The castle of Dineirth (or Dinarth) was probably founded by Richard de la Mare, a follower of Richard de Clare in 1110. Annes Cambriae 175. B.sa. 1130 mentions "castell de Ricardi deMare". It was most likely razed in Gruffydd ap Rhys' devestation of 1116 and destroyed again by Owain Gwynedd in 1136.In 1143 it passed to Hywel and in 1144 back to Cadwalader following inter tribal strife.
In 1158 it was among those castles ceded to Earl Roger of Hertford who garrisoned it. Destroyed again, this time by Rhys in 1164, it came into the possession of Maelgwyn who lost it to, and then recovered it from, his brother in 1199.
In 1202 it was dismantled by Maelgwyn to prevent his lands being held by Llewellyn. However Llewellyn succeded in capturing and holding all the territory to the Aeron and conferred Dineirth on the sons of Gruffydd ap Rhys.
It would appear that the dismantling of 1202 was final and that the site was then abandoned.

The site chosen for the castle was a narrow precipious promontory which juts out into the wooded gorge of the River Arth. The Normans may have found the earthworks of a small promontory fort which they heightened and strengthened. Two ramparts and dry moats of formidable proportions cut off the tip of the promontory from the level of the rising ground to the south. On the west the slope to the Arth is steep and is further defended by scarping, while on the east there is a cliff of varying height with a precipitous slope below it to the river.
A large large round motte was piled up in the north east half of the small bailey against the edge of the cliff. There appears to have been no intervening ditch between it and the bailey, but to the north where the top of the peninsular slopes down at a
slightly less steep pitch is a short straight fosse, beyond which is a short broad rampart or platform.
Below this and separated by traces of a slight fosse is a large semi-oval platform or horn work defended by scarping.
The motte appears to be partly natural; the summit is rather disturbed but shows traces of of depressions and platforms. A number of loose stones lie on the slopes, and on top are numerous loose stones which appear to be part of the construction material of the mound and of some structure that it supported.
The defences of the castle do not seem to have progressed beyond the earthwork and timber construction stage.
South of the motte and close to the cliff face is a curious square
mound, partly cleared away, which may have been a continuation
of the inner rampart.
The entrance was between this and and the termination of the rampart proper close against the edge of the cliff. It passed by means of a causeway over the fosses and by a straight cut through the outer rampart. It is traversed by a path of recent construction and some hedge banks of contemporary date rather confuse the detail of the outer entrance. There does not appear to have been an outer bailey which is rather unusual.