In 1827 the Whig historian Henry Hallam described the revolution of 1688 as ‘the triumph of those principles which, in the language of the present day, are denominated liberal or constitutional’ (quoted in ^{Fawcett, 2014}, p.8). The view of the ‘Glorious Revolution’ as a liberal revolution endured in British political discourse until the late twentieth century. This understanding of 1688 as a founding moment in the history of liberalism is encapsulated in Margaret Thatcher’s speech in a Commons debate on the tercentenary of the revolution. The revolution, the then Prime Minister urged had

established the tradition that political change should be sought and achieved through Parliament. It was this which saved us from the violent revolutions which shook our continental neighbours and made the revolution of 1688 the first step on the road, which, through successive Reform Acts, led to the establishment of universal suffrage and full parliamentary democracy. (^{Vallance, 2006}, p. 1)

Thatcher’s distillation of this triumphal Whig narrative of constitutional progress, however, was already being hotly contested in scholarly circles. The most recent accounts of 1688 by Tim Harris, Steven Pincus and others have asserted that, indeed, this was a revolution but one which shared more with its bloody mid-century predecessor or with the French revolution than it did with this account of stately, peaceful political progress. (^{Harris, 2006}; ^{Pincus, 2009}).

Piece by piece, the elements of the old Whig narrative have been dismantled. A range of scholars, but most notably Pincus, have demonstrated that it was anything but peaceful, unleashing war on land and sea in the dynastic contest between William III and James II. Nor was revolutionary change effected solely through Parliamentary initiative. Public opinion and crowd activity exerted a significant influence on the course of the revolution, not least the anti-Catholic rioting which encouraged James and his family to flee to the comparative safety of the continent. The principles associated with the revolution - toleration, liberty and freedom of the press - now appear far less a tangible consequence of the events of 1688. The statutory toleration for some Protestant groups achieved in 1689 was far less generous than that proposed by James II before the revolution. Rather it was anti-Catholicism itself which was a key driver of the revolution (^{Sowerby, 2013}). ‘Anti-popery’ remained a powerful force after 1688 as Elizabethan and Stuart penal laws were reinforced in the 1690s with further penalties against English Catholics (^{Glickman, 2009}). Instead of ushering in a more secular age, historians have instead argued that the revolution was part of England’s long reformation.

The decades immediately following the revolution would be marked by public campaigns for moral regulation and with the passage of new laws against blasphemy. This remained an intensely religious age, in which doctrinal controversy spilled out into furious debates in the press. As Alex Barber’s work has shown, these disputes test the claim that the revolution and the subsequent lapsing of the licensing act in 1695 established the beginnings of a free press. Rather, the mechanics of censorship shifted from pre to post-publication while the principle of press freedom itself remained little valued (^{Barber, 2013}). The revolution had certainly been cast in Williamite propaganda as a struggle to defend English liberties but the consequences of the revolution could also be seen in the expansion of unfreedom. As recent work has demonstrated, the assault on the monopoly of the Royal African Company that followed the revolution enabled a massive increase in Britain’s participation in the slave trade: free trade outweighed the freedom of persons (^{Pettigrew, 2013}). In England too, as John Marshall has recently argued, members of Locke’s circle endorsed forms of unfreedom as necessary measures to deal with the ‘idle poor’: the ‘Glorious Revolution’ here represents the birth of the workhouse, not parliamentary democracy (^{Marshall, 2019}).

Consequently, little appears to be left of the revolution of 1688 as the triumph of liberal ideals. Indeed, a recent major interpretation of liberalism has urged that, viewed as a ‘practice of politics’, there was no meaningful liberalism before the nineteenth century (^{Fawcett, 2014}, p. 22-23). In its place we can broadly see two readings of the revolution solidifying. One is of 1688 less as a standalone event and more as part of wider processes and crises, whether as a further element of England’s long reformation or whether as the last of a series of ruptures prompted by the threat of Popery and arbitrary government (^{Knights, 2015}). The second is of the revolution as historical watershed and harbinger of modernity but here the underpinning theoretical framework is Marxian rather than liberal. In this reading 1688 is modern precisely because it was everything the liberal account said it was not: popular, violent, disruptive, even radical. According to Steven Pincus, this was a revolution in which political economy was more important than anti-popery, and economic interests more forceful drivers of historical change than confessional divisions or dynastic allegiances (^{Pincus, 2009}).

We are left, then, with the Revolution of 1688 as either the end of an old world or the beginning of a new one. Reinhart Koselleck identified the notion of temporal rupture as critical to a modern understanding of revolution and the opening up of a new ‘horizon of expectation’ (Koselleck, 2004, p. 43-57, quote at p. 49). Koselleck located this moment of chronological caesura in the French revolution, the intellectual preparation for which was undertaken in the work of French philosophes. As has been noted elsewhere, however, there is a paradox here that the writings of these philosophes were in part inspired by a backward-looking glance towards the revolution of 1688 in England. The imaginative leap that Koselleck identifies these writers as making from an earlier understanding of revolution as the cyclical violence of civil war to a transformative revolution was enabled by accepting the myth of the ‘Glorious Revolution’ as peaceful and progressive (^{Vallance, 2019a}). As Tim Harris has noted, Koselleck’s account of the transformation of ‘revolution’ as a concept is also chronologically flawed: we have good evidence that seventeenth-century observers understood both the mid-century revolution and 1688 as moments of political rupture in which a new political ‘fabric’ was woven (^{Harris, 2015}).

In the rest of this article, it will be suggested that, rather than thinking of 1688 as either a moment of closure or rupture, we might identify it with another of Koselleck’s concepts - ‘Sattelzeit’, literally ‘saddle-time’, a transitional period between two epochs (Koselleck, 2016, p. 34). Instead of representing the end of the ‘long seventeenth century’ or the beginning of the ‘long eighteenth century’, 1688 represents a moment in which multiple understandings of time and history co-habited. In an influential article, Helge Jordheim has argued that Koselleck’s work, in fact, eschews simplistic periodization and instead argues for the existence of multiple temporalities in the same moment. At a given time, multiple understandings could co-exist. ‘Democracy’ could be understood, therefore, as past (its manifestation in the ancient world), present (a subject of contemporary political contestation) and future (an end point on an imagined trajectory of societal development) (^{Jordheim, 2012}).

In this reading of Koselleck’s idea of temporality, the Sattelzeit is not a ‘transition’ in the sense that we might equate with artistic transitions in film or in music - one age does not simply dissolve into another. Instead, as Gabriel Motzkin has suggested, contemporaries exist in state of ‘partial awareness’ in relation to historical transition: 1688 could be understood both as a restorative revolution and as an act of ‘regime change’ (^{Motzkin, 2005}). Similarly, in his study of understandings of time during the revolution of 1688, Tony Claydon has argued that the period is characterized by ‘hybridity’ rather than an overtly ‘modern’ sense of chronology. Older chronological frameworks, whether those of the ‘ancient constitution’, divine providence and even apocalypticism, co-existed with the sense of the revolution as a moment of temporal rupture. Consequently, for Claydon, ‘if modernity happened at the end of the seventeenth century, it cannot have been brought about by self-conscious modernizers.’ (^{Claydon, 2020}, p. 15).

To explore this idea, the article will focus on a particular moment, the tendering of a loyal association to William III in 1696 which, it will be argued, provides evidence of the transitional nature of this period. The 1696 Association in defence of William III was prompted by an assassination plot against the king, revealed in February of that year but which may have been connected to broader Jacobite schemes, initiated by James II himself, for a general rising in opposition to William’s regime. To defend the king from this threat, the ‘Association’ bound subscribers to declare that William was ‘rightful and lawful’ king of these realms, a breach from the de factoist position established by the new oaths of allegiance to William and Mary in 1689. These oaths had only required subscribers to promise loyalty to the joint monarchs and required no explicit acknowledgment of their legitimacy (^{Pincus, 2009}, ch. 14). This declaration can be seen as the fulfilment of schemes for an oath of abjuration against James’s right to the throne introduced in Parliament in 1690 and 1693. These schemes were an extension of the frustrated Whig campaign in the immediate aftermath of the Revolution for legal retribution against their Tory opponents (^{Goldie, 2021}).

Certainly, a number of historians have identified this declaration as targeted at Tory office holders who had managed to accommodate themselves to the regime on the basis of being able to accept William as king de facto if not de jure. The Association, however, was not tendered to office holders alone and instead was distributed for national subscription. The scale of the returns in some areas (the largest reported, for Suffolk, allegedly contained over 70,000 subscribers) meant that the Association was taken by nearly all adult males (^{Vallance, 2019b}). In total, 473 Association rolls are preserved in the UK National Archives, with a handful of others kept in local record offices. This article employs a sample of roughly a third of these rolls to examine what they can reveal about the nature of the revolution of 1688.

Some historians have seen the Association as indicative of a profound break with the past. For Pincus, the Association provides us with a window through which to see how the nation had been transformed by the revolution of 1688. Though the Association was modeled on an Elizabethan precedent, Pincus argues that it demonstrates ‘the social gulf separating the Williamite political nation from its Elizabethan predecessor.’ (^{Pincus, 2009}, p. 443). With a slightly different emphasis, Mark Knights has seen the social reach of the Williamite Association as testimony of an ‘inclusive, consensual society’ (^{Knights, 2005}, p. 160). It was, both Knights and Pincus suggest, an attempt to secure active, public consent to ‘revolution principles’ (^{Knights, 2005}, p. 156). Both note that the Association was promoted by sermons and pamphlets which advanced a Lockeian interpretation of political authority. Authority was ultimately founded on a contract between the monarch and the people. Mass subscriptions to the Association effectively represented the popular consent that was essential to legitimate political authority. Moreover, despite the fact that the pulpit remained a key channel through which these messages were communicated and although William’s survival was identified as a providential blessing, Pincus in particular urges that the Association of 1696 was shorn of many of the confessional, eschatological aspirations of earlier Protestant associations. For Pincus, the Association not only provides an important indication of the transformation of political culture after the revolution but also the transformation of the state. The capacity of the Williamite state to secure this demonstration of popular loyalty, unparalleled then and since, not only from English counties but also from England’s far-flung colonies, demonstrates an administration that had also been effectively modernized by the mid-1690s (^{Knights, 2005}, p. 156-160; ^{Pincus, 2009}, p. 450-471).

Some elements of this picture cannot be disputed: the 1696 Association represents the most extensive exercise in canvassing public loyalty undertaken in the early modern period. As Mark Knights has noted, press reporting of the Association frequently focused on the numbers of individuals subscribing, suggesting a political culture in which public opinion played an increasingly important role in legitimizing the regime¹. Even here, though, post-revolutionary developments appear less of a dramatic departure from the past if the 1640s rather than the 1580s are used as a point of comparison. During that revolutionary decade, several tests of loyalty were introduced, such as the Solemn League and Covenant of 1643 and the Engagement of Loyalty to the Commonwealth of 1649 which were intended to be taken by all men over the age of eighteen (^{Vallance, 2005}; Firth & Rait, 1911, i, p. 376-378; ii, p. 325-329).

Indeed, in some respects, subscription patterns to the Association demonstrate a return to more traditional understandings of the political nation than we can see evidenced through similar exercises in the 1640s. For example, subscription returns for the Commons’ Protestation of 1641 frequently include the names of women (^{Walter, 2016}). The same is not true of the Association which was predominately signed by men and in which we can see the enduring idea of the male, propertied householder as the sole legitimate political actor. The constable of Wilstone, Hertfordshire, for example, certified that ‘all ye Inhabitants’ had taken the Association but his return only contained the names of men². The care with which those administering the Suffolk association demonstrated in distinguishing between the status of male subscribers, whether householders, tenants or lodgers, also indicates that property holding as well as gender remained a critical marker of political status (^{Vallance, 2019b}). While some rolls may have recorded the subscriptions of nearly all adult, male inhabitants, others made clear that the text had been tendered only to persons of ‘Note and Quality’³.

These established hierarchical and gendered understandings of authority were evident in England’s colonies as well. When two councillors of the Province of New Hampshire, Richard Waldern and Thomas Vaughan, were dismissed from their offices for failure to take the Association, they contested their expulsion, stating that they had ‘expected to sign it together with the Governor and Council, and thought it improper for Councillors to sign an Association with the rabble’⁴. So while the evidence from subscription returns certainly suggests that the Association penetrated down the social scale, this does not mean that the activity erased established understandings of political entitlement.

Pincus has also argued that the range of groups submitting Association rolls is also indicative of the transformation of the state as well as society in the early modern period. Again, in contrast to the Elizabethan association, the Crown was presented with Association rolls from its growing army, navy and bureaucracy, as well as from the public (^{Pincus, 2009}, p. 466-467). The greater range of groups and organisations presenting Associations and their much wider geographical spread in comparison with 1584 is clearly evident. However, if the post-revolutionary state had undeniably grown in size, the 1696 Association did not necessarily show it as a fundamentally modern organization. The Association rolls that survive are remarkably varied in their organization. While many appear to provide the original subscription records, in other cases, duplicates were submitted. In the case of the Derbyshire Association roll, containing some four thousand names, the document represented a transcription of the original subscription vouchers which remained with the clerk of the peace⁵. In some urban centres, Durham and Hereford for example, subscriptions were organized first by office-holders and then by occupational groupings⁶.

In other corporations, however, no such organizational strategies were used⁷. The majority of these rolls remained handwritten, with only some Hertfordshire hundreds in the sample consulted here making use of printed texts of the Association⁸. The administration of the Association and the formatting of Association rolls therefore seems to have largely been dictated by local decisions and assumptions, rather than by central direction. Consequently, while the Association could certainly provide evidence of the growth of the state and the expansion of the Crown’s dominions, it does not provide evidence of increasing bureaucratization and rationalization of administration which has been identified as a hallmark of modernization.

In this way, the Association may be symptomatic of the diverging patterns of bureaucratic change in the post-revolutionary state identified by Gary W. Cox. Cox contrasts the increasing professionalization and bureaucratization of the fiscal-military state under Parliamentary oversight with the under-development of the British civil state, including of its domestic security apparatus. Cox attributes this disparity in state development to ongoing Parliamentary concern as to the Crown’s use of powers of appointment to effectively ‘buy’ political influence and, therefore, a related reluctance to support any increase of Crown bureaucracy which could be exploited to increase this influence, resulting, effectively in a two-track pattern of state modernization (^{Cox, 2020}). The Association of 1696 and later anti-Jacobite associations were notable in this respect for ostensibly seeking to address serious threats to the regime not via security apparatus directly under the control of government but through encouraging public, voluntary action in defence of the Crown (^{Vallance, 2005}, p. 200-216).

Further continuities are evident in the employment of the Association as an expression of popular loyalty. Association rolls were regularly accompanied by or included a loyal address. Addresses, public expressions of loyalty, usually directed towards the monarch, had been a significant feature of post-Restoration political culture and were an important weapon of the Crown both in the ‘Tory Reaction’ and again in support of James II’s policy of toleration. The practice was continued post-revolution and, indeed, mass addressing campaigns were launched in support of William’s title not only in 1696 but also in 1694-1695 (following the death of Mary II), in 1697-1698 to give thanks for the peace of Nijmegen and again in 1701-2 in response to Louis XIV’s recognition of the ‘Old Pretender’. Addressing would remain an important way of expressing popular loyalty on into the Hanoverian period (^{Knights, 2005}, ch. 3; ^{Vallance, 2019c}). This, in itself, could be seen as a further counter to the ‘liberal’ view of the revolution of 1688: Kenneth Minogue has argued that liberalism’s defining characteristic is an aversion to ‘excessive loyalty’, meaning those commitments which could be seen as limiting or jeopardizing individual autonomy (^{Minogue, 1987}, p. 203).

While there was an undoubted continuity in the use of addresses by the Crown, there was, arguably a difference in how the public responded to the demands to demonstrate loyalty which, in turn, casts doubt on whether the Association of 1696 represents a successful attempt to gain active public support for ‘revolution principles’. In the immediate aftermath of the revolution of 1688, the so-called ‘Allegiance Controversy’ saw the triumph of arguments urging loyalty to William and Mary as monarchs de facto rather than de jure (^{Goldie, 1980}). As has already been noted, this de facto position was even enshrined in modified oaths of allegiance to the new monarchs. The 1696 Association, however, required subscribers to acknowledge William as ‘rightful and lawful’ monarch and was defended on the basis of arguments grounded in Lockeian contract theory. According to Pincus, it was therefore ‘impossible for someone to believe that William was merely de facto king and sign the association.’ Consequently, while the majority subscribed, Pincus suggests around 20% of the (adult male) population could not be persuaded to take the Association ^{(Pincus, 2009}, p. 470).

As Jeffrey Chambers has recently noted, however, the very fact that the vast majority of the public did feel able to subscribe indicates that the monarchy enjoyed broad support. In one respect, this was simply a consequence of the immediate circumstances in which the Association was tendered, with the nation rallying around the Crown in the face of the threat of an assassination attempt linked with the possibility of foreign invasion. But by 1696 too, the choice was not simply between de facto and de jure allegiance as it had been understood in 1688-1689. Chambers contends that many contemporaries now argued for giving allegiance to William III on the grounds that his title was now legally and constitutionally recognized (^{Chamber, 2016}, ch. 7). Consequently, while there may have remained many who would struggle with acknowledging William as a ‘rightful’ monarch (that is, having the right to the throne by hereditary succession), there was now a significant body of opinion which could concede that the king’s title was ‘lawful’ (meaning enshrined in law) (^{Vallance, 2005}, p. 206-207). Hence, while the difference between describing William as ‘rightful and lawful’ monarch or having a ‘right by law’, as stated in the Lichfield Association could be seen as minimal, they in fact represented distinct ideological positions (^{Kettle, 2018}, p. 230).

Pincus has also suggested that the Association rolls demonstrated that the Revolution ‘was not understood in confessional terms.’ (^{Pincus, 2009}, p. 469). He has noted how even some English Catholics sought to pledge allegiance to the Williamite regime and how the plot was understood as involving Protestants as well as ‘papists’ (Pincus, 2009, p. 469-470). Yet, the addresses of loyalty which accompanied these Association rolls suggested that post-revolutionary society was perhaps less removed from the era of religious warfare than this indicates. It is true that few addresses shared the apocalyptic feel of that from Bere Regis in Dorset, which, echoing 2 Samuel 14-17 hoped that William would become ‘an Angell of God’ and so ‘discern between ye good & ye wicked that yor searching & trying may not be over till a compleat purge be made, of all yor Maties Monstrous Bloodthirsty Enemies, both nigh and far off’⁹.

Yet other communities did associate the assassination attempt with Catholicism, as that from the Protestant dissenters of Congleton which described the conspirators as seeking ‘to enslave these Kingdoms & all Europe to an arbitrary & Popish power, to the utter subverting our Religion Laws & Liberties.’¹⁰. Similarly, the address from Lyme Regis described the plotters as enemies of ‘our Religion’ as well as liberty¹¹. In Lancashire, a county with a significant Catholic population and a recent history of engagement in Jacobite plotting, ‘papists’ were clearly identified as the main threat to the regime. While some non-subscribers were identified as ‘protestants’ or ‘Quakers’, other lists were simply headed ‘Recusants’¹². Many other addresses expressed thanks and acknowledgement of the personal jeopardy the king had placed himself in defending the Protestant religion¹³. These addresses also frequently acknowledged the ‘Signall Providence of God’ in discovering the plot and preserving the king¹⁴.

The addresses not only demonstrated the survival of a providential worldview, they also revealed relatively minimal engagement with recent, human history. One notable feature of the addresses to William is actually how rarely they reflected on the revolution. For example, if we look at the addresses issued after the death of Queen Mary, texts such as that from Kingston upon Thames which referred to their rights having been ‘late Invaded’ by James II were extremely rare¹⁵. The address from Cornwall acknowledging William as ‘rightful and lawful’ monarch was similarly exceptional¹⁶.

This pattern was repeated in the addresses accompanying the Association returns. While addresses from non-conformist churches occasionally gave thanks for toleration as a direct benefit of the revolution, texts which commented upon recent political events were scant¹⁷. The address from the Devon grand jury, deputy lieutenants and justices of the peace was unusual in reflecting on the events of 1688: playing on the memory of the reception of William of Orange and his troops in Exeter in November of that year, the address proclaimed that ‘as wee had the honor to bee the first recevers of yor Sacred person soe wee have still the zeale to bee some of the most forward in defence thereof and Associate as followeth’¹⁸. This was in contrast to the addressing activity of the late Restoration period in which recent history, especially that of the civil wars, was frequently invoked, often as a means of demonstrating the loyalty of a particular community (^{Vallance, 2019c}, ch. 3).

Indeed, the addresses made to William often displayed a remarkable amnesia in general about recent events: the address from Cockermouth was exceptional in presenting their text as a renewal of the promises of loyalty to the king made upon the death of ‘our late Queen Mary of ever blessed memory’¹⁹. Overall, despite the huge scale of the returns, the Association of 1696 appears to have made relatively little impression on public memory. Only four of the addresses produced in 1701-1702 in the wake of Louis XIV’s acknowledgment of the title of the ‘Old Pretender’, James Francis Edward Stuart, made reference to the Association to King William made only a few years earlier (^{Vallance, 2019c}, p. 100). Moreover, while the intention of the campaigns from 1695 onwards may have been to drive public acceptance of a broadly ‘legitimist’, if not Whig/Lockeian position with regard to William’s title, the returns here appear to have been less impressive, even by the end of the king’s reign, than may be indicated by the subscriptions of hundreds of thousands of William’s subjects. In both 1695 and in 1696 it was noticeable that the full text of addresses was rarely reproduced in the contemporary press. This was a marked contrast with the 1680s when the periodical press was often dominated by addressing activity for weeks or even months on end (^{Vallance, 2019c}, ch. 3). The lack of historical comment and the absence of large-scale reproduction of these texts may suggest that if these campaigns were intended to secure active, public support for ‘revolution principles’ they were largely a failure in this regard.

There were powerful continuities, then, in both the demand for popular loyalty and in the medium through which it was expressed. At a conceptual and linguistic level, though, we can arguably see important changes taking place in the meaning of loyalty. Addresses to William frequently presented him as the defender of the Protestant religion and European as well as English liberties. For example, that issued from the corporation of Southampton after Mary’s death expressed the wish that his grief would not hinder the king from his ‘Heroick Designs, for freeing us and all Europe from Slavery and Oppression’²⁰. Similar expressions can be found in many of the addresses accompanying the 1696 Association (^{Vallance, 2005}, p. 203). The preservation of the king was identified as critical not only to protecting Protestantism but ensuring the ‘safety and Happiness of Europe’²¹. In this respect, Pincus is undoubtedly right that this was not a straightforwardly ‘Protestant’ association: most often, the reformed religion was one component within a broader European struggle against the French king’s ambition to establish himself as ‘universal monarch’, holding sway across Protestant and Catholic states alike. Of course, these sorts of expressions had the additional benefit of avoiding the question of William’s hereditary or legal right and instead focusing on the personal characteristics that made him an effective defender of the liberties of Europe.

Even so, these addresses focused on William less as a man and more as a medium through which certain objectives could be achieved: the liberty of Europe, the defence of the Protestant religion. The king was not so much the direct object of loyalty as the conduit to the true target of public affection, whether England’s church or its laws. This fits with the assessment of Matthew McCormack regarding the changing ‘lexicon of loyalty’ in the post-revolutionary era (^{McCormack, 2012}). For supporters of the revolution, loyalty was increasingly focused upon concepts and institutions rather than individuals. The absence of printed addresses in the reign of William III prevents us from undertaking quantitative analysis of this in the 1690s but corpus analysis of compendia of addresses from the reigns of Queen Anne and George II allow us to identify general patterns which support McCormack’s observations. The conception of loyalty developed in these texts did nod towards a broadly contractual relationship in which the affection of the public was the reward for the blessings they enjoyed as a ‘free People’. It remained, nonetheless, a value that was expressed in affective terms, though the understandings of public feeling would also be transformed later in the eighteenth century by notions of ‘sensibility’ (^{Vallance, 2019c}, ch. 7).

In conclusion, the 1696 Association arguably demonstrates the considerable continuities in the political culture of the late seventeenth and early eighteenth centuries, rather than demonstrating a revolutionary rupture. Mass oath-taking and loyal addressing were subscriptional practices that spanned the period from the outbreak of the civil war to the Hanoverian succession. Indeed, the Williamite Association was consciously backward-looking, in its appeal to its Elizabethan precursor. Similarly, these practices were supported by an expansive printed news culture that pre-dated the lapsing of the Licensing Act in 1695. This was, though, a political culture in transition. Koselleck’s notion of ‘Sattelzeit’ implies an awareness of temporal change. It has been argued in an early modern English context that this awareness was in part a product of these popular practices of mass addressing and oath-taking which created a self-reflective national political culture that was conscious of and responsive to shifts in public loyalty. This temporally structured news culture facilitated the creation of party identities through mapping political affiliation over time. It continued to express loyalty primarily in affective terms but these public affections were directed at values and institutions as well as individuals. If the de-personalisation of the professions of loyalty to William III had often been a means of avoiding an explicit endorsement of his right to the throne, by the end of the Stuart age, the language of loyalty arguably did demonstrate an implicit acceptance of revolution principles in locating the value in concepts such as civil and religious liberty and institutions such as the constitution (^{Vallance, 2019c}).

Yet, the Association of 1696 also presented a snapshot of a society in which the transformation of both state and its political culture appeared uneven and incomplete. As impressive as the scale of this exercise in surveying public loyalty was, it was an enterprise which still appeared dependent on that ‘unacknowledged republic’ of unpaid local officeholders identified by Mark Goldie, rather than a developed state bureaucracy (^{Goldie, 2001}). The absence of such administrative machinery was reflected in the organization of these returns on the basis of long-standing assumptions regarding political authority which remained hierarchical and gendered, even in those areas such as Suffolk where nearly all adult males may have taken the Association.

The addresses which accompanied many of these Association rolls also revealed a society in which divine providence remained an important framework for understanding events. Some addresses acknowledged the benefit of the revolution in delivering toleration to Protestant groups. However, many others depicted the greatest threat to the Williamite regime as coming from domestic ‘papists’ and foreign Catholic powers. For a minority, such as the inhabitants of Bere Regis, Dorset, this struggle between Protestantism and Catholicism, could still be cast in biblical terms. The general avoidance of commentary on recent history in these 1696 texts, went against the wider role of addressing campaigns in fashioning a self-reflective, temporally structured political culture. Here, the legitimist position endorsed in 1696, acknowledging the legal fact of the revolution but avoiding comment on the circumstances through which these laws were implemented disrupted linear understandings of time. Instead, it encouraged a form of loyalty to the Crown arguably shorn from reflections on either Britain’s past or its future.

This short article has attempted to critique readings of 1688 which either represent it as the coda of the early modern period or the prologue of modernity. Instead, it has taken Koselleck’s notion of Sattelziet to argue for a moment in which multiple understandings of time co-existed. It was a moment, contrary to Pincus’ interpretation, in which possible futures were not fully realized or articulated. Instead, conceptual change could be the product of seemingly contradictory impulses. Demands to recognize William III as ‘rightful and lawful’ monarch and therefore endorse the revolution could, at the same time, actually discourage reflection on the recent past. A call to defend the reigning monarch self-consciously modelled on historical precedent might nonetheless generate understandings of loyalty which gradually relocated the focus of that value from the person of the king to an abstraction such as liberty.