Thurso Castle

Even the seventeenth-century date needs care. Highland Council’s Historic Environment Record reproduces an Ordnance Survey Name Book description dating the earlier house to 1664, while MacGibbon and Ross placed it in 1660. An engraving preserved by the University of Edinburgh’s Statistical Accounts project identifies the Arch in the castle courtyard as built in 1665. These references need not describe precisely the same stage of construction. The responsible summary is a Sinclair building campaign in the 1660s, associated with George, Earl of Caithness.

That earlier residence was repaired and enlarged in 1806 and 1835, according to the Name Book. Its 1872 account is especially helpful because it describes extensive work still in progress, expected to hide the old building from view. Later architectural recording described the replacement as dating from about 1875. Rather than compressing the whole transformation into a single completion date, the surviving evidence points to rebuilding during the 1870s, under Sir Tollemache Sinclair.

The change was architectural as well as chronological. A large Gothic or Scottish Baronial mansion could borrow the outline of a fortress while serving the expectations of a Victorian country house. From the opposite riverbank, consider the breadth of the surviving elevations and their relationship with the water. The setting explains their visual impact without requiring every tower or opening to have served a medieval defensive function.

The Caithness Broch Project records a visit by former United States president Ulysses S. Grant in 1877. This is a worthwhile local historical association, but it should be attributed to that account: the evidence reviewed here did not include a contemporary record placing Grant inside the castle. His reported visit must not become an invented description of a banquet, conversation or overnight stay.

The same caution applies to destruction. Later accounts mention a wartime mine explosion below the castle, but reliable evidence establishing that incident as the cause of demolition was not found in this review. The local heritage project instead emphasises poor construction and safety concerns, dating substantial demolition to 1952. An Ordnance Survey visit in 1962 recorded that the roof had been removed about eleven years earlier. Together these support major loss in the early 1950s, without proving one simple cause.

The remaining walls are best appreciated from permitted public viewpoints. Their uneven survival is part of the story, not an invitation to explore unstable interiors. Look back across the river before leaving: Thurso’s memorable outline belongs to several successive buildings and to the selective losses of the last century. Keeping those layers distinct makes the view more interesting than an unsupported tale of a single ancient castle.