

• Imposing costs on violators of communal solidarity by threatening or attacking peasants who might make payments to church or lord despite a boycott; who might work as laborers at unacceptable wages; who might be hauling grain to market at unacceptable prices.

As one would expect, these are all local targets. Since the peasantry could not identify grain speculators, they could not direct their attacks against them. Mutatis mutandis, the following observation on Paris bread riots probably also applied to the countryside: "consumers would be inclined to discharge much of their venom on the baker - the immediate oppressor - rather than on the far more olympian speculators" (Kaplan 1982, 8). Nor could the peasant easily march on Paris to attack the counterrevolutionary nobles at the court who were thought to pay brigands to cut the grain while still green. Instead, they diverted their anger to the local elites.

The actions had local targets and were local in their origin. Yet they were to some extent coordinated in their timing by common external causes, such as inclement weather leading to a bad harvest and, in 1789, news from Paris. "The targets of peasants actions may have been very much local ones, but the causes of peasant actions [. ..] included the decisions on national policies of revolutionary elites" (Markoff 1996, 331). As in America, there were also effects of diffusion and contagion. In France, these took three forms: "one revolt stimulates another as the repressive forces appear weaker than previously known, as those forces are actually weakened by failure, and as organizational possibilities and tactics are debated and knowledge of successful organizational models and tactics becomes widely diffused" (id. 369). The first two mechanisms were unlikely to be observed in America: weakness of repression in one state could

neither signal nor cause weakness in other states. Although in theory the failure of the Confederation to raise troops against the insurrection in Massachusetts might have served as a signal to rebels in other states, this Bayesian mechanism does not seem to have operated.

The Great Fear was not one movement, but many. According to its foremost historian, Georges Lefebvre (1973), it originated simultaneously and independently in six regions of France in mid-July 1789. Later research has identified a seventh current (Ramsay 1991). Although Lefebvre carefully distinguishes the July Fear and the actions it inspired from effects of the hunger in March through June 1789 (no region saw the occurrence of both), the two movements exhibited many of the same mechanisms.

In the spring, the dearth of grain after a bad harvest in 1788 increased the number of vagabonds, beggars, and "brigands" in the countryside. Whereas individual wanderers and beggars had always been a regular (and often intimidating) presence, the famine increased their numbers to form larger bands. I Their actions against the peasantry took different forms, from simple protection rackets to cutting the grain before harvest time. The farmer refused to sell grain to the day laborers on the grounds that he was obliged to sell it in the market of the local town. Since as nonresidents these laborers were not admitted into the towns, "there was only one course left to them - to stop the wagons in transit and seize the sacks of corn and flour, paying either the appropriate price or none at all" (Lefebvre 1973, 26). At the same time, the peasants feared the "sudden and apparently spontaneous expeditions which came out from the towns and went from farm to farm buying corn - or, more accurately, forcing

the farmers to sell their stocks" (id. 28). The effect of these actions was to create i

a climate of generalized fear and suspicion, where not only town and country- S

side were in a state of reciprocal terror, but "peasants in rebellion became O

objects of fear for one another. Those who revolted rarely accepted a refusal >

to join them. [...] Every revolt made the peasant want to join it, while at the 3

same time scaring him. The people frightened itself (se faisait peur à lui-même) " 2

(id. 55-56). 55

The dearth of grain in the spring triggered suspicions that it was a result of hoarding by speculators. "The people were never willing to admit that the forces of nature alone might be responsible for their poverty and distress" (Lefebvre 1973, 24). In eighteenth century France, every famine triggered conspiratorial explanations. Although these rumors might be false in any given case, they were often justified (Kaplan 1982). The suspicions might take one of two forms. In its less virulent and more rational version, the peasantry believed that the conspirators were profit-seekers who were merely indifferent to their welfare. In a more extreme form, they believed that the main goal of the conspirators was to reduce them to starvation. In the spring of 1789, the first form seems to have

predominated, which may explain why "subsistence events" dominated "anti-seigneurial events" in this period (Markoff 1996, 276). The fear of brigands was endemic but local: there was no thought that the plundering was orchestrated at a national level. Also, the assumption was that the brigands were acting, like the peasants themselves, out of hunger. In July, the second form of suspicion emerged. The peasants feared that the aristocrats had enlisted the brigands to cut the unripe corn, for the purpose of creating chaos and anarchy that would undermine the Revolution. Mixed with this was a fear of foreign invasion, supposedly organized by the King's reactionary brother Comte d'Artois from his exile in Savoie. As noted, this fear was completely groundless. There was no conspiracy to ruin the countryside.

Although irrational, the Great Fear was intelligible. To understand it, we can cite motivational as well as cognitive mechanisms. On the motivational side, we can go back to the convocation of the Estates-General at the end of 1788. The peasantry tended to interpret this event as a promise that the King was going to I alleviate their misery. Great hopes were raised. When the Estates met and remained completely inactive for six weeks in May-June, the hope was replaced by a growing suspicion that the privileged orders were sabotaging the will of the King. When news about the dismissal of the King's liberal minister Necker on July 11, the fall of the Bastille on July 14, and the flight of Comte d'Artois on July 17 reached the provinces, the suspicion of an aristocratic conspiracy against the people hardened into a certainty.

On the cognitive side, many peasants were probably afraid of expressing their disbelief in the rumors, thereby creating a situation of pluralistic ignorance (few

people believe that p but most people believe that most people believe that p). g

Also, "there was a risk in revealing one's skepticism. Those who made too obviously a play of it and refused to take defensive measures might perhaps 0 be seeking to lull the people's suspicion. [...] The danger arose all the more >

rapidly because the people who brought the news felt their amour-propre 3; damaged if they were not taken seriously and they were very likely to spread 2 malignant gossip about those who refused to believe them." (Lefebvre 1973, 5 153.). Refugees tended to exaggerate the danger lest they be accused of cowardice for having run away (id. 148). In many towns "authorities were delighted to be able to shift the blame [for violence] from local people to unknown brigands [...]; the intendants accepted these versions without batting an eyelid and contributed to spread them" (id. 128). Misinterpretations of accidents and natural phenomena also contributed to the general panic (id. 94, 131, 144, 145, 164, 166, 168, 189). As in the spring, the people frightened itself (id. 116, 123).

The Great Fear inspired a great number of "anti-seigneurial events", including violence against persons or property, invasion of castles with varying

degrees of damage, destruction (rather than seizure) of food sources, coerced renunciation of rights, seizures of land charters, damage to seigneurial mills, ovens and winepresses, refusal to pay rent, and numerous others (Markoff 1996, 221). Lefebvre (1973, ch. II.5) also cites refusal to pay rent, destruction of seigneurial pigeon houses, burning of archives, sacking or looting of castles, arson, and many others. Personal violence against the lords, on some occasions resulting in their death, occurred in 3 percent of antiseigneurial events, as against 53 percent that involved property damage (Markoff 1996, 221). Although small in relative terms, the number and gruesome detail of the massacres were large enough to focus the attention of contemporaries (Ferrieres 1880, 120-121).

The anger towards the seigneurs had, as explained earlier, deep historical roots. It was strengthened by fear of the (nonexistent) brigands that they believed the lords had organized against them. Hence anything that could strengthen the fear would strengthen the anger. The peasant belief that the /

aristocrats had enlisted the brigands to cut their unripe grain would simultan- .

eously trigger fear of the proximate cause of destruction and anger towards the C

ultimate cause; the greater the fear, the greater the anger. o

3.2 The Night of August 4th

The abolition of feudalism on the night of August 4th (confirmed by the decrees one week later) was triggered by information from the provinces to Paris. Taking account of the time pattern of antiseigneurial actions and of the time lag, Markoff (1996, 437) calculates that the reception of bad news from the n

provinces had two sharp spikes around July 28 and August 2. Many of the deputies were personally affected or threatened. In addition to the nobles, many members of the third estate held important landed properties (Kessel 1966, 19-21; Tackett 1996, 38-39). It took the deputies a few days to absorb the shock and start debating and enacting countermeasures.

Generally speaking, in the face of actual or potential rebellion a government has the choice of four responses: preemption, concession, moderate repression, and severe repression. Wisdom dictates preemption—meeting popular demands before they are formulated, or granting more than is demanded. After July 14, 1789, that option was not on the table. Jaures (1968, 443) was probably right in asserting that "one had the choice between organizing a very difficult and very dangerous repression throughout the countryside, and giving in to the demands from the rebellious peasantry". Moderate repression was unlikely to work. Although the government had used this strategy in the decades prior to the Revolution, its targets "were subjected to enough restraint to provoke resistance but not the heavy yoke that might quell it" (de Tocqueville 2011, 139).

Severe repression might seem more promising, and in fact "the first motions made in the Assembly all went in the direction of repression" (Jaures 1968, 443). This statement refers to a motion that the deputy Solomon presented on August 3, on behalf of the Committee on Reports (AP 8, p. 336). The first paragraph of the motion describes the violence in the countryside; the second affirms that the Assembly cannot take time off from its main task to deal with particular matters; and the third states in intransigent terms that no pretext whatsoever could justify the refusal to pay taxes or feudal dues. The proposal was sent to another committee, which reported back in almost equally intransigent terms the next evening. o After the second report had been read, the first speaker, the Vicomte de § Noailles, argued that the peasantry had to be met with concessions, not with d

repression. His speech was the first event in "the night of August 4" (see Elster r

2007 for a narrative). Other speakers followed, and at the end of the night the 3

Assembly had abolished not only the feudal regime, but virtually the whole |

system of privileges and exemptions that constituted the ancien regime, the a

courts (parlements), and the guilds (jurandes) being the only, and temporary, f

exceptions. The assembly caved in. O

Some deputies probably made up their mind because they feared a return to g

the absolute monarchy and the fear of a civil war. In terms of the distinction O

made above, these are prudential fears, not—or not necessarily—visceral fears. a

Some deputies may indeed have viewed the situation in this detached perspec- e

tive. They may have believed that a repression would cause a return to the Y

absolute monarchy or a civil war, and they did not want any of these outcomes 3

to happen. g

In contrast, deputies who believed that their property and family were under S an imminent threat from the peasantry could easily experience visceral fear. O Although they would not themselves be targets of attacks by the peasantry, >

personal danger is not a necessary condition for the triggering of visceral fear 3 (as any parent knows). In several letters from August 7 onwards by the noble 2 deputy Comte de Ferrieres (1932, 109 seq.) to his wife, one can easily read his 5 anxiety between the lines. The first letter contains very detailed instructions that she is to sell his sheep and his oxen, at any price, for cash; to gather all the money and documents in his castle in Mirebeau and transfer them to their house in Poitiers, making sure nobody observes her doing so; to ship their mattresses, bed covers and sheet to Poitiers ("in case of an event, at least something will be saved"). Three days later, he tells her to go with their daughters to Poitiers, even if the harvest should suffer: "do not consider the costs, and do not ask for [the protection of] soldiers, which would cause alarm in the countryside". He does not care if after these precautions his castle is burned, as he is never going to live there again.

His fears also affected his political behavior, as shown by a letter from August 7 addressed both to his constituency and to his friend Rabreuil:

[To his constituency:] It would have been dangerous even for you if I had expressed opposition to the general wish of the nation. It would have been to designate you and your possessions to the fury of the multitude, and to have exposed you to seeing your houses burned down.

[To Rabreuil:] Mme de Ferrieres tells me that you would like me to get into the newspapers; that would be the means to lose the little credit I have in the third estate, for, at this moment, I could only speak out in opposition to what is being done; at least in great part; that would be pointless. Thus I keep silent, as do M. de Clermont., M. de Sulli, Mounier, and wise people. If I alienated the third estate in questions touching on the interests of my electoral districts, I would experience difficulties, if only because of the spirit of revenge (id. 118119).

In a letter of August 12 he also asserts that "the inhabitants of Mirebeau, r

who had so many good reasons to treat me well and who even, because of the g

way I have always behaved towards them, ought to have special consideration el

for me, have acted with such insolence and fury that I cannot count on their g

good will". His fear may, then, have spurred him to act on two fronts: to reduce £

his vulnerability to invasion of his castle by transferring or selling his most §

valuable possessions, and to reduce the likelihood of an invasion by voting for ^

measures that might satisfy and pacify the peasantry. His example was not an U

isolated one. e

Tocqueville (2011, 157) claimed that concession, like moderate repression, is jy

likely to be ineffective in quelling rebellion. "The evil that one endures patiently £

because it appears inevitable becomes unbearable the moment its elimination i

becomes conceivable." Lefebvre (1973, 38) offered a similar argument: "as hope ,

sprang in the people's breast, so did hatred for the nobility: in the certainty of 1 royal support, the peasants, invited to speak their minds [in the grievance books], reiterated with growing bitterness their present miseries and from the depths of their memory the stifled remembrance of past injuries" (emphasis added). Along similar lines, Jaures (1968, 469) wrote that

Not only did the nobles think that the abolition of the tithe without compensation would increase their income from land, but they believed above all that this immediate satisfaction obtained at the expense of the clergy would make the peasantry less eager to

pursue the abolition of the feudal dues: they hoped to divert the storm towards the goods of the church. What a poor calculation! Quite to the contrary, the peasants were all the more unlikely to accept the need for compensation with regard to the feudal dues as they had been dispensed with compensation for the tithe.

From these arguments it follows that when news about the decrees adopted on August 4th reached the provinces, antiseigneurial actions ought to become more rather than less frequent. This consequence was in fact asserted by contemporaries such as Rivarol (1824, 152) and Dumont (1832, 104). Markoff (1996, 443) claims, however, that the opposite effect was observed: "No sooner did [the deputies] complete their legislative work on the eleventh [of August] than the countryside, almost instantly, subsided into something which if not quite peace was at least far less dramatically threatening than for a long several weeks. Their own words must have seemed to possess magical powers". They would indeed have needed magical retroactive powers, as the diagram on page 437 of Markoff (1996) shows that the troubles subsided before news about the decrees of the 4th and a fortiori those of the 11th could possibly have reached the provinces. Nevertheless, they did not resume when the news arrived. In a longer time perspective, however, it seems clear that the effect of the measures of August 1789 was to inflame rather than to pacify the peasant furies.

3.3 A Counterrevolution that Triggered a Revolution

At the outset, nobody expected the Estates-General to take very radical meas- g ures. The main demand from the third estate was for the abolition of the privileges of the nobility, in particular their exemption from most taxes and their exclusive right to higher military office. In addition, many demanded reform of the many arbitrary features of the regime. There was no demand for the abolition of the feudal property system or even for a constitutional monarchy. Although the events I shall describe shortly are commonly referred to as an attempted counterrevolution, this label is in fact somewhat misleading. It was rather the other way around: it was the failed attempt by the King to intimidate the assembly by gathering troops to Paris that triggered the revolution, both in Paris and in the provinces (see above).

On June 17, after six weeks of bickering over procedure, the Estates-General finally constituted themselves as a National Assembly in a single chamber, rather than as three separate estates. The King took this decision as a direct attack on his authority, since he had ordered the estates to deliberately

separately and to vote by order, except if they all agreed to merge and to vote by head. His initial reaction was to close the assembly hall to the deputies, but they reassembled in an indoor tennis court and swore to remain together until the constitution had been adopted. On June 23, the King spoke to the assembly, promised some reforms, and again ordered the deputies to deliberate separately by estate. When they refused, he backed down on June 27, but began to prepare a military operation against the assembly.

In this "counterrevolutionary" effort, the King was largely a passive tool in the hands of a small coterie that included his wife, his youngest brother Comte d'Artois, and some nobles. On June 26 and on July 1 the King issued orders for troops to converge on Paris, probably amounting to a total of 20 000 (Caron 1906-1907, 14). Rumors of their imminent arrival began in late June, and triggered what an historian has called, with reference to the events described above, a "Great Fear" among many deputies (id. 22).

The precise intentions of the conspirators remain unclear to this day. P

Because of the indecisive nature of the King, plans vacillated constantly. It is .

likely that some scheme of either arresting the most radical deputies or of 0

moving the whole assembly to a location more remote from Paris was envi- o

saged. Many deputies certainly perceived the troop movements as direct threat 3

to the assembly, and some feared being arrested. In a masterful speech, o

Mirabeau, addressing himself directly to the king, cleverly avoided the poten- 33

tially treasonable language of threats and used instead the less objectionable— e

but equally explosive—language of warnings (see Elster 2000 for this ^

distinction): £

The danger exists for the people of the provinces. Once alarms have e

been raised alarmed about our freedom, we do know what can retain them. The very distance makes everything appear larger, exaggerates everything, multiplies, envenoms and embitters the worries.

The danger exists for the capital. How will the people - in the midst of scarcity of food and tormented by the most cruel anxieties -perceive it when a crowd of threatening soldiers are fighting over what remains of their subsistence? The presence of the troops will heat up and draw out the public opinion, and produce a universal fermentation; the first act of violence, carried out under the pretext of maintaining order, may trigger a horrible series of disasters.

The danger exists for the troops. French soldiers, who are close to the center of discussions and share the passions as well as the interests of the people, may forget that a contract made them soldiers and remember that nature made them men.

The danger, Sire, is threatening the efforts that are our main duty, and which can only succeed fully and durably if the people perceive

them as entirely free. Passionate movements are subject to contagion; we are only men; our distrust of ourselves, the fear of appearing to be weak, may carry us beyond the goal; we will become obsessed by violent and excessive proposals (AP 8, 213).

The troops did indeed remember that they were men and citizens before they were soldiers. Although they had been called in from the provinces because they were supposedly more reliable than the French Guards in Paris, they soon melted into the population and become utterly unreliable as an instrument of repression. The conspirators did not master the most elementary techniques of a coup d'Etat (Caron 1906-1907, 659):

Instead of dispersing the troops in Paris and around Paris and exhibiting them everywhere, which had the triple disadvantage of fragmenting the forces, exciting the spirits and exposing the soldiers to demoralizing influences, one should have assembled them some distance from Paris in a compact body, fed them well, had the King visit them, ensured their cohesion, fortified their military spirit, impose - if necessary by some severe examples - a strict discipline, take them in full charge; then, once all the troops were in place, strike quickly and strongly (id. 657-658).

When the King dismissed his liberal minister Necker on July 11, public opinion in Paris immediately interpreted the decision as part of a scheme to attack the assembly. The troops offered no resistance to the people of Paris when they invaded the Hotel des Invalides in search of arms on July 13 and the Bastille in search of gunpowder on July 14. Moreover, the strong group of i rentiers in Paris with large investments in state bonds supported the insurrec- | tion wholeheartedly (Caron 1906-1907, 666; Godechot 1989, 312ff.). They O counted on the assembly to authorize new taxes and on Necker, a renowned A financier, to work with the assembly to rescue the state finances. Conversely, H during the decisive debates about the location of the Estates-General, when 2 many of the King's advisers wanted them to be held at a safe distance from 5 the crowds in Paris, Necker preferred Paris because he thought the proximity to the capital market in Paris would have a moderating influence on the assembly (Egret 1975, 249-250).

Louis XVI, once again, had to back down by recalling Necker on July 17. After the failure of the counterrevolution, some moderate members of the assembly wanted to relocate it to the provinces to keep it sheltered from the crowds in Paris. Clermont-Tonnerre asked his fellow deputies, "You did not obey armed despotism; are you going to obey popular effervescence? The former commanded crimes, the latter will command vileness. You cannot deliberate in the midst of fifteen thousand armed men whose projects are

unknown and whose character is in tatters" (AR 8, 513-514). The dominant group of the day—the "triumvirs" Barnave, Duport, and Alexandre Lameth— successfully resisted this motion. According to his biographer, Duport thought that "the conquests of the revolution far from being consolidated - the large judicial and administrative reforms had not yet been undertaken -were at the mercy of an aristocratic counteroffensive, and that the moment had not yet arrived to calm the popular ardors" (Michon 1924, 67; emphasis added). This widely shared idea that popular violence could be switched off and on according to the political goals of the day (Droz 1860, vol. II, 213) was to prove disastrous.

These developments decisively undermined the efforts of the monarchiens in the assembly to create a constitutional monarchy on the English model (Egret 1950). The proposals to give the King an absolute veto and to create a bicameral assembly were defeated by large majorities on September 10 and 11. In producing these majorities, both the fear of violence and the hope of violence played a crucial role. The vote on bicameralism is particularly instructive. The supporters of unicameralism formed a coalition of the extremes. The left supported the measure because they feared that the upper house of a bicameral legislature would become a tool of the aristocracy and veto radical proposals. The right supported it because they thought a unicameral system would produce anarchy and chaos, thus preparing the grounds for a restoration of the ancien régime. r

Some supporters of bicameralism may have voted against it because the left t

insisted on a roll-call vote, which would expose them to popular violence if they e

voted sincerely (Egret 1950, 132). The second and third of these motives are ^

clearly stated in a letter that one of the monarchiens wrote to his constituency: §

Some deputies from the third estate have told me, I do not want my r wife and children to have their throats cut. The bicameralist proposal

had yet another kind of adversary, those who regret the ancien régime §

and want the new one to be so bad that it cannot subsist. I have i

received on this topic confidential communications that I met with ,

neither gratitude nor politeness. These are two strange bases for a 0 constitution, the fear of being assassinated and the desire to make it collapse (Lally-Tolendal 1790, 141).

One of these communications may have come from the eloquent and reactionary Abbe de Maury, who repeatedly expressed the wish that things get worse so that they would eventually get better (Droz 1860, vol. II, 343; Montgaillard 1827, vol. 1, 428). Some of the votes in the night of August 4th may also have been motivated by the goal of crisis maximization, the politique du pire (Kessel 1969, 132). Thus although the fears of the peasants that the nobility was deliberately starving them to create anarchy were unjustified, the strategy of crisis maximization was not a mere product of their fantasy. The King himself

increasingly relied on it (de Priest 1929, vol. II, 25). The decision by the assembly on May 16, 1791 to render the constituants ineligible to the first ordinary legislature was also due to a coalition of the two extremes. Both wanted the first legislature to be made up of inexperienced men: the right believed that they would easily be dominated by the King, and the left that they would easily be dominated by the Jacobin clubs. Ex post, the left proved to be right. Ex ante, the outcome was less obvious, since the vote was taken before the King's flight to Varennes that fatally undermined his legitimacy.

4. CONCLUSION: JUSTICE AND EFFICIENCY

Both constitution-making processes had the effect of shifting the locus of decision-making from social or geographical subunits towards a centralized government. In France, the three estates were abolished and replaced by a national assembly. In America, the thirteen states lost much of their power to the Union. Historically, the decentralized forms had proved to be inefficient. Each estate or state wanted to benefit from government protection, while being reluctant to pay the taxes needed to fund it. The new constitutions made it possible to overcome this suboptimal situation.

These consequences of the constitutions do not, however, provide anything like a full explanation of their adoption. Madison's notes on "The vices of the present system" can probably be taken as representative of the views of leading members of the Convention. Although they certainly dwell on the inefficiency of the Confederation, the strongest language is devoted to denouncing the injustice n of state legislation. In France, the calling of the Estates-General was certainly motivated by the need to establish a more efficient tax system. Yet as events unfolded, that aim receded in comparisons with the demand for economic justice and political representation.

The idea of justice was in fact at the core of both processes. In America, two conceptions of justice confronted each other head-on.

On the one hand, an elite conception of justice held that issuing money and bonds entailed a morally binding promise that these instruments would keep their value. Comparisons with female chastity were common. In 1779, the Continental Congress issued a statement that a bankrupt republic would "appear among reputable nations like a common prostitute among chaste and respectable matrons" (Journals of the Continental Congress, vol. 15, 1060). In 1784, "Philadelphia bondholders petitioning against plans to withhold their annual interest declared that 'credit may be considered as the chastity of the state'. For the government to pick and choose among its creditors - to allow original holders' claims while denying 'an interest of 40 or 50 per cent [to] a few

speculators' - would be 'as indelicate, as it would be to measure female honor by calculations in arithmetic'" (Holton 2007, 94-95). In 1786, a correspondent to a New York newspaper recalled with nostalgia the times "before the commencement of the late war, when public faith was still in the possession of vestal chastity [and paper money] circulated freely and at its full nominal value on a perfect equality with specie" (Ferguson 1961, 18).

Normative and ideological overtones are common in debates over monetary policy, inflation, and depreciation. In A Tract on Monetary Reform, Keynes (1923, 67-68) denounced the economists and bankers who fulminated against devaluations and capital levies "on the grounds that they infringe the untouchable sacredness of contract" and regarded it "as more consonant with their cloth, and also as economising thought, to shift public discussion of financial topics off the logical on to an alleged 'moral' plane". In a remark whose relevance to my main topic here is obvious, he added "the fact that in time of war it is easier for the State to borrow than to tax [cannot] be allowed permanently to enslave the tax-payer to the bond-holder". In a much-quoted and equally relevant phrase, he referred to "the absolutists of contract" as "the real parents of Revolution".

In the Great Depression, "the gold standard provided just such an ideology, supported by a rhetoric of morality and rectitude. Its rhetoric dominated discussions of public policy in the years leading up to the Great Depression, and it sustained central bankers and political leaders as they imposed ever greater costs on ordinary people" (Eichengreen & Temin 2000, 207). More specifically,

Treasury Secretary Andrew Mellon advised President Hoover that the

only way to restore the economy to a sustainable footing was to u

"liquidate labor, liquidate stocks, liquidate the farmers, liquidate real e

estate... purge the rottenness out of the system...". "People will work harder", Mellon insisted, and "live a more moral life". Those espousing the puritanical strand of gold-standard dogma grew more strident as unemployment mounted. Hoover himself regarded the gold standard as "little short of a sacred formula". Any deviation he dismissed as "collectivism", an all-embracing label for economic and social decay (id. 196).

In the 1780s, some Americans took a similar puritanical and moralistic point of

Early in 1787, a Marylander contended that the "lax principle in our laws, and the administration of justice, ha[d] greatly tended ... to relax the natural springs of industry". "A Native of Virginia" was blunter, declaring that "the relaxation of our laws" had led to "inactivity and torpor". It followed that "accelerating the Adm[inistrat]ion of Justice" would actually relieve "the present distresses of the Countrey", as

another Virginian, Edmund Pendleton, contended in a December 1786 letter to James Madison. How? By "producing Industry & Oeconomy" among debtors" (Holton 2007, 99).

These views rest on a confused amalgam of morality and causality, appealing to the sacred character of promises, the purifying effects of hard work, and to an assumption of a backward-sloping supply curve of labor. Many no doubt believed that strict adherence to the gold standard and redemption in specie of war bonds at full value were to be recommended on grounds both of justice and efficiency. Just as Hoover saw any deviation from the gold standard as collectivism, critics of paper money claimed that it was equivalent to leveling and confiscatory agrarian laws. The 1780s and the 1930s seem to have differed in one respect. Hoover was a deluded ideologue: he did not defend the gold standard because he stood to benefit from it personally. In contrast, many advocates of hard money redemption had much to gain if that policy were adopted, at least if it were restricted to domestic debts. Many Virginians, including Washington and Jefferson, tried to pay off their British debts in depreciated paper money (Smith 1998, 153-154).

On the other hand, a popular conception of justice condemned policies that caused hardships for taxpayers—farmer and veterans—for the benefit of speculators. In 1784, the son of the general who was to suppress Shays' Rebellion "said the conflict between creditors and the parallel dispute pitting 'the creditors of the public, particularly of the army', against taxpayers had both 'arisen from [a] principle of opposition, against the interests of those, whose subsistence is derived from the labours of others'" (Holton 2004, 283). A correspond- n ent writing under the name of "Justice" "urged the Connecticut legislature to scale down the war bonds so citizens would not be 'unjustly taxed to pay more than the real value'" (id. 285). Another Connecticut writer said that the value of the securities when they bought them, with interest, "is all [the speculators] can justly demand" (Holton 2007, 56).

The popular sense of injustice was fueled by the belief that the speculators were behind the legislation that would enable them to reap astronomical profits. Their gains were not the fruits of productive labor, but the reward to lobbying. The elite, to be sure, disagreed. In a careful weighing of the claims of different bondholders, Madison (1790) affirmed that the profit of speculators was a just reward to risk: "the holders by assignment, have claims, which I by no means wish to depreciate. They will say, that whatever pretensions others may have against the public, these cannot effect the validity of theirs. That if they gain by the risk taken upon themselves, it is but the just reward of that risk. That as they hold the public promise, they have an undeniable demand on the public faith".

To this argument, the popular conception of justice could retort that the original sales of the bonds had been bargains of desperation (Holton 2007, 90).

The sense of injustice among the French peasantry had different roots. Their three adversaries—the agents of the seigneur, the officials of the royal administration, and the speculators—did not trigger quite the same emotions. Each of the first two groups was part of a social contract with the peasantry, providing law and order, such as it was, in exchange for feudal dues and taxes. There were innumerable occasions for arbitrary and exploitative behavior that triggered émeutes in the people (the word signifies "riot", but has the connotation of "emotion" as well). For the period 1661-1789, Nicolas (2008, 53 and passim) c

has identified 439 antiseigneurial events and 3336 antifiscal events, an event §

being defined as an act of violence against person or property committed by at |

least four individuals (id. 39-40). The real numbers are certainly much higher. c

Yet I conjecture that in the main these events were reactions to deviations from h

practices that were, themselves, unquestioned. /

The speculators had no redeeming features. In the ancien régime, "the trader c

remained the prototype of the liar who menaced the well-being and the bonds °

of solidarity of society. The grain trader was especially odious, for who but U

vicious men would speculate on the subsistence of their fellow citizens?" §

(Kaplan 1982, 63). In the abstract, speculation may have some benefits. r

Arrow (1982) observes that "when situations of scarcity arise, hoarding always blamed. But the evidence for the degree and effects of hoarding is usually W

difficult to come by. [...] If the famine is prolonged, then hoarding at the c

beginning means greater stores will be available later on". (Today, some econo- £

mists defend short-selling and even "naked short selling" as providing a valu- e

able social service.) Yet such notions had no grip on the popular imagination. y

The idea that speculators in grain might be the functional equivalent of gran-

aries and provide an intertemporal redistribution of consumption does not TT

seem to have crossed anyone's mind. Speculators and hoarders were hated 0

because of their perceived indifference to the welfare of the people. 0

In America, the calling of the Federal Convention occurred when the elite's u'

conception of justice was hurt by violent actions inspired by the popular conception of justice. By a small miracle—the Hamilton plan—the federal government was able to resolve the tension between these two conceptions of justice, by assuming the debts at full value without resorting to direct taxation. It is difficult to know how much one should credit the Convention or the Constitution for this result. A federal impost or excise tax made it possible to fund an army to crush insurrections, as the Whiskey Rebellion of 1794 demonstrated very effectively. It also enabled the establishment of a sinking fund to service the public debt. I conjecture that in the minds of the framers, the first

effect loomed larger than the second. The American constitution created an instrument of state violence to repress popular violence.

In France, the constituent assembly abolished feudalism and effectively ended the monarchy. Popular and state violence had a crucial role in bringing about both effects. The assembly acted as the American state legislatures had done a few years before, responding to rural violence with large concessions. The attempted state violence generated urban counterviolence that fatally undermined the authority of the King. Even before the Terror, the scale of violence in France was vastly larger than anything seen in America, a fact easily explained by the depth and strength of the institutions that had to be overcome. The French constitution was born of popular violence, as a response both to secular feudal exploitation and to failed state violence.

REFERENCES f

Amar, Akhil. 2005. America's Constitution: A Biography. New York: Random f House. o

AP = Archives Parlementaires, Serie I: 1787-1799, Paris 1875-1888. Arrow, Kenneth. 1982. Review of A. Sen, Poverty and Famines. In The New York

Review of Books, 29, pp. 24-26. Benson, Lee. 1960. Turner and Beard. Glencoe IL: The Free Press. Bouton, Terry. 1996. Tying up the Revolution, PhD Dissertation, Duke University.

-. 2007. Taming Democracy. New York: Oxford University Press.

Brown, Roger. 1993. Redeeming the Republic. Baltimore: John Hopkins S University Press. O

Brynner, Rock. 1993. Fire Beneath our Feet: Shays's Rebellion and its >

Constitutional Impact, PhD Dissertation, Columbia University, New York. 3 Caron, Pierre. 1906-1907. La tentative de contre-révolution de Juin-Juillet 2 1789. Revue d'histoire moderne et contemporaine. 8, 5-34, 649-678. 5

de Ferrieres, Marquise. 1932. Correspondance inédite. Paris: Armand Colin. de Montaigne, Michel. 1991. Essays. London: Penguin.

Dougherty, Keith. 2001. Collective Action under the Articles of Confederation.

Cambridge: Cambridge University Press. Droz, Joseph. 1860. Histoire du règne de Louis XVI. Paris: Renouard. Dumont, Etienne. 1832. Souvenirs sur Mirabeau. London: Bull. Egret, Jean. 1950. La révolution des notables. Paris: Armand Colin.

-. 1975. Necker, Ministre de Louis XVI. Paris: Champion.

Eichengreen, Barry, & Peter Temin. 2000. The Gold Standard and the Great Depression. Contemp. Eur. Hist. 9, 183-207.

Elster, Jon. 1984. Ulysses and the Sirens, rev. edn. Cambridge: Cambridge University Press.

-. 2000. Arguing and Bargaining in Two Constituent Assemblies. Univ.

Pennsylvania J. Constitutional Law 2, 345-421. -. 2007. The night of August 4, 1789. A study of social interaction in

collective decision-making. Revue Européenne des sciences sociales 45, 71-94.

Eyck, Frank. 1968. The Frankfurt Parliament. London: Macmillan.

Farrand, Max. 1966. Records of the Federal Convention. New Haven CO: Yale University Press.

Ferguson, E. James. 1961. The Power of the Purse. Williamsburg VA: University of North Carolina Press.

Ferrieres, Charles-Elie, & Marquise de. 1880. Memoires. In M. de Lescure (ed.)Bibliothèque des mémoires, t, XXXV. Paris: Firmin-Didot.

Gazell, James. 1970. One Man, One Voice. Western Pol. Q. 23, 445-462.

Godechot, Jacques. 1989. La prise de la Bastille. Paris: Gallimard.

Grubb, Farley. 2006. The US Constitution and Monetary Powers. J. Finan. Hist. 13, 43-71.

Gordon, Robert. 1987. The Structure of Emotions. Cambridge University Press.

Holton, Woody. 2004. 'From the Labours of Others': The War Bonds

Controversy and the Origins of the Constitution in New England. r

William Mary Q 61, 271-316. t

-. 2007. Unruly Americans. New York: Hill and Wang. e

Jaurès, Jean. 1968. Histoire socialiste de la Révolution Française, vol. I. Paris: ^

Editions Sociales. R

Jefferson, Thomas. 1784. Notes on the State of Virginia. Paris: Private printing. B.

Jensen, Merrill. 1968. The Founding of a Nation. Indianapolis IN: Hackett. i

Kaplan, Steven. 1982. The Famine-plot Persuasion in Eighteenth-century n

France. Trans. Am. Phil. Soc. 72, 1-79. P

Kessel, Patrick. 1969. La nuit du 4 août 1789. Paris: Artaud. 3

Keynes, John Maynard. 1923. A Tract

on Monetary Reform. London: Macmillan. 2

Lally-Tolendal, Trophime Gerard, & Comte de. 1790. Seconde lettre è ses com- 5

mettans. Paris: Desenne.

Lefebvre, Georges. 1973. The Great Fear of 1789. New York: Schocken.

Madison, James. 1790. Speech on the Public Debt, February 11. In The Writings of James Madison, vol. 5.. New York: Putnam.

Markoff, John. 1996. The Abolition of Feudalism. University Park PA: Pennsylvania State University Press.

McDonald, Forrest. 1982. We the People: The Economic Origins of the Constitution. New Brunswick NJ: Transaction Books.

McGuire, Robert. 2003. To form a More Perfect Union. Oxford University Press.

Michon, Georges. 1924. Essai sur l'histoire du parti feuillant: Adrien Duport. Paris: Payot.

Montgaillard, Guillaume Honore Rocques, & Abbe de. 1827. Histoire de France.

Paris: Moutardier. Nicolas, Jean. 2008. La rebellion française. Paris: Gallimard. Peyrefitte, Alain. 1995. C'était de Gaulle, vol. I. Paris: Fayard. Pole, Jack R. 1966. Political Representation in England and the Origins of the

American Republic. Berkeley: University of California Press. Ramsay, Clay. 1991. The Ideology of the Great Fear. Baltimore: Johns Hopkins Press.

Richards, Leonard. 2002. Shays's Rebellion. Philadelphia PA: University of

Pennsylvania Press. Rivarol, Antoine. 1824. Mémoires. Paris: Baudoin Frères. Roll, Charles. 1969. We, Some of the People. J. Am. Hist. 56, 21-40. Rossiter, Clinton. 1987. The Grand Convention. New York: Norton. Russell, Peter. 1993. Constitutional Odyssey. Toronto: University of Toronto Press.

Saint-Priest, Francois-Emanuel Guignard, & Comte de. 1929. Mémoires. Paris: Calmann-Levy.

Smith, Jean Edward. 1998. John Marshall: Definer of a Nation. New York: Holt. Tackett, Timothy. 1996. Becoming a Revolutionary. University Park, PA: Penn

State University Press. Tocqueville, Alexis. 2011. The Ancien Régime and the French Revolution.

Cambridge: Cambridge University Press. Webster, Pelatiah. 1785. A Plea for the Poor Soldiers. Political Essays (1791), 269-305.

Zagarri, Rosemarie. 1987. The Politics of Size. Ithaca NY: Cornell University

Press. P