---
abstract: |
  The literature on elite conflict has paid limited attention to how locally powerful settlers challenge the colonial order under which they have prospered. We examine the 1779 Cape Patriots petition, linking its 404 signatories to household tax censuses, rural officeholding records, and deceased-estate auctions. Among households observed in Stellenbosch and Drakenstein, petitioners were wealthier, concentrated in the top quartile and wine economy, and disproportionately likely to have held rural office under our preferred linkage. Half of the eight serving district heemraden signed. Petitioners also occupied more central positions in auction networks, although much of their brokerage advantage reflected greater purchasing activity. These findings support an interpretation of elite decoupling: local wealth, standing, and subordinate office could sustain a public challenge to institutions over which settlers lacked equal authority. The evidence identifies the movement’s rural elite base; it does not establish why conflict became public in 1779.
author:
- Jonathan Schoots[^1]
- Johan Fourie[^2]
bibliography: references.bib
reference-section-title: References
title: |
  Protest from Above:\
  Elite Decoupling and Public Petitioning in the VOC Cape Colony[^3]
---

> Figures and typeset tables are omitted from this Markdown version.
> The complete paper, with all figures, is in JS_JF_ProtestFromAbove_v2.pdf.


**Keywords:** intra-elite conflict; petitioning; colonial governance; officeholding; Cape Colony

**JEL codes:** N47; D74; P16

# Introduction

In 1779, three years after the American Declaration of Independence, Dutch settlers at the Cape of Good Hope mounted their own challenge to colonial rule. The Cape Patriots (*Kaapse Patriotte*) petitioned the governing board of the Dutch East India Company (VOC) for access to trade, constraints on officials’ commerce, legal protections, and settler appointments in key councils. Like contemporaries elsewhere in the Atlantic world, they invoked citizen rights and natural law. Their programme sought institutional reform rather than secession: a reconfiguration of the boundary between local power and central political authority.

This paper asks who these petitioners were and what their economic, political, and social positions reveal about conflict in this settler-colonial setting. Were they the dispossessed, a cross-class settler coalition, or locally prominent people seeking greater control over the institutions that governed them? We find a local elite: wealthy, well connected, and unusually likely to have held rural office. We then ask what this composition implies for the movement’s public institutional demands. Petitioning could combine a recognised right of request with collective claims on government; its public use had a history long before the Cape Patriots (Zaret 1996, 2019; Oddens 2017). The puzzle, then, is why prominent Cape settlers used it to challenge Company rule.

We use this case to develop an interpretation of *elite decoupling*: a separation between locally accumulated economic and social power and control over the institutions that set the terms of colonial rule. Imperial authorities depended on local allies to project power, yet retained authority over those allies. Collaboration could therefore strengthen people whose claim to a greater role in government remained unsettled (Robinson 1972; Walters et al. 2026). The Cape case brings this tension into view through the social composition of a public challenge. It does not establish that such a separation inevitably produces protest, or that it alone explains the timing or form of the 1779 petition.

The contrast is with settings where overlapping wealth, standing, and office help elites manage conflict through negotiation and settlement (Burton and Higley 1987; Higley 2010). At the Cape, settlers accumulated land, enslaved labour, social reach, and offices that projected colonial power without giving them equal control over central policy. Informal channels existed: kinship, patronage, and commercial ties linked settlers and Company officials, and factions crossed that institutional boundary (Baartman 2011). Such connections could secure individual accommodation without satisfying demands for a collective institutional voice. Decoupling, in this account, concerns unequal authority within a connected elite, rather than the absence of relations between two separate groups.

We link the 404 signatories to household tax censuses measuring rural production assets and slaveholding; a new roll of heemraden, settler officials who projected VOC administration in rural districts (Walters et al. 2026); and deceased-estate auctions that reveal social-network position. These records locate petitioners relative to thousands of non-petitioners.

Across all three domains, challengers were more elite. Petitioners were wealthier and concentrated in the top quartile, where one in four households signed. Within that quartile they clustered in wine rather than sheep farming. Under our preferred linkage, they were about three times more likely to have served as heemraden; four of eight serving Stellenbosch–Drakenstein officeholders signed. Petitioners also purchased at more auctions and occupied more bridging positions in the purchase network. Most of the projected-network brokerage gap is reproduced when the number of distinct auctions at which each person purchased is held fixed and ties are rewired; a smaller residual remains.

The analysis contributes to debates about the social bases of mobilisation and the institutional conditions of elite conflict. Relative deprivation, populist mobilisation, and settler nationalism direct attention to different possible constituencies (Davies 1962; Gurr 2011; Laclau 2005; Jansen 2017; Veracini 2010). The rural signatories instead resemble the locally powerful contenders emphasised in accounts of elite competition (Lachmann 2000; Khan 2012; Ansell and Samuels 2014). Linking their wealth, officeholding, and auction positions makes that composition visible. The institutional interpretation connects this evidence to work on colonial collaboration and company rule (Robinson 1972; Adams 1996; Phillips and Sharman 2020): a public challenge could arise among settlers who had prospered under Company government and administered it locally.

# Structural accounts of mobilisation

Three structural accounts suggest different compositional benchmarks for the Cape Patriots: mobilisation from deprivation, cross-class mobilisation through populism or settler nationalism, and mobilisation by elites. These are guides to interpreting recruitment, not mutually exclusive theories that a signatory list can decisively adjudicate. Wealthy participants could experience relative deprivation or employ nationalist and populist language.

## Mobilisation by social and economic deprivation

Relative deprivation theories locate political discontent in a perceived gap between expectations and circumstances, which may follow hardship or a reversal of earlier gains (Davies 1962; Gurr 2011). The Patriots’ memorandum gives grounds for examining this possibility: in Baartman’s translation it warned that burghers would be “reduced to extreme poverty” (Baartman 2011, 33). Baartman also describes complaints that VOC officials used their positions for private financial advantage (Baartman 2011, 36). Our records can test material disadvantage and relative changes in recorded production assets; they cannot measure petitioners’ expectations.

Mass-society accounts draw attention to social isolation and alienation (Kornhauser 1959/2013). These are not interchangeable with low wealth or low observed network centrality. Broćić’s (2023) analysis, for example, distinguishes subjective forms of alienation and their development before mobilisation. The auction records allow us to examine commercial embeddedness, one observable dimension of social position, but not the subjective experience of alienation.

## Cross-class recruitment through populism or settler nationalism

A second body of scholarship examines how outsider-elite leaders reframe grievances to rally the economically disadvantaged and socially marginalised, channelling widespread frustration into collective action with rhetoric that pits an exploited majority against a corrupt ruling class (Laclau 2005; Jansen 2017). The Patriots’ invocation of economic hardship and VOC corruption (Baartman 2011) could signal this rhetorical strategy, with contender elites drawing on latent economic or social exclusion to bolster their numbers. A coalition of this kind could recruit both wealthy leaders and poorer supporters. Strong recruitment at both ends of the wealth distribution would be consistent with that possibility, although bimodality is not a necessary implication of populism.

An alternative axis of cross-class unity is predicted by scholarship on settler colonial nationalism. Such studies hold that settler colonies create a tripartite social structure of imperial, settler, and indigenous interrelations that shape political identity (Johnston and Lawson 2005; Veracini 2010). A common settler identity can emerge in opposition to indigenous populations whom settlers seek to dominate, destroy, or displace (Wolfe 2006). It can simultaneously emerge from opposition to imperial elites who deploy the power of colonial institutions, where settlers encounter greater political subjugation and have fewer rights than their metropolitan counterparts (Johnston and Lawson 2005). A strand of South African nationalist historiography identifies the Cape Patriot movement as a forerunner of anti-imperial nationalism in South Africa (Beyers 1967). A broad settler coalition would imply recruitment across the settler population. Concentration in one stratum would qualify that account of the movement’s social base, without ruling out nationalist language or identification among its members.

## Counter-elite challenge

Elite theory directs attention to the concentration of resources and to competition among those who control them (Khan 2012; Lachmann 2000). Network studies examine how relations among prominent actors organise political alliances (Bearman 1993; Padgett and Ansell 1993); Ansell and Samuels (2014) emphasise demands by rising economic groups to constrain incumbent political power. If the Patriots were such a challenger elite, participants should be drawn disproportionately from wealthy and locally prominent settlers. Whether they were *rising*, and why they challenged publicly, require evidence beyond that composition.

# Elite decoupling under colonial rule

If the Patriots represented an elite challenge, why did conflict in a small colony become a public contest rather than backstage mediation? We propose elite decoupling: public intra-elite challenge can emerge when economic and social power are separated from political-institutional control.

## Elite cohesion and conflict

One body of scholarship has sought to explain long-running political stability, seen in cases such as the United States, Great Britain, or Sweden. Scholarship here has pointed to key moments of elite integration and settlement between elite groups (Burton and Higley 1987; Higley 2010), followed by a context of elite cohesion, facilitated by a small, densely overlapping elite who combine close network connections with a common normative consensus (Mills 1956; Burton and Higley 1987; Higley 2010; Rohr 2025). In the Cape context – a small settler society governed by a large and powerful international chartered company – one might expect unity among a small and consolidated elite, with power established through VOC governance institutions, backed by force, and with patronage networks or personalised concessions to dissipate elite conflict.

A second body of work examines how elite conflict can transform political orders. Lachmann’s (2000) comparative analysis places competition among elites controlling different resources at the centre of institutional change. Bearman’s (1993) study of Norfolk relates changes in local elite social structure to political alignments, while Padgett and Ansell (1993) explain Medici power through the organisation of multiple networks and the possibilities these created for ambiguous, flexible political action. These accounts make relations among elites consequential; they do not imply that social connectedness always absorbs conflict.

## The decoupling of economic, social, and political power in colonial settings

Where wealth, status, and office overlap, shared roles and agreement on political rules can facilitate negotiation. Economic prominence may convert into institutional access: in the United States, for example, property qualifications persisted in local voting and officeholding long after their retreat at other levels (Einstein and Palmer 2021). But wealth can also challenge existing political arrangements. Ansell and Samuels (2014) link rising economic groups’ demands for institutional protection to elite competition. Puga and Trefler (2014) trace a different sequence in Venice: commercial expansion first widened political participation, before a wealthy group made access hereditary and restricted economic competition. Neither history supplies a ready-made model of the Cape; both show why the relationship between economic resources and political access must be established rather than assumed.

Colonial government gives this relationship a particular institutional form. Imperial authorities seek to retain control while relying on locally embedded allies for knowledge, resources, and administrative capacity (Robinson 1972). At the Cape, that reliance linked the Company’s officials to a settler elite whose prosperity and local authority were themselves products of colonial rule (Walters et al. 2026). Collaboration could maintain the governing order while also furnishing local actors with resources to press for a greater share of authority.

The chartered company combined sovereign powers with commercial objectives (Phillips and Sharman 2020). The relevant institutional distinction at the Cape was a specific one. The Council of Policy was reserved for senior Company officials. Burghers did participate in the Council of Justice when cases concerned burghers, and they held other civic and district offices (Baartman 2011, 68). What they lacked was equal authority over central policy, rather than all access to colonial government. The petition’s demands to enlarge burgher representation in the Council of Justice reveal this distinction particularly clearly (Baartman 2011, 38–39).

An instructive comparison is the 1653 New Netherland Remonstrance. Middleton (2010) locates the delegates’ demands within disputes over legitimate authority under another Dutch chartered company. The delegates included experienced local officeholders who asserted the standing of established settlers while professing allegiance to the Company. This is a comparison of institutional claims, not evidence that identical divisions or causes produced the two petitions.

The proposed mechanism therefore concerns the limits of incorporation. Locally prominent actors may have resources for public dissent and personal connections to rulers, while lacking an accepted collective role in decisions that constrain their interests. The Cape allows us to examine the wealth, officeholding, and network positions of people who made such demands. Whether their informal channels had failed, and whether that failure caused the petition’s public form, remain questions for process evidence beyond the comparisons reported here.

# Historical context: the Cape Colony, the VOC, and the 1779 petition

Established as a VOC refreshment station in 1652, the Cape remained Company territory until 1795. Its governor and Council of Policy exercised central authority through the Company hierarchy; burgher participation in justice and local administration remained subordinate (Baartman 2011). The station grew into a trade waypoint with an expanding agrarian slave economy (Fourie and von Fintel 2011; Fourie 2013).

The Cape Patriots asserted the political and economic rights of *Vrij Burghers*, settlers outside VOC employment who could own land, trade within Company limits, and participate in civic institutions. Their demands sought to constrain officials’ private commerce, widen trading opportunities, strengthen legal protections, and increase burgher authority within colonial government (Baartman 2011; Baartman 2012). These demands arose within a society whose official and burgher families were connected by marriage, commerce, and patronage. The resulting factions did not map neatly onto employment by the VOC (Baartman 2011; Baartman 2019).

One historiographical tradition treats burgher identity as early anti-imperial settler nationalism (Beyers 1967); revisionist work stresses transnational Dutch political connections (Baartman 2011; Baartman 2019). Burgher status was both a legal category and an idiom of civic rights (Baartman 2012).

At the same time, the Cape was a settler colony built on land appropriation and coerced labour. Settler prosperity depended on enslaved labour and on the gradual displacement, destruction, and incorporation of indigenous Khoe and San polities (Green 2022; Wolfe 2006). This tripartite structure – metropolitan officials, settlers, and subordinated indigenous and enslaved populations – shaped the distribution of resources and insecurity. Settlers’ claims to rights and representation existed alongside, and were enabled by, regimes of domination and exclusion. Our data capture the internal stratification of the settler household economy and the relationship between that stratification and public claim-making. We do not directly observe political agency among enslaved or indigenous actors, even though those groups shaped the colony’s political economy and security environment. We return to this limitation in the discussion.

# Data and methods

## Petition signatories

Our dependent variable is signing the 1779 Cape Patriots memorandum. We use Baartman’s list of all 404 signatories (Baartman 2011); the archival reference is Western Cape Archives and Record Service, C 2665, pp. 2–9. A signature records declared participation rather than private support. It combines willingness to sign with an opportunity to encounter the document, a distinction relevant to the network analysis. Four petitioners signed with an X, so the ability to write a name was not required.

## Tax censuses and the construction of production-asset value

Annual agricultural tax censuses (*opgaafrollen*) list household heads, spouses, children, enslaved and Khoesan workers, agricultural assets, and outputs (Fourie 2013; Fourie and von Fintel 2011).

Our linked census data cover Stellenbosch and Drakenstein, spanning the towns, commercial farms, and frontier homesteads of these rural districts. They do not include the Cape District or provide comparable coverage of Swellendam. Urban artisans, merchants, and administrators are consequently outside this wealth comparison. The results describe the observed rural agrarian settler population, not the colony as a whole.

*Matching signatories to the census.* We manually matched petition names to census names using conservative criteria that avoid false positives at the cost of missed matches: when multiple individuals with identical first and last names appeared in the tax censuses, we excluded the match. We matched 165 of 404 signatories (41%). Match rates by signature decile suggest recruitment began in Cape Town, whose signers are outside our census frame, and later moved into the rural districts (Online Appendix Table S1). High rates in deciles six to nine are consistent with stronger coverage of rural petitioners, while the final decline may reflect Swellendam or late-added Cape signers. Cape Town VOC retirees and merchants are a specific gap. Section [6.7](#sec:robustness) asks what those signers would need to look like for the rural wealth finding to reverse.

For the common 1779 wealth sample we retain all households recording a positive quantity on at least one of the five priced asset series. This yields $N = 1{,}139$ households, including $153$ petitioners and $986$ non-petitioners. The filter removes $236$ households, $12$ of them petitioners, that report none of the five: some are sparse records, but others are genuine households whose assets lay outside the priced series, 29 of them recording enslaved workers. We note this because it is relevant to the sensitivity exercise in §[6.7](#sec:robustness).

The censuses reduce enslaved and indigenous people to household counts and may understate self-reported wealth (Fourie 2013; Fourie and von Fintel 2011). They remain the most comprehensive source on the eighteenth-century settler economy.

We value cattle, sheep, horses, wheat, and wine using decadal prices from Fourie and von Fintel (2011). This production-asset measure excludes land, vines, brandy, wagons, firearms, credit, and enslaved-labour capital (Fourie 2013); slave counts enter separately.

## Heemraden and local officeholding

Petition titles identify current and former (*oud*) heemraden. We also use the appointment roll underlying Walters et al. (2026), constructed from Council of Policy minutes. The extract we use labels appointment cohorts from 1719 to 1783. It omits 1746–1749, the gap reported in the source study, so earlier service cannot be observed completely. Late-year appointments began service the next January for two-year terms.

Table [3](#tab:stellenbosch-heemraden) asks what share of Stellenbosch–Drakenstein officeholders signed. We retain the four district appointments in each annual cohort, excluding two Swellendam positions, and use a hand-constructed crosswalk to title-identified petitioners. Eleven of the twelve titled petitioners are linked to the table’s 1762–1778 appointment cohorts. We consolidate spelling variants while preserving distinguishing suffixes. One further proposed link, between François du Toit de Oude in 1779 and a 1722 appointee, lacks independent biographical confirmation and is not used in this table.

Table [4](#tab:heemraad_quartile) instead compares prior service among matched households. Petitioner titles provide direct indicators; non-petitioner heads are linked to the roll, preserving patronymics and distinguishing suffixes. Of 21 namesake collisions, panel tenure resolves eight by service-year overlap and assigns eleven to the longest-tenured candidate; two remain unresolved and are dropped. Online Appendix Section E reports alternatives. This linkage is separate from Table [3](#tab:stellenbosch-heemraden)’s direct hand match.

## Auction networks

Deceased-estate auction records (*vendurollen*, MOOC Series 10, volumes 1–36) capture public, often multi-day gatherings. Because households attended for commerce and sociability, auctions proxy the broad weak ties of settler society (Mitchell 2009; Fourie et al. 2025).

We reconstruct a two-mode network linking people to purchases at 1775–1779 auctions. For the descriptive comparisons and regressions, we compute centrality on the simple bipartite graph, collapsing repeated person–auction edges. The supplementary permutation tests use a person projection and therefore a different brokerage measure. Fuzzy name linkage was verified by hand for petitioners. The sample contains $2{,}225$ people ($172$ petitioners and $2{,}053$ non-petitioners). Degree counts distinct auctions at which a person purchased; closeness is the reciprocal of mean distance to other nodes; betweenness counts shortest paths passing through a node (Freeman 1978; Wasserman and Faust 1994). Figure [2](#fig:5) uses a Fruchterman–Reingold layout (Fruchterman and Reingold 1991, seed $= 1779$).

We observe participants only when they won a bid, so purchasing power filters the network. We therefore distinguish auction activity (degree) from brokerage after accounting for degree.

## Analytical strategy

We compare petitioners and non-petitioners in economic position, officeholding, and network position, testing mean and distributional differences. Annual census panels also test whether future petitioners experienced different pre-1779 wealth trends.

# Results

## Petitioners’ wealth

How did petitioners’ wealth compare with that of non-petitioners? Petitioners were markedly wealthier overall, concentrated in the top wealth quartile and in the wine sector, a marker of the established rural economy, and under-represented in sheep farming, a marker of peripheral pastoral agriculture.

Figure [1](#fig:1) visualises petitioners’ wealth relative to the rest of the observed rural population, showing the joint distribution of production-asset value and slaveholding. Petitioners cluster among wealthier households, with the three wealthiest petitioners ranking among the ten wealthiest households in the two districts.

*[Figure not reproduced here — see JS_JF_ProtestFromAbove_v2.pdf]*
Production-asset value and slaveholding, 1779. Petitioners are filled triangles ($n = 153$); non-petitioners are open circles ($n = 986$). Both axes are $\log(x+1)$ so that households holding none of an asset remain on the panel; tick labels give the underlying raw values. Points are jittered slightly to separate ties. The three labelled households are the wealthiest petitioners.

Table [1](#tab:1) compares mean asset holdings between petitioners and non-petitioners in the common 1779 wealth sample ($N = 1{,}139$; petitioners $= 153$). Petitioners held more horses, wheat, wine, slaves, and settler household members than non-petitioners. The gaps are largest where the established rural economy was concentrated: petitioners held roughly four times the wine wealth and nearly twice the enslaved workforce of non-petitioners. Cattle differences are modest, and the mean sheep difference is not statistically distinguishable from zero.

|  | Pet. mean | Non-pet. mean | Diff. | Pet. $n$ | Non-pet. $n$ | $p$ |  |
|:---|---:|---:|---:|---:|---:|:---|:---|
| Cattle (Rds) | 109.5 | 88.3 | 21.2 | 153 | 986 | 0.06 |  |
| Sheep (Rds) | 64.5 | 74.2 | $-$9.8 | 153 | 986 | 0.31 |  |
| Horses (Rds) | 116.0 | 59.1 | 57.0 | 153 | 986 | $<$.001\*\*\* |  |
| Wheat (Rds) | 32.0 | 14.0 | 18.0 | 153 | 986 | 0.01\*\* |  |
| Wine (Rds) | 194.5 | 46.4 | 148.1 | 153 | 986 | $<$.001\*\*\* |  |
| Slaves (no.) | 7.25 | 3.84 | 3.41 | 153 | 986 | $<$.001\*\*\* |  |
| Settlers (no.) | 4.33 | 3.46 | 0.87 | 153 | 986 | $<$.001\*\*\* |  |

Wealth and household characteristics: petitioners vs. non-petitioners, 1779.

*Note:* Values in *rijksdaalders* (Rds). Two-sided $t$-tests. \*\*\* $p < 0.001$, \*\* $p < 0.01$, \* $p < 0.05$.

Distributional and rank tests point to the same conclusion (Online Appendix Section A): petitioners tend to rank higher on the overall index and on every asset except sheep. The rank advantages are greatest for wine, total wealth, and enslaved-worker counts; all seven non-sheep comparisons survive false-discovery correction. The empirical distributions cross, so this is not a claim of first-order stochastic dominance. Online Appendix Figure S1 visualises the sectoral differences. The wine gap is especially stark: $43$ per cent of petitioners held wine compared with $11$ per cent of non-petitioners, and the difference persists at higher thresholds.

Petitioners therefore tended to occupy higher positions in the asset distribution, especially for the commodities that defined the wine-and-slave elite. They were disproportionately wealthy and, among the wealthy, concentrated among established farmers at the centre of the Cape’s rural economy.

## Concentration at the top and within-elite differentiation

The mean differences in Table [1](#tab:1) are not driven only by a long left tail of poorer non-petitioners. Panel A of Table [2](#tab:2) shows a clear concentration of petitioners among the wealthier households: petitioning rates are 8.4 per cent in the bottom quartile of rural wealth, 8.0 per cent in the second quartile, 12.0 per cent in the third, and 25.3 per cent in the top quartile. This pattern offers little support for a predominantly poor rural constituency or for equally strong recruitment across wealth groups. Settlers throughout the distribution did sign, however, and composition alone cannot establish whether their rhetoric or identities were populist or nationalist. Petitioning was nonetheless concentrated in the upper tail of observable rural wealth.

| *Panel A. Petitioning by wealth quartile* |  |  |  |  |
|:---|---:|---:|---:|---:|
| Wealth quartile | $n$ | Petitioners | Petition rate | Mean wealth (Rds) |
| Q1 | 371 | 31 | 8.4% | 11.2 |
| Q2 | 200 | 16 | 8.0% | 101.9 |
| Q3 | 283 | 34 | 12.0% | 276.5 |
| Q4 | 285 | 72 | 25.3% | 892.1 |
| *Panel B. Petitioners vs. non-petitioners within top wealth quartile* |  |  |  |  |
| Variable | Petitioners | Non-petitioners | Difference | $p$ |
| Total wealth (Rds) | 930.7 | 879.0 | 51.8 | 0.52 |
| Wine wealth (Rds) | 376.7 | 190.2 | 186.5 | 0.002\*\* |
| Wheat wealth (Rds) | 59.4 | 50.0 | 9.3 | 0.52 |
| Sheep wealth (Rds) | 103.2 | 203.7 | $-$100.5 | $<$.001\*\*\* |
| Slaves (no.) | 13.1 | 12.6 | 0.5 | 0.74 |

Upper-tail concentration and within-elite differentiation.

*Note:* Panel A uses the common 1779 wealth sample ($N = 1{,}139$). Quantile cut-points retain ties together, so group sizes differ. Panel B restricts the sample to the top wealth quartile ($n = 285$) and reports group means, with the Difference column the mean difference. \*\*\* $p < 0.001$, \*\* $p < 0.01$, \* $p < 0.05$.

Panel B of Table [2](#tab:2) focuses on the wealthiest quartile. Among elite households, which elites signed? Within the top quartile, the mean total-wealth difference between petitioners and non-petitioners is small relative to its uncertainty. What distinguished them was their domain of economic activity: elite petitioners held roughly twice the wine wealth and half the sheep wealth of their non-petitioning peers, while the wheat and enslaved-worker differences are imprecisely estimated. Petitioners were thus disproportionately drawn from commercially oriented wine households and less often from sheep-oriented pastoral elites.

This compositional contrast locates the petitioning group within the commercial agrarian economy. Wine production and enslaved labour were central to the wealth of established farming households (Fourie and von Fintel 2011; Green 2022). The contrast with sheep holdings suggests that recruitment differed among wealthy producers, not simply between rich and poor. This pattern is consistent with an interest in the trading opportunities sought by the petition. Our sectoral measures do not, however, identify particular lineages or establish how Company restrictions affected wine producers relative to pastoralists.

## Wealth trajectories: testing the deprivation mechanism

Petitioners were wealthier, but could they have been a prosperous stratum experiencing relative decline? We examine pre-1779 trajectories using the panel structure of the *opgaafrollen*. Online Appendix Figure S2 plots mean production-asset values for households linked to the common 1779 sample. Future petitioners remain wealthier on average. These are changing samples of surviving, linked households, however, and neither the levels nor the visual comparison settles the question of relative decline.

A within-household model for 1764–1778 estimates the petitioner-specific trend at $-1.59$ Rds per year, with a standard error of $7.27$ (Online Appendix Section B). The estimate does not detect a differential trend, but its uncertainty permits economically meaningful relative decline as well as improvement. The year-specific estimates compare each year with 1778, not with the preceding year; several suggest a smaller petitioner advantage in earlier years, without establishing a sustained upward path. The panel evidence therefore gives no clear indication of relative decline. It cannot rule out a deprivation mechanism, particularly one concerning expectations, unmeasured wealth, or petitioners outside the rural sample.

## Political office: testing access to institutional power

We turn to officeholding to ask whether petitioners were marginal to colonial institutions or recognised within them. Forty-seven of the 404 signatories (11.6%) list civic, militia, or church titles, spread fairly evenly across those institutional domains (Online Appendix Table S2).

We now focus on the office of heemraad (pl. heemraden), whose holders deployed VOC power in settler communities. Heemraden were elite settler officers who exercised local administrative and lower judicial authority in the rural districts. They were unsalaried officials serving under a salaried landdrost (magistrate), rather than VOC employees. Drawn from the district’s “most notable settlers”, they linked central authority and rural settler society. Appointment was for two years and staggered, with half of the eight seats renewed each year. Selection was “strictly controlled”: the local board compiled a list from which the central government chose appointees (Dooling 1991, 7–9). The board nominated twice the number required (Walters et al. 2026). Heemraden projected VOC administrative and judicial power into the countryside but had no authority over colony-level decisions. The office therefore exemplifies the colonial incorporation of locally recognised leaders beneath a formal colonial official.

Twelve petitioners identified themselves as current or former (*oud*) heemraden. The crosswalk to the Stellenbosch–Drakenstein roll links four to the eight men serving in 1779, seven to terms beginning in 1775–1779, and eleven to terms beginning in 1763–1779. These links locate petitioners among recognised local notables who had administered VOC power at district level. The proposed 1722 match for the twelfth titled signer remains unresolved; the three reported cohorts do not depend on it.

|                       |             |     |     |     |
|:----------------------|------------:|----:|----:|----:|
| Service status        |             |     |     |     |
| officeholders         |             |     |     |     |
| slots                 | Petitioners |     |     |     |
| petitioning           |             |     |     |     |
| Current in 1779       |           8 |   8 |   4 | 50% |
| Terms began 1775–1779 |          17 |  20 |   7 | 41% |
| Terms began 1763–1779 |          39 |  68 |  11 | 28% |

Petition participation among Stellenbosch–Drakenstein heemraden.

*Note:* Only the four Stellenbosch–Drakenstein appointments in each annual cohort are used, not the two Swellendam ones. Two-year terms mean the 1777 and 1778 cohorts comprise the eight officeholders serving in 1779. The longer rows count terms beginning in the stated years: appointment cohorts 1774–1778 and 1762–1778. They exclude terms beginning earlier that still overlapped the first year. Unique officeholders consolidate hand-verified spelling variants; unqualified common names are not merged with men explicitly marked *d’oude*. Shares use unique officeholders as the denominator. The full crosswalk is in the replication package.

Table [3](#tab:stellenbosch-heemraden) reverses the question: rather than asking how many petitioners were heemraden, it asks how many heemraden joined the petition. Half of the serving officeholders signed; participation was also substantial among men whose terms had begun earlier. The longer windows include repeated appointments but count each identified officeholder once. They are historical cohorts rather than populations known to be alive in 1779, so their participation shares should not be read as recruitment rates among surviving former officeholders.

| Wealth | Sample |  | Prior heemraad service % |  | Diff. | $p$ |
|:---|---:|---:|---:|---:|---:|:---|
| 2-3(lr)4-5 quartile | $n$ | Pet. | Pet.  | Non-pet.  | (pp) | (Fisher) |
| Q1 | 370 | 31 | 0.0 | 1.5 | $-$1.5 | 1.00 |
| Q2 | 200 | 16 | 0.0 | 2.2 | $-$2.2 | 1.00 |
| Q3 | 283 | 34 | 5.9 | 2.8 | $+$3.1 | .30 |
| Q4 | 282 | 72 | 12.5 | 3.8 | $+$8.7 | .017\* |
| All | 1,135 | 153 | 7.2 | 2.4 | $+$4.7 | .004\*\* |

Prior heemraad service by wealth quartile and petition status.

*Note:* Common 1779 wealth sample (after wealth filter; $N = 1{,}135$ rows where the heemraad indicator is non-missing). Petitioner-side heemraad indicator is from the petition name title (heemraad or oud heemraad). Non-petitioner-side indicator is from name match against the heemraden list, using year and household head data to disambiguate populous lineages with duplicate names. Fisher exact two-sided $p$. The four non-petitioner rows in the two unresolvable collision groups are excluded as missing on the indicator. \*\*\* $p < 0.001$, \*\* $p < 0.01$, \* $p < 0.05$.

Table [4](#tab:heemraad_quartile) compares heemraad service across petitioners and non-petitioners in our matched households, by wealth quartile. This is a different estimand from Table [3](#tab:stellenbosch-heemraden): it asks whether matched petitioner households were more likely than matched non-petitioner households to contain a prior officeholder. To label non-petitioners, we match household heads in the common wealth sample to the list of prior heemraden. Prior officeholding increases with wealth: among non-petitioners, prior service rises steadily across the wealth quartiles, and all eleven matched petitioner heemraden are in the top two quartiles. The gap is concentrated at the top: one in eight top-quartile petitioners had served as heemraad, against roughly one in twenty-six top-quartile non-petitioners. In the full rural sample, 7.2 per cent of petitioners and 2.4 per cent of non-petitioners had served – the threefold difference reported in the introduction. Heemraden are thus over-represented among petitioners, especially at the top of the wealth distribution. This comparison rests on a linkage judgement that the wealth and network results do not require, because petitioners declare the office themselves while for non-petitioners it must be inferred from names. The differential is positive under every rule that uses the panel to resolve namesake collisions, but its magnitude is sensitive to how ambiguous namesakes are treated, and two rules that discard or adversely assign them turn it slightly negative. Online Appendix Section E reports the alternatives, the accuracy of the name match against the roster, and the limits of the tenure-based continuity rule.

Taken together, these analyses show that petitioners were not excluded from all political power: they were often locally prominent. The findings are consistent with an institutional asymmetry between central Company authority and locally generated power. Heemraden are an archetypal case of colonial power projection. They were empowered local notables who brought VOC administration to the hinterland, yet remained subordinate to the VOC landdrost. In our framing, they were local elite allies recruited to the VOC governing project but subordinate in the making of central policy. They were also well placed to fill any new seats in the councils that the petition demanded. The combination of local recognition, administrative experience, and limited access to central authority is consistent with the capacity to organise a public institutional challenge without resembling either a deprived mass movement or a broad settler-nationalist mobilisation.

## Network position of petitioners

Finally, we turn to the social position of petitioners. Using auction networks as a proxy for the weak ties of Cape settler society, we find that petitioners were more socially central figures than non-petitioners, indicating greater embeddedness in observed purchasing relationships. These relationships provide evidence of potential social reach rather than a complete map of settler sociability.

Table [5](#tab:3) compares centrality in the auction purchase network. Petitioners purchased at more distinct auctions, had higher closeness on average, and had roughly twice the mean betweenness of non-petitioners; rank tests also distinguish the groups. Betweenness is strongly right-skewed and often zero. At the median, non-petitioners purchased at one auction and had zero betweenness, while petitioners purchased at two and occupied positions through which some shortest paths passed.

|  | Mean |  | Median |  | $p$ |  |
|:---|---:|---:|---:|---:|---:|---:|
| 2-3(lr)4-5(lr)6-7 | Non-pet. | Pet. | Non-pet. | Pet. | $t$-test | M–W |
| Degree | 2.03 | 2.74 | 1 | 2 | .001\*\* | $<$.001\*\*\* |
| Closeness | 0.234 | 0.244 | 0.238 | 0.244 | $<$.001\*\*\* | $<$.001\*\*\* |
| Betweenness | 1,327 | 2,754 | 0 | 171 | .04\* | $<$.001\*\*\* |

Network centrality: petitioners vs. non-petitioners.

*Note:* Auction purchase network, 1775–1779, largest connected component. Non-petitioners $n = 2{,}053$; petitioners $n = 172$. M–W is the Mann–Whitney $U$ test. Test statistics and Cliff’s $\Delta$ are reported in Online Appendix Table S4. \*\*\* $p < 0.001$, \*\* $p < 0.01$, \* $p < 0.05$.

Figure [2](#fig:5) visualises the two-mode auction network. Each node represents either a person (petitioner or non-petitioner) or an auction event; edges connect individuals to the auctions at which they purchased goods. Petitioners (filled nodes) are visibly dispersed across the network rather than clustered in a single region of it, and many sit between otherwise separate auction clusters. Shared purchasing connects people across auction events; the plotted distances reflect a graph layout rather than physical proximity. They often bridged to more peripheral auctions, though not to the most isolated ones.

*[Figure not reproduced here — see JS_JF_ProtestFromAbove_v2.pdf]*
Two-mode auction network, 1775–1779. Filled circles are petitioners ($n = 172$); open circles are non-petitioners ($n = 2{,}053$); squares are auction events ($n = 95$). Edges connect individuals to auctions at which they purchased goods. Node size increases with degree under a common, capped rule for all persons. Fruchterman–Reingold layout, seed $= 1779$; the one disconnected person–auction pair is omitted.

Petitioners were disproportionately central figures in this network. A larger share purchased at more than one auction and occupied the top quintile of betweenness centrality. Online Appendix Figure S3 shows the density distributions of betweenness centrality and the rightward shift of the petitioner distribution.

Together, the statistics and visualisation show that petitioners were distributed across the purchase network and often connected different auction groups. Network paths do not show that information or the petition actually travelled along them. Historians’ accounts of auctions as social gatherings make such links plausible channels of contact (Mitchell 2009); our evidence establishes purchasing connections and potential brokerage.

## Brokerage and auction activity

How much of this network advantage reflects higher auction activity? People who purchase at more auctions have more opportunities to connect otherwise separate groups. We therefore examine betweenness together with degree, which counts distinct auctions with a recorded purchase. Degree adjusts for one dimension of activity; it is not a control for wealth or for all opportunities to encounter the petition.

Online Appendix Table S3 reports logistic regressions of signing on degree and betweenness. Each is associated with signing on its own. In the joint model, betweenness retains a positive association while degree is imprecisely estimated. Firth penalisation (Firth 1993) produces similar estimates. The residualised specification in Online Appendix Section C1 expresses the same joint model in different coordinates, rather than providing an independent robustness test. These regressions show an association with brokerage conditional on recorded auction activity; they do not establish that activity has no separate role, and their conventional standard errors do not account for network dependence.

Permutation tests provide a complementary comparison using brokerage on the person projection of the auction network (Online Appendix Section C2). The observed gap exceeds that in all 2,000 random reallocations of petition labels. Rewiring ties while preserving each person’s and auction’s degree produces a mean log-betweenness gap of $1.370$, against the observed $1.443$. Thus most of the gap is reproduced by this degree-preserving benchmark, with a smaller residual advantage for petitioners. The $500$ rewired networks place that residual in the upper tail ($p = .006$). This is conditional evidence under a specified randomisation, not a decomposition of the causes of signing; its interpretation also depends on the edge-swap sampler adequately exploring the permitted networks.

The firmest conclusion is that petitioners were active purchasers with broad connections in the recorded auction economy. A smaller component of their brokerage advantage remains relative to the degree-preserving benchmark. Wealth, residence, family connections, and the circulation of the document could still explain these associations. The limited linkage between census and network records prevents a persuasive joint adjustment for those factors.

A caveat applies. A signature reflects both the willingness to sign and the chance of being reached as the document circulated: brokers may have organised the petition, or it may have reached them along the same ties that give them high betweenness. We read bridging position in this combined sense, as the relational structure along which the petition circulated, and do not claim to identify organisers.

## Signatories outside the wealth sample

Only $153$ signatories enter the positive-asset wealth sample. Another $12$ have recorded zero production assets and $239$ are unmatched. Online Appendix Section D retains those 12 zeros and varies the assets of the unmatched signers. Their mean would need to be about $146$ Rds for the combined petitioner mean to equal the observed positive-asset non-petitioner mean of $282$ Rds. Some plausible assignments preserve a premium; others reverse it. Because agricultural assets may poorly measure the standing of Cape Town merchants, artisans, and Company retirees, these scenarios cannot establish the wealth rank of the whole movement. The rural comparison also survives the inclusion of the observed zero-asset households, which addresses the sample filter directly.

# Discussion

Tax-census wealth, officeholding, and auction networks locate the observed petitioners among relatively prominent settlers. They were concentrated in the upper wealth quartile and wine economy; four of eight serving Stellenbosch–Drakenstein heemraden signed. They were also unusually active and central in auction purchasing, with a smaller brokerage advantage remaining under a degree-preserving network benchmark. The wealth panel does not establish an absence of relative decline. Neither do these records measure political motives directly. Their contribution is to identify the resources and institutional experience present among those who signed.

## Elite decoupling and historical political sociology

The public political challenge of those who were disproportionately wealthy, socially connected, and likely to be local officials points to an elite cleavage rather than a non-elite challenger group. This public intra-elite challenge is puzzling twice over: in a small settler colony, a compact elite should be cross-connected; and where political-institutional power overlaps with wealth and prestige, we might expect conflict to be absorbed through elite negotiation and settlement (Burton and Higley 1987; Higley 2010).

We propose *elite decoupling* as an interpretation of this public challenge. Imperial authorities retained control over central policy while depending on locally embedded allies to administer colonial society (Robinson 1972; Walters et al. 2026). That dependence helped local elites accumulate wealth, standing, and organisational capacity. The resulting asymmetry could give locally powerful actors both a stake in the governing order and reasons to demand a greater share of authority. Our compositional evidence establishes the first part of this account more directly than the second: it identifies prominent participants, while the explanation of public contestation draws on the institutional setting and the petition’s demands.

The heemraden illustrate that asymmetry. Settlers with wealth, public standing, and experience of administering rural society challenged the Company’s authority. Four of eight serving district officeholders signed, as did eleven of thirty-nine men whose terms began in 1763–1779. Their local responsibilities did not confer an equal role in central policy. Nor did local office leave them wholly outside colonial institutions. It placed them within the governing order, with experience of both its operation and the limits of their authority.

Baartman’s account supplies a necessary qualification (Baartman 2011; Baartman 2019). He describes factions linked by kinship, patronage, and commercial interest that included both burghers and Company officials. We therefore do not treat the Patriots as the expression of a socially isolated settler bloc confronting an internally unified Company elite. Institutional inequality coexisted with social overlap. Local incorporation, in this reading, could give prominent settlers both experience of colonial government and resources to seek a greater share of its authority.

This interpretation places the Cape alongside accounts of elite competition without treating those accounts as versions of the same argument (Lachmann 2000; Bearman 1993; Padgett and Ansell 1993; Ansell and Samuels 2014). The relevant institutional division need not originate in the breakdown of a previously unified ruling group. It can instead rest on the unequal allocation of authority across local and central institutions. Adams’s (1996) analysis concerns a different relation, between Company principals and their overseas agents; our focus is on locally prominent settlers’ position within a Company-governed society. These relations could intersect, and the evidence does not establish a single cleavage governing all factional alignments.

This mechanism for intra-elite conflict should not be understood as an automatic consequence of the separation of wealth from office. Public elite challenge becomes more likely when several conditions coincide. A local elite must possess sufficient independent resources to withstand retaliation; it must be socially embedded enough to coordinate trans-local claim-making; and its exclusion must concern institutions that materially constrain its interests and status. At the same time, ordinary channels of incorporation, rotation, or backstage bargaining must remain limited. Where local elites can readily purchase office, enter representative institutions, or secure durable patronage, conflict may be absorbed rather than made public. Where they lack resources or cross-community connections, exclusion may instead produce acquiescence, fragmented resistance, or exit. The Cape evidence documents resources, local administrative standing, and broad purchasing connections alongside subordinate authority. It does not establish that informal bargaining channels had been exhausted.

The comparison with New Netherland suggests where further work could test this interpretation. Both petitions involved locally prominent settlers asserting institutional standing under chartered-company government (Middleton 2010). Establishing whether restricted incorporation caused public challenge would require tracing failed negotiations, changing alliances, and the alternatives available to participants. Our cross-sectional comparisons cannot supply that process evidence or explain why the Cape petition emerged in 1779 rather than in another year.

Finally, identifying the Cape Patriots as a challenger elite should not be confused with treating their challenge as democratising or emancipatory. The political capacities of Cape settler elites rested on appropriated land, enslaved labour, and the subordination of Khoe and San populations. Their language of burgher rights contested a local elite’s exclusion from the central institutions of Company rule without seeking to disrupt the domination of populations whose labour and dispossession sustained settler prosperity. The movement therefore reveals how opposition to imperial domination can coexist with, and be enabled by, colonial domination of others (Wolfe 2006; Johnston and Lawson 2005). Elite decoupling highlights how challenges to rulers can emerge from intra-elite cleavages that preserve elite–non-elite power structures. In the colonial case, it clarifies how struggles within the dominant colonial order could generate public political innovation while leaving its foundational exclusions intact.

## Scope conditions

Three gaps limit our findings. First, the wealth comparison concerns linked rural households in Stellenbosch and Drakenstein. It excludes Cape District households and lacks comparable Swellendam coverage. The matched sample is concentrated in the second half of the signatory list. The sensitivity exercise makes the importance of the unobserved signers explicit, but cannot substitute for a full account of the movement.

Second, the social action of indigenous and enslaved people remains largely invisible in our tax, office, and auction records. The Cape Patriot movement created a local political public and framed a conflict between settler burghers and VOC-appointed officials, yet this conflict rested on enslaved labour and the displacement of indigenous Khoe and San polities (Wolfe 2006; Veracini 2010; Johnston and Lawson 2005). Settler-colonial theory insists that this public was constituted through the structural exclusion of indigenous and enslaved political voices. The movement’s conflict with the VOC did not challenge settler elites’ domination of enslaved workers or their dependence on Company governors to mediate labour and land dispossession. Nonetheless, indigenous and enslaved peoples shaped the possibilities and limits of intra-elite conflict. A fuller account of why settler elites mounted their challenge at this moment would trace frontier violence, the threat of slave resistance, and the Company’s limited capacity to control force and property. We treat these wider conflicts, and their effects on different groups’ willingness to contest, as questions for further research.

Third, key measurement choices and linkage limits remain open to alternative interpretation. The 41 per cent census match rate, the winners-only structure of the auction network, and the production-asset measure’s omission of land, physical assets, and the capitalised value of enslaved labour all shape our analysis. The matched-household heemraad differential is also sensitive to how namesake collisions are resolved, as Online Appendix Section E shows. By contrast, the officeholder participation rates in Table [3](#tab:stellenbosch-heemraden) use direct title evidence and an explicit hand-constructed district crosswalk, whose underlying identities remain open to biographical verification. We hope these analytical and empirical choices prompt further research using alternative linkage judgements and measures.

# Conclusion

Among the rural households we observe, the Cape Patriots were disproportionately wealthy, concentrated in the upper quartile and wine economy, and prominent among district officeholders. Petitioners in the auction records were active purchasers with unusually broad connections. This composition is difficult to reconcile with a primarily impoverished rural constituency. It does not establish an absence of perceived deprivation or nationalist identification, and it cannot describe all signatories’ wealth.

Elite decoupling identifies the institutional tension these findings bring into view. Colonial rulers relied on local allies while retaining central authority, enabling some settlers to accumulate wealth, standing, and subordinate governing experience without incorporating them as equals. At the Cape, public challenge arose among people who had prospered within the colonial order and administered it locally. Social ties between officials and burghers crossed that institutional boundary without removing the unequal distribution of authority.

The contribution concerns the relationship between elite incorporation and public conflict. Local office and prosperity need not secure acceptance of the institutions that confer them. They may also supply resources and experience for demands to redistribute authority. Our evidence supports this interpretation of the Patriots’ social base; it does not identify the causal effect of institutional exclusion or explain the timing of mobilisation. Those questions require tracing the movement’s alliances and negotiations. What the signatory list does make clear is that a colonial political order could face a visible challenge from people who had benefited substantially from it.

# Data availability

The replication data and code supporting this article will be deposited in the JHPE Dataverse before publication, subject to the journal’s replication requirements. The underlying archival sources are the tax censuses (*opgaafrollen*) and deceased-estate auction records (*vendurollen*, MOOC Series 10) held at the Western Cape Archives and Record Service, Cape Town, and the 1779 petition signatory list (WCARS, C 2665, pp. 2–9).

# References

Adams, Julia. 1996. “Principals and Agents, Colonialists and Company Men: The Decay of Colonial Control in the Dutch East Indies.” *American Sociological Review* 61 (1): 12–28.

Ansell, Ben W., and David J. Samuels. 2014. *Inequality and Democratization: An Elite-Competition Approach*. Cambridge University Press.

Baartman, Teun. 2012. “Protest and Dutch Burgher Identity.” In *Cape Town Between East and West: Social Identities in a Dutch Colonial Town*, edited by Nigel Worden. Uitgeverij Verloren.

Baartman, Teun. 2019. *Cape Conflict: Protest and Political Alliances in a Dutch Settlement*. Leiden University Press.

Baartman, Teunis. 2011. “Fighting for the Spoils: Cape Burgerschap and Faction Disputes in Cape Town in the 1770s.” PhD thesis, University of Cape Town.

Bearman, Peter S. 1993. *Relations into Rhetorics: Local Elite Social Structure in Norfolk, England, 1540–1640*. Rutgers University Press.

Beyers, Coenraad. 1967. *Die Kaapse Patriotte Gedurende Die Laaste Kwart van Die Agtiende Eeu En Die Voortlewing van Hul Denkbeelde*. Van Schaik.

Broćić, Miloš. 2023. “Alienation and Activism.” *American Journal of Sociology* 128 (5): 1291–334.

Burton, Michael G., and John Higley. 1987. “Elite Settlements.” *American Sociological Review* 52 (3): 295–307. <https://doi.org/10.2307/2095351>.

Davies, James C. 1962. “Toward a Theory of Revolution.” *American Sociological Review* 27 (1): 5–19.

Dooling, Wayne. 1991. “Law and Community in a Slave Society: Stellenbosch District, c. 1760–1820.” MA thesis, University of Cape Town.

Einstein, Katherine Levine, and Maxwell Palmer. 2021. “Land of the Freeholder: How Property Rights Make Local Voting Rights.” *Journal of Historical Political Economy* 1 (4): 499–530.

Firth, David. 1993. “Bias Reduction of Maximum Likelihood Estimates.” *Biometrika* 80 (1): 27–38. <https://doi.org/10.2307/2336755>.

Fourie, Johan. 2013. “The Remarkable Wealth of the Dutch Cape Colony: Measurements from Eighteenth-Century Probate Inventories.” *The Economic History Review* 66 (2): 419–48.

Fourie, Johan, Helena Liebenberg, Jonathan Schoots, and Paul van der Linde. 2025. “Auction Records, Consumer Culture and Social Networks in the Cape Colony, 1701–1825.” *South African Journal of Cultural History* 39 (1): 27–49.

Fourie, Johan, and Dieter von Fintel. 2011. “A History with Evidence: Income Inequality in the Dutch Cape Colony.” *Economic History of Developing Regions* 26 (1): 16–48.

Freeman, Linton C. 1978. “Centrality in Social Networks: Conceptual Clarification.” *Social Networks* 1 (3): 215–39.

Fruchterman, Thomas M. J., and Edward M. Reingold. 1991. “Graph Drawing by Force-Directed Placement.” *Software: Practice and Experience* 21 (11): 1129–64.

Green, Erik. 2022. *Creating the Cape Colony: The Political Economy of Settler Colonization*. Bloomsbury Publishing.

Gurr, Ted Robert. 2011. *Why Men Rebel*. Paradigm Publishers.

Higley, John. 2010. “Elite Theory and Elites.” In *Handbook of Politics: State and Society in Global Perspective*, edited by Kevin T. Leicht and J. Craig Jenkins. Springer New York.

Jansen, Robert S. 2017. *Revolutionizing Repertoires: The Rise of Populist Mobilization in Peru*. University of Chicago Press.

Johnston, Anna, and Alan Lawson. 2005. “Settler Colonies.” In *A Companion to Postcolonial Studies*, edited by Henry Schwarz and Sangeeta Ray. Blackwell.

Khan, Shamus Rahman. 2012. “The Sociology of Elites.” *Annual Review of Sociology* 38 (1): 361–77.

Kornhauser, William. 1959/2013. *The Politics of Mass Society*. Routledge.

Lachmann, Richard. 2000. *Capitalists in Spite of Themselves: Elite Conflict and Economic Transitions in Early Modern Europe*. Oxford University Press.

Laclau, Ernesto. 2005. *On Populist Reason*. Verso.

Middleton, Simon. 2010. “Order and Authority in New Netherland: The 1653 Remonstrance and Early Settlement Politics.” *William and Mary Quarterly, 3d Ser.* 67 (1): 31–68.

Mills, C. Wright. 1956. *The Power Elite*. Oxford University Press.

Mitchell, Laura Jane. 2009. *Belongings: Property, Family, and Identity in Colonial South Africa (an Exploration of Frontiers, 1725–c. 1830)*. Columbia University Press.

Oddens, Joris. 2017. “The Greatest Right of Them All: The Debate on the Right to Petition in the Netherlands from the Dutch Republic to the Kingdom (c. 1750–1830).” *European History Quarterly* 47 (4): 634–56.

Padgett, John F., and Christopher K. Ansell. 1993. “Robust Action and the Rise of the Medici, 1400–1434.” *American Journal of Sociology* 98 (6): 1259–319.

Phillips, Andrew, and JC Sharman. 2020. “Company-States and the Creation of the Global International System.” *European Journal of International Relations* 26 (4): 1249–72. <https://doi.org/10.1177/1354066120928127>.

Puga, Diego, and Daniel Trefler. 2014. “International Trade and Institutional Change: Medieval Venice’s Response to Globalization.” *The Quarterly Journal of Economics* 129 (2): 753–821.

Robinson, Ronald. 1972. “Non-European Foundations of European Imperialism: Sketch for a Theory of Collaboration.” In *Studies in the Theory of Imperialism*, edited by Roger Owen and Bob Sutcliffe. Longman.

Rohr, Benjamin. 2025. “Elite Cohesion in the American Administrative State, 1898–1998.” *Social Science History* 49 (2): 505–32. <https://doi.org/10.1017/ssh.2025.16>.

Veracini, Lorenzo. 2010. *Settler Colonialism: A Theoretical Overview*. Palgrave Macmillan.

Walters, Leoné, Johan Fourie, and Robert Ross. 2026. “Colonial Elites in the Eighteenth-Century Cape Colony.” *Revista de Historia Industrial—Industrial History Review*. <https://doi.org/10.1344/rhi-ihr.48401>.

Wasserman, Stanley, and Katherine Faust. 1994. *Social Network Analysis: Methods and Applications*. Cambridge University Press.

Wolfe, Patrick. 2006. “Settler Colonialism and the Elimination of the Native.” *Journal of Genocide Research* 8 (4): 387–409.

Zaret, David. 1996. “Petitions and the ‘Invention’ of Public Opinion in the English Revolution.” *American Journal of Sociology* 101 (6): 1497–555.

Zaret, David. 2019. “Petition-and-Response and Liminal Petitioning in Comparative/Historical Perspective.” *Social Science History* 43 (3): 431–51.

[^1]: Department of Sociology, University of the Witwatersrand. Email: <jonathan.schoots@wits.ac.za>.

[^2]: Department of Economics, Stellenbosch University. Email: <johanf@sun.ac.za>.

[^3]: We thank Anne McCants, Bronwen Everill, Erik Green, Benjamin Rohr, Robert Ross, Joris van der Tol, Dieter von Fintel, Leoné Walters, and participants at the African Economic History Network meetings (Gran Canaria) and Social Science History Conference (Chicago) for valuable comments on earlier drafts, and Riksbankens Jubileumsfond (M20-0041) for financial support. This paper was created with the help of Anthropic’s Claude Code and OpenAI’s Codex. Cite this paper as: Schoots, Jonathan, and Johan Fourie. 2026. “Protest from Above: Elite Decoupling and Public Petitioning in the VOC Cape Colony.” Working Paper, Department of Economics, Stellenbosch University.