Don't even think about it...

In 1786, Margaret Nicholson, a 36-year-old needlewoman, drew a knife and attempted to stab George III while presenting him with a petition for his assistance following the loss of her job and her lover. Was this a politically motivated assassination attempt, a fit of madness, or a symbolic expression of popular grievances? These questions become more pressing when there are several such attacks (the ambiguous assaults associated with the Two Acts in 1795 followed by two incidents on May 15 1800 and an epidemic between these and the close of 1802). Increasing frequency makes it harder to dismiss individual instances as fits of madness.

As Steve Poole's book shows, the identification of individuals as mentally unbalanced had the advantage of deflating the potential political significance of such attacks, just as the presentation of the fortitude of the king in the face of such threats contributed to his public standing. Using the Vagrancy Act and subsequently the Hadfield Act (itself named after one of George's May 1800 assailants), it was possible to consign culprits to an asylum for the rest of their natural lives - a simple solution that avoided the problems associated with jury trials for treason. Such disciplining of these "troublesome subjects" fits neatly with accounts of the ritualising and dignifying of the monarchy, but there remains the question of what the assailants thought they were doing.

Even if there was madness, there was also, as Poole shows, a certain amount of method. Mental disorder, religious mania and emotional disturbance may seem more prominent in the final unorthodox approaches to the king - involving stalking, knife-throwing and public suicide - but the attacks were usually the culmination of unrequited petitioning, indistinguishable from traditional practices. Does this moderate the madness of those involved? Does it increase the political dimensions of the act and thereby the significance of the courts' attempts to take such cases out of the domain of treason and into that of insanity?

After a slightly stodgy first chapter, Poole tells a lively story about the way the British monarchy after George III became increasingly insulated from the practice of public petitioning. The book succeeds partly by sustaining the uncertainties about the stability and intentions of those who lobbed various things at the king. Yet its exclusive focus on these extreme "addresses" gives us little sense of the extent of more polite petitioning or of the shift of petitioning from the crown to Parliament. In Britain, unlike in other European monarchies, Parliament provided an alternate focus for popular concerns. By concentrating on the benighted, if intelligible actions of Poole's subjects, there is a danger that we miss the more profound change taking place in the practices of popular petitioning and the associated transformation in people's interpretation of the monarchy.

While there was rhetorical value to the view that one would have to be mad to attempt to harm the king, the government believed that designing individuals associated with popular radicalism in the 1790s had more treasonous intentions. In Imagining the King's Death , John Barrell examines prosecutions for treason instigated between 1793 and 1796. The detail is meticulous and the account magisterial. The statute of treasons of 1351, 25 Edward III, identified treason with the "compassing or imagining" of the death of the king - "imagining" having the sense of purposing or intending.

Barrell shows that the interpretation of the statute became progressively inflated and contested in the 1790s. The established interpretation was that "the law considereth the wicked imagination of the Heart in the same Degree of Guilt as if carried into actual Execution".

Thomas Erskine, defending those charged with treason, agreed that the statute so completely took the will for the deed that a person could not be indicted for killing the king, only for intending so do so, and accordingly based his defence on the absence of factual support for the assertion of treasonous intent.

In the controversies over interpretation in the 1790s, some prosecutors did treat overt acts as matters of fact and as in themselves definitively settling intention (leaving the jury with nothing to judge), while others encouraged a conflation of attacks on the constitution with attacks on the person of the king. Lord Chief Justice Eyre, during Horne Tooke's trial in 1794, enunciated an interpretation in which any attempt to overawe Parliament, because it sought to overturn the laws of the land, threatened kingly authority - and thereby compassed the death of the king. By such a construction, popular conventionism became not merely politically imprudent, but necessarily treasonous. Rival interpretations were fought out with ingenuity and (on the reformers' side) with considerable wit. One irony they were quick to spot was that those who instigated the prosecutions were imagining the reformers imagining the death of the king, allowing them to ask whether it was their own or their prosecutors' imaginations that were the more treasonous. Against the backdrop of events in France, the sentimental imagining of Louis's execution, and wild accusations of regicidal intent among reformers, who could avoid thinking what seemed increasingly proscribed as unthinkable?

Barrell focuses on the deployment of the treason statute against the extra-parliamentary reform movement of the 1790s. He also covers the government's discomfiture over Richard Brothers's millennial imaginings, John Reeves's libel on the constitution, and the farcical "pop-gun" plot. He demonstrates that the character of the reform movements, and the corresponding dynamics of political confrontation and control at the end of the 18th century, were profoundly influenced by the state's attempt to exploit the ambiguities of the original statute and by radicals' attempts to resist these constructions.

Barrell's grasp of legal argument and distortion, his detailed reconstruction of the activities of the reformers and those who sought to restrain them, and his literary eye for ambiguity and rhetorical play make the book a magnificent achievement. It offers a detailed discussion of the way in which the legal apparatus of the British state was deployed in defence of its established practices against the first stirrings of a democratic movement. It delivers a convincing judgement against William Pitt's government for its willingness to resort to the interpretation of treason in cases that two years earlier would have been no more than misdemeanours.

Poole tells the story of a set of apparently isolated events that threatened the death of the king and generally led to the detention of those involved for insanity. Barrell narrates the progress of the extra-parliamentary reform movement whose leaders, while not conspiring against the life of the king, were seen as deserving a place on the executioner's block. The apparent clemency contrasts with the reckless disregard for the reformers' fondness for their heads. There is no doubt that some radicals imagined a life without kings. But it takes a particular mixture of conspiracy theory and arrogance to call this treason. Moreover, the refusal to settle for sedition or riotous assembly charges, while showing a certain tolerance for individuals who attacked the king, suggests that the principal concern of the officers of the state was not the safety of the king but the preservation of the unreformed parliamentary system. The treason statute became a tool for those who sought to assert the sovereignty of the state against the popular extra-parliamentary movement.

The signal failure of the prosecutions, with juries acquitting all those arraigned in the period, save the hapless Scots Robert Watt and David Downie, was rectified after the popular disturbances of October 1795 by the 1796 Bill for the Safety and Protection of his Majesty's Person. Although never formally applied, the act disambiguated the earlier statute to the government's benefit. It closed the space for the kind of contestation engaged in earlier in the decade. Later statutes, such as the acts of 1842 and 1848, were exercises in depriving Poole's "troublesome subjects" of the rights that defendants in treason trials were accorded. In doing so, however, they confirmed the low status accorded to attacks on the person of the monarch, relative to that attributed to extra-parliamentary attempts to reform the established order - providing further evidence that the monarch was becoming an increasingly dignified, if occasionally targeted, part of the constitution.

Mark Philp is head of the department of politics and international relations, University of Oxford.