Abstract. In the third volume of the History of England, David Hume considers the political ramifications of the Protestant reformation with a “Digression concerning the ecclesiastical state.” He advocates the establishment of a state church, believing it will dampen religious “enthusiasm” in the polity. Unlike later secularization theorists, Hume assumes an intractable basis for religion in the human passions. Tensions in Hume’s “cooptation” strategy are evident from Adam Smith’s famous attack upon it in section five of The Wealth of Nations, and in Hume’s own treatment of seventeenth century independency in the fifth volume of the History. Smith argues that public competition among sects facilitates political moderation. In History V Hume stresses the positive role of enthusiasm in fostering civil liberty. This article traces Hume’s indecision to his “external” mode of moral and historical analysis, arguing that a secular policy on religion cannot proceed fruitfully without engaging the theological particulars of the religions at issue.
Keywords: David Hume, religion, secularism, Adam Smith, politics.
David Hume’s political treatment of religion can appear ironic to contemporary readers. After Marx, Weber, Durkheim, and others, the disappearance of religion from public life has been described as an inevitable and positive development. In light of this view on the process of secularization, it is striking to see a thinker with a reputation for atheism and hostility to religion adopting a policy of engagement and establishment.  As Hume made clear in his Natural History of Religion, he believed the passions giving rise to religion were natural, ineradicable, and most often, harmful.  His political response to this predicament was to advocate – most explicitly in the third volume of the History of England – church establishment.
In assessing Hume’s philosophy of religion, Jennifer Herdt has argued that Hume “dismembered theistic belief.”  That is, he isolated purely religious questions from social ones, creating a space where theological disputes were rendered independent of any practical ramification. In this, Hume was actively shaping the public definition of religion so as to minimize its impact. Following Herdt’s thesis, I will suggest that we may view Hume’s philosophy of religion as one half in a division of intellectual labor; his political considerations on religious institutions represent the second half. Hume’s establishment reflections apply to religious politics a secularized, non-religious interpretive lens, the goal being to stabilize and moderate the polity. In the following pages, I will locate the basis of Hume’s support for Anglican establishment in his worries on enthusiasm, with the idea that his approach to the question of religious passions is conditioned by his fully “external” mode of moral and political analysis. I will examine his “cooptation” strategy from two critical angles. The first is Adam Smith’s famous attack on it in section V of the Wealth of Nations; Smith’s “competition of sects” approach calls into question Hume’s phenomenology of religious passion. Second is Hume’s own prevarication on establishment in the fifth volume of the History, where he connects toleration to Protestant independency. I conclude by suggesting that some lessons emerge from Hume’s indecision. His grudging adoption of “independent” interpretive premises suggests the futility of a thoroughly instrumental account of religion; there are limits to an external and non-theological hermeneutic of religious politics.
Hume’s most direct statement on church-state relations appears in quite succinct form at the beginning of History III, Chapter 29.  In discussing the initial stages of the Protestant Reformation, Hume says he will address the theological disputes at issue, and the particular “abuses” that generated a need for reform. He nonetheless takes initial leave of theology, arguing that we should consider the reasons “why there must be an ecclesiastical order, and a public establishment of religion in every civilized community.”  Summarized briefly, his argument is as follows: In every state, there are professions needing no help from a magistrate because they benefit their practitioners directly. Others need encouragement, since they aid the public, but do not necessarily benefit the people who would engage in them. The question requiring examination is whether the clergy can be thus described, and by extension, whether religion itself needs encouragement as a public force. Hume’s suggestion is that any prudent leader will wish to discourage “interested diligence” in the clergy, because in all religion “except the true” such diligence is pernicious. Each religious leader will strive to outpace his competitors by exciting “disorderly affections of the human frame,” appealing to the most outlandish and superstitious doctrines. This reading understands religious passions to be publicly disruptive, requiring at least some direction, and perhaps more, a dampening. The solution is to “bribe [the leaders’ – A.Sz.] indolence” by fixing salaried offices, rendering this would-be contest in “ghostly practice” unnecessary. Hume concludes by observing that most establishments begin with religious justifications, though they prove useful for political interests, and should be maintained on this basis. 
Some obvious questions emerge: First, Hume typically emphasizes particulars before drawing broad generalizations, but he pointedly reverses the order in this case – why? Second, Hume does not specify whether religious “enterprise” needs state incorporation because it wants direction or because it requires attenuating: Is religion ever beneficial? Third, there is the logic of incorporation itself. If one wishes to mitigate religious enthusiasm, what reasons are there to suppose that the “bribing” will be effective?
Hume discusses generalization and theoretical inference in several of his essays, particularly Of National Characters, Of the Rise and Progress of the Arts and Sciences [RPAS], and Of Commerce. In the last-mentioned work, he cautions against the too-quick use of “abstruse” thought, but supports nonetheless some theorizing in “general” subjects where “our speculations can scarcely ever be too fine, provided they be just.”  He sees this as a practical matter, saying that “it is the chief business of philosophers to regard the general course of things. I may add, that it is also the chief business of politicians.”  A similar claim appears in the RPAS essay, where Hume expresses higher confidence regarding the discernment of causes where forces operate on great numbers.  These themes were solidly instituted by the time Hume was writing the History. We must therefore regard his “digression” on establishment as a deliberate claim on general causes for a political end. The goal of managing religion, in other words, involved a “just” generalization. The nature of this generalization, however, is opaque. The problem is magnified by the fact that in speaking of religious matters elsewhere, Hume is quite circumspect concerning just how much of general causes can be known from limited effects. 
This raises our second question, “a generalization on what?” or “what is being treated by establishment?” Hume addresses this matter, albeit obliquely, in the essay Of Superstition and Enthusiasm [SE], and further in the second Enquiry. SE precedes the History by a decade and introduces some basic themes on the political working of the passions. Hume classed older religious sects, suffused with ritual and dominated by priests, as “superstition,” producing “weakness, fear, melancholy, [and] ignorance.” His obvious target here is Roman Catholicism. By contrast he placed the newer Protestant sects – especially those derived from the teachings of Calvin – under the category of “enthusiasm.” These came from “hope, pride, presumption, a warm imagination, together with ignorance.”  Superstition and enthusiasm corrupt “true religion.” I assume here a minimal sense of this curious term, meaning that where “true religion” appears, men act as they would by force of nature alone.  This assumption, like the SE argument itself, follows the tenor of Hume’s epistemology and historiography: passions generate doctrines, not the other way around. Upon observing that violent and presumptive emotions generate a certain species of doctrine or institution, Hume deduces from English events that enthusiasm flares up violently, settles down, and facilitates civil liberty in the process.  But just how universal is this process? Does it indicate a general cause, upon which a magistrate may work? Is the process of weakening enthusiasm the same as removing religion itself, or merely its worst effects?  And to these questions we should add one concerning Hume’s aim: Is it in fostering civil liberty, or rather, as seems more immediately the case in History III, preserving order and moderation? 
Even if we assume a clear answer to this last matter, Hume’s retrospective mode of analysis poses some troubling issues for a politician, who will face our third question. Should he favor establishment of any religion, or do only some of them admit of “bribing?” On what grounds do we expect that our best chance at dampening harmful emotion is establishment? Hume’s answer in History III is that establishment dealt with a persistent force, that it diverted “metaphysical” impulses into lower, economic ones, and perhaps more basically, that a certain portion of religious motivation is always tied up with a thirst for pre-eminence. Religious feeling is a given, but its form and subsequently its strength are elastic, responsive to political incentives. The aim is to fix a “languid” default and prevent its being prevailed upon by enterprising innovators.  This facilitates moderation in the polity, though not because religion itself generates moderation. Hume assumes a rough but stable middle in the passions; religious fanaticism decenters the polity from this middle. Attenuate fanaticism, Hume argues, and one thereby fosters moderation in the polity. But, it is far from necessary that Hume’s incorporating position should follow from his basic assumption of a natural middle. This is clear from Adam Smith’s critique, which appears in Section V of WN. Smith shares Hume’s view on the political dangers of religious enthusiasm, but differs in his prescription.
Politically speaking, Adam Smith’s goal for religion is similar to Hume’s.  In Smith’s words, it is to foster “pure and rational religion, free from every mixture of absurdity, imposture, or fanaticism.”  Still, Smith advocates religious liberty in the WN by repudiating Hume. After quoting Hume’s History III argument at length, Smith advances a two-part account for the reduction of religious power. His analysis of secularization is more thorough and consistent than Hume’s, though it may be asked whether analytic rigor is purchased at the cost of moral subtlety. Smith observes that political goals are seldom the ends for which states establish churches. As we have seen, Hume had already introduced this point. Yet Smith pushes it in a new direction, tracing the connection of religion and politics explicitly to conflict. The idea is not, as some in later centuries have it, that conflict vanishes without religion. It is instead that religion is the natural repository of, and vehicle for, inflamed passions. Each party “has either found it, or imagined it, for its interest, to league itself with some one or other of the contending religious sects.”  Once one has “called in the aid of religion,” then its tenets are magnified in power with the victory of the adopting party.
Smith suggests that it might be advantageous to have a multiplicity of sects, and with it a growth of individual conscience. By facilitating a “market” in faiths, a general moderation of enthusiasm will follow more effectively from competition than it would from an establishment, itself set in place by merely one dominant form of irrationality. Smith is skeptical of even the initial reasons for adopting religion: political parties either “find it,” or “imagine it” in their interest to enlist religion’s help. But where does “real” interest shade into “imagined” interest? If it is difficult for the philosopher to fix boundaries here, then it is more so for the politician. Seemingly undermining his own prescription, Smith doubts whether a law for complete religious toleration could be established in any country, since “positive law always has been, and probably always will be, more or less influenced by popular superstition and enthusiasm.”  Because the boundaries between enthusiasm and “pure” religion are never entirely clear, it would be folly to expect the politician alone fully to act for them. The best he can do in facilitating the cause of moderation is to blunt some of the excesses of “austere” morality by encouraging science and philosophy among the upper orders of society, and by supplying “publick diversions” for the lower.  There is one more part to Smith’s policy: emoluments to clergy – if any – should be low and equal, redirecting the drive to preeminence into academic pursuit. Crucially, Smith did not think these prescriptions effective – especially the last – in just any circumstance. 
At this juncture some problems should be evident: Why prescribe science and entertainments? Indeed, why bother with approaching “rational religion” at all if it is doomed to fall before popular superstition and enthusiasm? Hume’s view appears comparatively more sensible: own what cannot be rooted out. Smith’s point on the equality of clerical offices seems at best a mere adjustment of Hume’s policy, at worst a multiplication of its difficulties. Most obviously, it presumes an establishment – is Smith blatantly contradicting himself? Further, how does Smith warrant a Presbyterian establishment? These matters are resolved by the second, historical-materialist part of the WN account. Smith saw developments under the Presbyterians as manifesting certain psychological and economic forces, and his argument represents a more clearly drawn doctrine on this matter than Hume’s.
We began by exploring Hume’s moral/political view on generalizations, noting his claim that religious passions were amenable to redirection through economic incentives. Hume did not, though, believe that such passions could be directed entirely in this way. Smith goes further, saying that redirection is impossible – at least from conscious decision. His innovation is in the claim that history and social development would accomplish what political design might not. He presses this theory by way of a long discursus on the property relations of the Catholic Church. The monopoly of the Church on land created a “rent” both in money and morality. The moral economy of beneficial acts was perforce “cornered,” defined to suit one organization. Under such conditions, competition among religious views was impossible. The Reformation broke this monopoly, because developments in religion and morals were inseparable from those in land and commerce.
Most contemporary readers will register a forerunning of Marx in this argument, but Smith is not working in such a determinist framework. His secularization argument in WN fits comfortably with the “invisible hand” of TMS IV:  it connects an inadvertent moral or economic phenomenon at the individual level with a social process. Indeed, the proper frame of reference here is the “stadial” speculative history found in Millar, Robertson, Kames, and elsewhere in Smith.  The direct forerunning is as much moral as economic, more of Kant rather than of Marx. As in Kant’s “guarantee” of perpetual peace, Smith argues that we might plausibly infer a direction of events, and that this suffices morally for adopting a goal. He projects that with the relative “independence” achieved through Protestantism in general, and Presbyterianism in particular, the economic and spiritual power of the Catholic Church “is now likely, in the course of a few centuries more, perhaps, to crumble into ruins altogether.”  Having adopted what we might call a loose historical telos, Smith defines high-church establishment as inherently “anti-civil.” This produces a much clearer repudiation of Catholicism – and conceivably, high Anglicanism – than we find in Hume. In Smith’s telling, the purpose of secularization is the replacement of superstition with enthusiasm, and the subsequent blunting of enthusiasm by means of competition. A market in belief will produce this result, but for reasons emerging from the historical narrative, it will do so only at a particular juncture.
Smith, therefore, advances some very specific answers to the questions we directed at Hume. Religion serves political and moral ends, facilitating the development of individual conscience. It does so, however, through larger processes of social change, and not at the behest of any conscious political direction. Smith gives us a distinctly anti-Erastian argument, one specifically hostile to the older religious institutions Hume had described as “superstitious”; still, one cannot “bribe” religious power where socio-economic progress has not loosened its moral grip. Once that grip is broken, state measures (science, entertainments, low equality of offices) sustain market processes already underway. There can be no hope that high religious institutions will produce lasting political moderation; this is to misunderstand their place.  Hume’s SE parallels Smith’s history of religion in WN: superstition abates, enthusiasm rages, then moderates, producing liberty. But tellingly, he refuses there, and in NHR and DNR, to adopt Smith’s economistic teleology as an operating generalization. Moral concerns, in other words, do not strictly parallel economic ones. The very manner of generalizing is more circumspect: particular and psychological, rather than sweeping and historical.
This difference in generalizing follows from some basics of Hume’s moral philosophy, which puts larger speculative impulses at arm’s length, but does so in a different manner than Smith’s moral theory does. In moral description Smith focuses typically on internal mechanisms of the imagination.  His speculative inferences, meanwhile, depart from this second-person frame; they are external, as in the case of the teleological reflections we surveyed above. As a consequence of this distinction, we may observe that Smith can with some justification plead that political direction of religion rests on historical observations of a non-moral character (which then, of course, can be said to frame the moral ones). By contrast, Hume accounts for moral phenomena themselves solely in an external fashion, parsing or adjudging internal states by way of external facts. Lacking, and in fact rejecting, Smith’s distinction, Hume reminds us throughout his writings that historical investigation is an inherently moral species of inquiry.  For this reason his psychological inferences tend to be both particular and negative, exhibiting a reticence to freight historical sequences themselves with too much significance, and focusing instead on reoccurring types of behavior.  A paradigm instance of this tendency, particularly in relation to religious phenomena, appears in the Treatise, where Hume advanced the rather striking claim that most people do not actually believe in an afterlife. For this conclusion, he offered a proof invoking acts of religiously inspired violence:
[…] there scarce are any, who believe the immortality of the soul with a true and establish’d judgment; such as is deriv’d from the testimony of travellers and historians. This appears very conspicuously wherever men have occasion to compare the pleasures and pains, the rewards and punishments of this life with those of a future; even tho’ the case does not concern themselves, and there is no violent passion to disturb their judgment. The Roman Catholicks are certainly the most zealous of any sect in the christian world; and yet you’ll find few among the more sensible people of that communion, who do not blame the Gunpowder treason, and the massacre of St. Bartholomew, as cruel and barbarous, tho’ projected or executed against those very people, whom without any scruple they condemn to eternal and infinite punishments. All we can say in excuse for this inconsistency is, that they really do not believe what they affirm concerning a future state; nor is there any better proof of it than the very inconsistency. 
Here is Hume’s characteristic accent on external description. He argues that certain concepts – such as an eternal hereafter – are by virtue of their exceptionality and extremity not bases for “natural” moral thought or behavior. But this judgment – like all judgments for Hume – finds its evidence in repeated instances, and a lack of meaningful exceptions. There is no historical process at issue; the settled disposition of co-religionists speaks against the natural prevalence of the belief in question. What matters more than the spoken belief is the external evidence of behavior (or in this case, its absence). Hume concludes that real belief in a permanent hereafter requires an “artificial” state, as he would subsequently label such a condition in the EPM Dialogue.  This kind of situational-analogical reasoning involves discrete consideration of “secularizing” events, resisting the comprehensive hypotheses of Scottish conjectural history. An account like Smith’s – unambiguously anti-Catholic and descriptively sweeping – simply cannot sit well with Hume’s philosophy. So we might inquire whether, in the absence of any larger historical arc for analysis, Hume was able to trace a consistent picture of religious passions in political life. Since religious enthusiasm and regular moral passions are both discernible as matters of general observation, political treatment of them should reflect this uniformity. As it turns out, though, Hume’s goals and observations both exhibit some shifting of emphasis, particularly when it comes to the classic problem of order versus liberty. A view of some themes from History V demonstrates that these shifts were not merely products of a changing political outlook on Hume’s part; they originated, too, in discrete interpretation of religious doctrine. Here we find that extremity and “artificiality” in passion could explain only so much, even for Hume.
In the main, Hume’s History V account of the 17 th-century religious Independents reinforces the SE position: both establish that enthusiasm generated that peculiar and praiseworthy English capacity for civil liberty. However, this focus on liberty runs up against Hume’s earlier concerns on stability and moderation in History III. The question is not just a situational one, where we can ask whether one might enjoy enthusiasm’s fruits without taking the tree itself. Hume’s “external” account is beset by some deeper and more interesting paradoxes. The biggest of these is that while he connects independency with the development of civil conscience and tolerance, he maintains at the same time that the Independents’ enthusiasm was empty of reasoning.
Hume’s case turns on the relationship between Presbyterians, with their insistence on maintaining some diminished, contractual ecclesiastic authority, and Independents, who seemed to doubt the institutional salience of covenants altogether beyond temporary expedience. One might reasonably insist that a point of theology was at issue, and politique concerns would need to weigh it – but how? Is this possible on that same “external” mode of description that denied belief in the afterlife and presupposed the tractability of religious motives by security and money? Hume sketches the “genius” of the Independents with analyses of the new model army, the treaty of Uxbridge, and the self-denying ordinance, among other events. He largely omits theology – at least at first – by repeating the basic theory of SE, i.e., that religious fervor is a product of unnaturally enlivened passions. From here, he links the enlivening directly to emulation or competition.
Every man, as prompted by the warmth of his temper, excited by emulation, or supported by his habits of hypocrisy, endeavoured to distinguish himself beyond his fellows, and to arrive at a higher pitch of saintship and perfection. In proportion to its degree of fanaticism, each sect became dangerous and destructive; and as the independents went a note higher than the presbyterians, they could less be restrained within any bounds of temper and moderation. 
In History III, Hume’s argument was that one might channel the thirst for distinction into lower economic passions, preserving a “languid” institutional default. The Presbyterian and Independent forms of enthusiasm, though, seemed to resist any initial moderation, and Hume suggests that the differences between the two were matters entirely of “pitch.” His telling of the English civil war revolves around the idea that no one acting – not Charles, not Cromwell, not the religious factions themselves – trusted in the establishment of his opponent’s religion, even under the most moderated policies. So then where does toleration originate? For Hume, it is from the inner direction of feeling that comes from this “higher pitch.” No external validation satisfies, and all efforts among men to set down enduring compacts, even for church maintenance, are distrusted.
The enthusiasm of the presbyterians led them to reject the authority of prelates, to throw off the restraint of liturgies, to retrench ceremonies, to limit the riches and authority of the priestly office: The fanaticism of the independents, exalted to a higher pitch, abolished ecclesiastical government, disdained creeds and systems, neglected every ceremony, and confounded all ranks and orders. The soldier, the merchant, the mechanic, indulging the fervors of zeal, and guided by the illapses of the spirit, resigned himself to an inward and superior direction, and was consecrated, in a manner, by an immediate intercourse and communication with heaven. 
This inward-turning is a central feature of many religious experiences, though its meaning is frequently contested. Many tend to see hermetic mysticism more than political skepticism. So, a problem Hume faces is whether the political view of the Independents ought to be understood as a foreseeable development, something about which the historian or politician might draw a generalization. Further, is this a force to be encouraged? Should Protestant enthusiasm have created toleration and institutional pluralism, and does it represent a phenomenon upon which we might rely? We have seen that Smith’s secularization theory permits us to say “yes,” because he describes the matter in stadial-economic terms, framing the question teleologically. This view justifies confidence in both the growth of liberty and the stability of a new political order. But how to receive Hume’s thinking is a matter of some greater puzzlement, since as we have seen, he eschews teleological projection. Many current readings of this situation assume that Hume’s focus was primarily on toleration, though such readings differ on what normative conclusion we are to draw. There is an issue, in other words, concerning how enthusiasm is to be understood, even once we concede Hume’s point that it had some role in fostering liberty.
Greg Conti argues that Hume’s ES and History positions are actually different. In the first, toleration is an aftereffect of enthusiasm; in the second, it is tied up at some level psychologically with enthusiastic thoughts themselves. In the first stage, Hume still believed religion to be on the wane, whereas by the time of the second he no longer held this to be true. He therefore had to account for toleration by interpreting enthusiastic doctrines themselves. Such doctrines, though illogical, were the means by which regular, non-philosophical people came to toleration; in Conti’s rendering, enthusiastic religion is the “low road” to toleration.  Conti’s work builds on that of Andrew Sabl, who sees toleration as one of Hume’s “artificial virtues.” Like other virtues of this kind, it develops apart from any planning or “politique” considerations of the magistrate, and should thus never be viewed merely as a situational policy, at least not once it has arisen. Indeed, Sabl goes so far as to claim that Hume sees toleration as a “law of nature.”  In a more modest interpretation, Richard Dees highlights the fact that the social attractiveness of toleration was not a logical consequence of its ideological articulations. Dees says Hume accounts for the formation of toleration by seeing it as a “compromise, born out of a long-lasting suffering.”  Dees’ “toleration” is thus a more precarious and situational phenomenon than Sabl’s, even if the two concur on its artificiality. Neither sees Hume entertaining a psychological-doctrinal analysis of religious independency like Conti does.
These readings effectively frame a set of problems for Hume’s position on tolerance and civil liberty, but I believe his difficulties concerning enthusiasm are yet deeper. There is good reason to believe that Hume became even more concerned with civil order as he matured. It has long been noted that his reservations about Toryism lessened after the Jacobite defeat at Culloden attenuated its religious element.  More recently, James Harris has traced Hume’s rightward drift amidst the civil disturbances of the 1750s, sketching how these events resulted in a downplaying of consent in Hume’s political writings.  Patrick Whelan has compared Hume explicitly to Machiavelli, noting a constant tension in his work between “backward facing” emphasis on habit, and a “forward facing” utilitarianism.  These authors’ observations should lead us to understand, first, that Hume was indeed interested in normative prescriptions and, second, that such prescriptions had order and moderation in view as much as – if not more than – civil liberty or tolerance. He could not, therefore, advocate independent enthusiasm without reservation – even in retrospect, even in a “hedged” fashion. The problem is that enthusiasm’s products were not univocally positive, and Hume never quite isolates those positive elements as matters of cause and effect. He claims a uniqueness of commitment for independency, yet seems loath to attribute this uniqueness to doctrine:
Their mind, set afloat in the wide sea of inspiration, could confine itself within no certain limits; and the same variations, in which an enthusiast indulged himself, he was apt, by a natural train of thinking, to permit in others. Of all christian sects, this was the first, which, during its prosperity, as well as its adversity, always adopted the principle of toleration; and, it is remarkable, that so reasonable a doctrine owed its origin, not to reasoning, but to the height of extravagance and fanaticism. 
Thus, a “natural train of thinking” emerges from an “unnatural” set of emotions. Psychological paradoxes notwithstanding, this explanatory move generates a rather nuanced account of modern politics when Hume considers what actually followed. The problem, as he clearly understood, was that peaceful secularizing could not proceed under the Independents. Not content to neuter religious authority, they undermined political authority as well.
[T]his sect, more ardent in the pursuit of liberty, aspired to a total abolition of the monarchy, and even of the aristocracy; and projected an entire equality of rank and order, in a republic, quite free and independent. In consequence of this scheme, they were declared enemies to all proposals for peace, except on such terms as, they knew, it was impossible to obtain […] By terrifying others with the fear of vengeance from the offended prince, they had engaged greater numbers into the opposition against peace, than had adopted their other principles with regard to government and religion. 
This is a penetrating description of a familiar modern paradox: that tolerance and liberty as first-order principles undermine trust, foster paranoia, and in turn, breed intolerance.  By this point, establishment could no longer fix through dampening what had been riven through a cycle of violent emotion and distrust. Still, Hume refused pointedly to resolve this dialectic of liberty and order in the manner of thinkers from Hobbes to Kant, who envisioned a calculus for incompatible judgments. In this negative respect Hume, too, was an “Independent” of a sort, distrusting civil compacts.  His, however, was an “unenthusiastic” independency – an independency, to put the matter in Hutchesonian terms, of “calm passions.”  This resulted in an analysis, the sympathies of which lay decidedly outside of those held by its subjects. Hume weighed countervailing forces “unenthusiastically,” and nowhere was this expressed with more clarity than in his treatment of Archbishop William Laud, a chief opponent of Independent influence.
Laud’s tenure as a prelate, culminating with his accession to Archbishop of Canterbury under Charles I, was marked by steady resistance to the spirit of Puritanism developing in the Church. From an academic temperament and Arminian orientation, Laud introduced liturgical and ritual changes directly opposite to the reformed Calvinist worship found in many areas of the country by the 1630s. And at a time when fighting in Ireland and on the continent had induced a wave of fear concerning “popery,” Laud’s tolerance of Catholics was considered by many an unnecessary and egregious affront. This especially when he expelled from the English Church many who declined his revanchism. Laud was imprisoned at the behest of parliament, and after several years’ time (and in Hume’s view, on specious charges) was executed. Laud attempted, and ultimately failed, to temper humanely a species of religious enthusiasm, albeit by means detested by much of his flock. Hume’s sympathy for him is therefore revealing. Though generally striving in History V to vindicate the effects of Protestant enthusiasm on politics, Hume treats Laud so as to reinforce partly his History III establishment position. He does this by illustrating the absurdities inherent in adopting toleration as a telos, absent a context to grant it meaning. Hume finds both irony and profundity in the fate of a man like Laud, whose traditional theological sympathies led him to introduce some toleration, only to suffer for that fact.
Toleration had hitherto been so little the principle of any christian sect, that even the catholics, the remnant of the religion professed by their fore-fathers, could not obtain from the English the least indulgence [...]. And the enemies of the church were so fair from the beginning, as not to lay claim to liberty of conscience, which they called a toleration for foul murder. 
For Hume our first “natural” thought involves party attachment, and toleration is in this context an admission of weakness; Hume sees this position in the Presbyterians. When experience intervenes with violence, though, one is still left with the problem of what to tolerate. Hume cites Montesquieu’s maxim that “a sect, already formed and advanced, may, with good reason, demand a toleration.” Peace itself seems to demand as much. This premise fits well with Hume’s philosophy, where long-standing arrangements have moral weight.  The Presbyterians, though, cut themselves off from tradition, and the Independents went further, eschewing even the social ties of the present in favor of direct inspiration. So Hume’s position could not be theirs; they could demand toleration politically or rationally only on pain of absurdity. Recall that by Hume’s estimation, the Independents’ enthusiasm had led them “by a natural train of thinking” to extend freedom of conscience to others. The problem, politically speaking, was that any attachment of others to traditions, truths, or conclusions different from the ones the enthusiasts had reached was seen as a threat to their freedom of conscience. Thus, the predicament of Laud, who, as Hume was quick to point out, never actually denied the Independents the right to assemble and worship outside of the official church.
The maxims, however, of his administration were the same that had ever prevailed in England […]. To have changed them for the modern maxims of toleration, how reasonable soever, would have been deemed a very bold and dangerous enterprize. It is a principle advanced by president Montesquieu, that, where the magistrate is satisfied with the established religion, he ought to repress the first attempts towards innovation, and only grant a toleration to sects that are diffused and established. […] According to this principle, Laud’s indulgence to the catholics, and severity to the puritans, would admit of apology. I own, however, that it is very questionable, whether persecution can in any case be justifyed: But, at the same time, it would be hard to give that appellation to Laud’s conduct, who only enforced the act of uniformity, and expelled the clergymen that accepted benefices, and yet refused to observe the ceremonies, which they previously knew to be enjoined by law. He never refused them separate places of worship; because they themselves would have esteemed it impious to demand them, and no less impious to allow them. 
Reflection on Laud’s mitigated toleration led Hume to agree partly with Montesquieu, discarding the business of repressing new sects. Laud’s opponents, meanwhile, embraced absurdities the moment they would enter into a public debate on the matter of being tolerated. How can one object to exclusion from church organizations when one rejects, as a matter of principle, organization itself? How can one object to another’s being tolerated when one stands for complete liberty? Paradoxes aside, philosophical cogency was not the only consideration for Hume, and illogic, hardly unique to Protestant enthusiasts. While he yearned elsewhere for an extinction of popular religion, in the History Hume seemed to accept its inevitable persistence. Smith would propose to dampen the “austere” and “gloomy” parts of religion with science and popular entertainments. Hume argued in History V that the “dampening” would need to happen from within religion itself:
Whatever ridicule, to a philosophical mind, may be thrown on pious ceremonies, it must be confessed, that, during a very religious age, no institutions can be more advantageous to the rude multitude, and tend more to mollify that fierce and gloomy spirit of devotion, to which they are subject. Even the English church, though it had retained a share of popish ceremonies, may justly be thought too naked and unadorned, and still to approach too near the abstract and spiritual religion of the puritans. Laud and his associates, by reviving a few primitive institutions of this nature, corrected the error of the first reformers, and presented to the affrightened and astonished mind, some sensible, exterior observances, which might occupy it during its religious exercises, and abate the violence of its disappointed efforts. 
Hume’s chief criticism of Laud was that he acted “not with the enlarged sentiments and cool reflection of a legislator, but with the intemperate zeal of a sectary.” This, however, seems too fine a psychological point to put on behavior that, by Hume’s own admission, was bound for a less-than-charitable reception from Laud’s adversaries. Also, many of Laud’s reforms met Hume’s or Smith’s “politic” concerns. For instance, in addition to his changes to the altar and the prayer book, Laud had directed a loosening of behavioral strictures on the Sabbath, reincorporating feasts and other communal celebrations. Like Smith, he sought to dispel “gloom” and severity. Hume does not say that another policy could have averted civil war, but he does seem to be underscoring the general point that religious thinking was inadequate to politics.
That said, we are still left with the puzzle of whether a fully secular political-historical theory can generate meaningful political norms concerning the management of religion, without itself delving into the particulars of religious doctrine. On this question the previously-mentioned readings from the secondary literature generate answers, albeit ones that cleave too closely to Smith’s teleological position, or which neglect the depth of Hume’s predicament. Conti’s “dual position” reading is correct in noting that Hume needs an interpretation of enthusiastic doctrine, but there is little reason to conclude that this problem did not exist at the SE stage as well as that of the History. And from the side of prescription, our observations above would have it that Hume tilted away from the liberty-and- -tolerance concern in the 1750s, not toward it. That is to say, the History treatments should be viewed as an exploration of an ongoing concern, not as the product of a shifted estimation of the power of religion. Hume can fall back neither on a progression, nor on a retrenchment, as solutions to what is arguably a hermeneutic issue. This observation plays even more strongly against Sabl and Dees. Sabl’s “artificial virtue” position has it that Hume is rejecting “politique” management of religion. But we have argued that such management is precisely what Hume leaves open, even if he concedes its difficulty. Further, in Hume’s philosophy artificial virtues are fundamental to political society, and are not mere addenda. If religious toleration is an artificial virtue, it is a very special one on Hume’s account. The point is not merely that political society can do without it, which Hume clearly believed. It is also not just that Hume’s “Tory” sympathies led him increasingly to concern himself somewhat more with order than with liberty. Rather, the peculiarity is in the fact that all artificial virtue for Hume involves a backlog of unconscious, common performance. But religious argument is about the meaning and value of those patterns themselves. Religious debate involves specifically those things about which Hume’s entire philosophy is skeptical: notions of the good.
Thus, the difference between Conti’s and Dees’ reading is important. Dees gets right the “externalist” element of Hume’s theory that I have traced here. According to this way of seeing things, there is no inherent reason why toleration should result from this doctrine or that, one set of practices or another. Nonetheless, Conti is right that Hume is brought to engage in psychological speculation on doctrine, and to posit a kind of logic: the Independents acted on the basis of a “natural train of thinking.” If this is a “just generalization,” it is hardly a just Humean generalization. Hume’s establishment position cannot avoid a critical kind of tension: between a thoroughgoing external, empiricist analysis of religion as blind passions on one hand, and a necessary interpretation – and perhaps even evaluation – of doctrine on the other. Whether a figure like Laud is ultimately to be praised or excused for what followed his policies is dependent upon our assessment of the psychological and doctrinal qualities of those whom he opposed. It does not appear that Hume resolved, or even could resolve this problem. The one thing evidently consistent in his view is that he resisted any adoption of a telos, even an indirect one like Smith’s.
Our central question on the wisdom of ecclesiastic order seems to find slightly different answers in Hume’s History III and V accounts. In the former, there is an assertion that establishment is to be favored; in the latter – that it cannot generate the public goods of civil liberty and tolerance. But in both cases, some psychological interpretation is necessary to “read” the events generated by religious passion. Hume never embraces Smith’s late Presbyterian accommodation outright as a guarantor of toleration. But nor does his partial approval of Laud bespeak an unqualified high-church political allegiance. Hume’s relatively brief take on the issue in the DNR only confirms that he had reached a point of some ambivalence. Having worked on the DNR right up to his death, and having read WN, Hume was surely aware of Smith’s criticism of his History III position. This did not convince him that he was wrong, though it did induce a statement – in the voice of Philo – to the effect that there was probably no systematic and all-encompassing policy to be followed.
Cleanthes argues that all religion – even the “popular” variety actually found in the world – is salutary, even necessary, for good politics and character. “Religion, however corrupted, is still better than no religion at all.”  The idea here is the modest one that belief in a future state is a necessary incentive to morals.  Still, Philo is unconvinced. So how is a magistrate to handle popular religion?
Every expedient which he tries for so humble a purpose is surrounded with inconveniencies. If he admits only one religion among his subjects, he must sacrifice, to an uncertain prospect of tranquillity, every consideration of public liberty, science, reason, industry, and even his own independency. If he gives indulgence to several sects, which is the wiser maxim, he must preserve a very philosophical indifference to all of them, and carefully restrain the pretensions of the prevailing sect; otherwise he can expect nothing but endless disputes, quarrels, factions, persecutions, and civil commotions. 
It seems as if Hume has been moved closer to a position like Smith’s, but we should remember that the establishment of one sect did not entail the intolerance of others. Philosophical indifference is certainly not political indifference for Hume, and the History III position was exactly that political favor restrained pretension. One might, therefore, establish a church and at the same time tolerate others. However, Hume never advances this policy unambiguously. If it is to be surmised, we must do so indirectly from his hedged sympathy with Laudianism. As we have seen, though, Hume hesitates to derive overarching lessons from Laud’s policies. In light of such indecision, which I have sketched here as a function of Hume’s “external” hermeneutic, one must conclude against any positive tutelary function for religion. If Whelan is right concerning a “forward facing” element of utility in Hume’s political thought, then religion cannot be a tool in its service. Andrew Sabl has suggested that for Hume, the re-direction of passion induces adherence to an “enlarged,” “proto-Millian sense of fallibilism.” With religion, this means that missionary zeal comes to be associated with tolerance rather than with factional quarrel.  It is evident that Hume does in fact think this happened in the aftermath of the Independents, though the reasons remain obscure. What is more, it may be argued that Conti’s “aftereffects” observation still holds some truth: only when independent enthusiasm had “burned out” did the brutal part of its political relativism subside. With this in mind, “dampening” or “attenuation” seem better choices for describing Hume’s strategy than does “redirection.” Lacking the insight to predict specific developments as the outgrowths of religious doctrine, the safer course in both hermeneutic and practical terms was to restrain any public claims made on its behalf. Duncan Forbes argued that Hume’s political position was fully “institutional,” eschewing Machiavellian (or more proximately, Fergusonian) concerns with public morals.  One need not embrace the whole of this view to consider that it applied in some uneven manner to Hume’s posture on religion. “Popular” religion might damage critical institutions (and perhaps inhibit morals, too), but it did not and could not sustain them. Religious institutions, as would-be moral ones, could not be trusted. Still, this observation fails to resolve Hume’s difficulties. The problem is that his interpretive quandary touches also on the business of restraining pretension; it remains unclear whether establishment is generally the best avenue to achieve such restraint.
At this juncture, I want to avoid the suggestion that Hume’s dilemma – between external interpretation and piecemeal psychologizing – is fully a weakness in his approach. There is something admirable in Hume’s empirical care: his refusal of the historicist frame removes him from the dehumanizing hazards of one species of ideological generalization. Another salutary element is a concern for order and balance in the polity. Moreover, the perceptive analysis of the Independents demonstrates that some political goals unmoored – especially tolerance, freedom, equality – have a dark side. In demonstrating this, however, Hume is forced in spite of himself to take some of independency’s doctrinal claims on their face, adopting their linkages as his own. Without these suppositions, there would seem to be little reason to connect the development of civil liberty to the religion of the Independents, rather than solely to the violence in which they participated.
Interpretive virtues notwithstanding, we must also apportion to Hume’s external historical analysis at least some blame for generating that stubborn, late-modern insistence on treating all religion, and all religious conflicts, as matters of raw emotion. “Religion” as discussed politically has become a monolith, or perhaps a monochromatic scale of moderation and extremity, with doctrine itself irrelevant. Herdt explains this rather powerfully by noting that Hume’s tendency to “causal explanations which sidestep actors’ internal reasons” has the ironic effect of undermining his moral emphasis on sympathy, creating a “faction” resembling the religious ones of which he was critical.  She remarks that Hume treats religion – at least politically – as irrational and self-evidently false, such that it requires special explanation.  Whether one concurs with this assessment of Hume’s or not, I hope to have demonstrated some limitations in the mode of analysis associated with it. Hume assists us in careful investigation, reminding us that sensible politics are often derailed by religion. However, preventing this outcome requires that we augment our reflections with decidedly non-Humean considerations, even if healthy secular politics is a goal. In light of our present (multi) cultural difficulties, we do well to see that there is no avoiding the question of the good at the heart of religion. Civil liberty itself rests on how we answer it.