Please consider registering as a member of the International Spenser Society, the professional organization that supports The Spenser Review. There is no charge for membership; your contact information will be kept strictly confidential and will be used only to conduct the business of the ISS—chiefly to notify members when a new issue of SpR has been posted.

Don’t worry, be happy: reading Herron reviewing O’Keeffe making Spenser envy Norris
by Tadhg O’Keeffe


Informed conversation between architectural-historians and Spenserians on the buildings of Spenser’s world, especially when focused on how those buildings shaped (and indeed were shaped by) the experiences and perceptions of natives and planters alike, can only enrich our collective understanding of Ireland at the end of the sixteenth century. I am genuinely pleased that my paper has started a conversation about the relationship between Edmund Spenser and Thomas Norris as castle-owners. My response to Professor Herron is informed almost entirely by my familiarity with architecture – for good or bad, I said what I have to say about history and literature in the original paper – and I leave it to Professor Herron and his fellow Spenserians to decide at the end what can usefully be salvaged from my analysis.

Kilcolman’s grandeur

Prof. Herron pleads that the medieval castle at Kilcolman was worthy of Spenser, and invokes some great Munster castles (Barryscourt; Newcastle West) in his plea.[1] But, as an architectural historian visiting, thinking and writing about castles for the past thirty-five years, I struggle to get too excited about Kilcolman’s intrinsic qualities. Its tower had nice cut-stone quoins and window surrounds, but that was not unusual in towers built in limestone country. It was not a big tower in terms of floor area. In fact, it was small by County Cork standards (Fig. 1). And, contrary to the opinion of Professor Herron (who regards the tower’s height as a significant indicator of its status), it was not particularly tall. Although I allowed the possibility in 2017 that it might have been five storeys high originally,[2] I think that it might have been only four, with the corner turret rising to an additional storey and overlooking the roof of the main block. My suggestion is based on the observation that the truncated upper walls in many ruined towers demonstrably indicate the original parapet-levels. Upper rooms did not physically fall off towers, leaving nice horizontal levels, nor did those who ‘decommissioned’ towers through partial demolition carefully slice off the upper floors to leave such horizontal levels. Rather, battlements, or crenellations, were taken down or knocked off when castles were attacked and captured, and that is why the ascending walls of ruined tower-houses often stop abruptly. Removing crenellations, the signifying features of fortifications, was possibly a symbolic as well as a practical act of decommissioning a tower; a licence to crenellate one’s house was a mark of one’s status in English medieval culture.[3] By no stretch of the imagination could Kilcolman have been six or seven storeys high, as Prof. Herron suggests at one point;[4] Lohort tower-house, about 10km north-west of Mallow and owned by the Perceval earls of Egmont in the seventeenth century, is a five-storey tower-house (Fig. 2), and it is as high as such towers went.[5] Castlepook, Kilcolman’s neighbour, is not a good example to cite in an argument that Kilcolman was more substantial than I allow in my paper: it is not, as Professor Herron claims, a six-storey tower, but a four-storey tower – admittedly with one storey having a low attic – with a narrow additional storey rising above the roof. 

It is certainly true that we cannot speak authoritatively about Spenser’s alterations to Kilcolman. I doubt, on the basis of what can be seen today, that he had the top of the tower ‘retrofitted with windows’. More excavation would be needed to establish if changes were made to the original medieval hall. The best place to see what altering a hall might have entailed is Castlemartyr, County Cork (Fig. 3). The one change at Kilcolman that we can attribute to Spenser is the structure which Eric Klingelhofer discovered in excavations between the tower and the hall; I argued in 2017 that it could be regarded as a ‘presence chamber’[6] but Professor Herron follows Professor Klingelhofer in describing it as a parlour. I am sure that it was nicely fenestrated, but it was relatively small; it was probably only one storey in height. My two colleagues believe it to have been a half-timbered structure. Is it possible that Spenser’s half-timbered addition made Kilcolman look like an English yeoman’s house?

The chivalrous and the fortified

Turning now to Mallow, I need to begin by reclaiming from misrepresentation two of the points about that castle which I made in my ‘Home truths’ paper. First, I do not suggest, as Professor Herron asserts in his very first paragraph, that the drawing of a house on the plat by Arthur Robins ‘was a direct model’ for the design of Mallow. On the contrary, I specifically point out that the drawing on the plat is not an architect’s drawing, and that it and the castle must have had a common source, communicated in a manner – an actual architectural drawing, probably – that would have allowed the same basic form to appear on a plat and in an actual building. Second, and rather less critically, Professor Herron draws attention to Lord Chancellor Hatton’s now-lost, pre-1600, house at Cappoquin, County Waterford, in the context of taking me to task for describing the Norris house as ‘the oldest surviving residence built ab initio during the Munster Plantation, and the only one to have been built before 1600’. My point was that Mallow is the only surviving pre-1600 house. I thought the phraseology was clear; apparently it was not.

Professor Herron’s suggestion that the Norris family home of Rycote was the model for Mallow – ‘the Norrises may have been copying their own house’ – might seem at first glance to be supported by antiquarian images of the since-demolished mansion in Oxfordshire. Winstanley’s mid-eighteenth-century drawing of Rycote’s frontage shows polygonal corner turrets and front-facing gables not at all unlike those at Mallow. Indeed, I would add, perhaps further undermining my own argument, that the manner in which the string-courses ran under the gables at both houses is another parallel. The upshot of Professor Herron’s analysis of the Rycote-Mallow axis, were it correct, would be to deprive Sherborne of the prominence which I give it. Professor Herron makes the point in a striking and thought-provoking conclusion: ‘Whereas Sherborne in England was a fanciful reflection of Raleigh and Spenser’s romance-epic ‘chivalric’ aesthetic, Mallow in Ireland was a fortified Rycote’.

However, I want to suggest respectfully that Professor Herron’s understanding of the meanings of, and of the relationships between, the three houses – Mallow, Rycote and Sherborne – is problematic, and needs to be addressed because of its implications for how Spenserian scholarship understands Kilcolman.

Rycote-Mallow reconsidered

The comparative method in architectural-historical enquiry naturally privileges comparisons, so some derivation of Mallow from Rycote makes sense based on Professor Herron’s observations. But one must also explain differences, and the differences between Rycote and Mallow were significant. The former was a courtyard house with domestic accommodation in ranges, whereas the latter is a single-pile building. The gables at Rycote, for example, were not attic gables like those at Mallow but were the end-wall gables of suites of domestic spaces. Therefore, although design motifs such as polygonal corner towers were easily copied from one house to another, a Mallow-type house could not have been produced from a Rycote-type template.

Were the polygonal towers, at least, copied at Mallow? It is hard to believe that they were, notwithstanding the family connection between the houses. The Rycote towers were octagonal whereas those at Mallow are pentagonal. Why does this matter? The two Rycote towers were simply self-contained units of architecture framing the entire house frontage, and their octagonal shape, made by cutting the corners of a square, was the simplest polygonal shape to design. Towers of similar shape, size and relative proportion framed the Rycote-like frontages of other sixteenth-century English courtyard houses, such as Kentwell Hall in Suffolk (Fig 4), Hengrave Hall, also in Suffolk, Cobham Hall in Kent, Gorhambury House in Hertfordshire and so on. The two Mallow flankers, by comparison, as well as the angled half-tower at the front of the house, were planned in alignment with corners of the building (Fig. 5a) in a manner that clearly imitated the flanking-fire determinations associated with Renaissance star-shaped artillery forts and houses (Fig. 5b). The distinction is important: one does not need to look outside the English architectural tradition to understand fully the Rycote towers and their relationship to the frontage in which they are set, but one cannot explain the Mallow composition without reference to the ‘scientific’ architecture of fortification in Renaissance Europe. Mallow is not a copy of Rycote, even in spirit.

Mallow-Sherborne restated

At the time of writing the ‘Home truths’ paper, I had not yet come upon the (rather scratchy) drawing of c.1600 of Sherborne which was made by Simon Basil, Surveyor of the King’s Works from 1606 to 1615.[7] I have redrawn it here for clarity, as best I can (Fig. 5c). It is exactly the type of drawing which allowed designs to be transmitted from place to place. In its case, it is a plan of the standing building at Sherborne rather than a blueprint for a new building.[8] An important addition to the book of evidence, it firmly reinforces the link which I suggested between Raleigh’s Dorset house and Norris’s Cork house.

First, Basil drew the imaginary firing-lines by which the angles of its pentagonal towers were determined (accentuating them with little circles), and those lines are similar to the lines which gave Mallow its shape. The completed Sherborne certainly looked like other four-turreted houses, such as late fifteenth-century Wickham Court in Kent, or the residential gatehouse of c.1580 at Tixall in Staffordshire (Fig. 6), but its plan reflected – fairly uniquely for its date – the influence of artillery fortifications. Second, the tower on the left-hand-side of the front of Sherborne had a service stairs, as had the tower in the same position in Mallow. Third, the main stairs in Sherborne was abutted by a cross-wall on its right-hand-side at the back of the building; Mallow had the same arrangement, differing only in having the stairs projected outwards in a turret.[9] I know of no house in these islands – there is certainly none in Ireland – which resembles in overall plan and stairs-arrangement either one of these two houses even more than they resemble each other. I now realise that, just as the largest house on the Robins plat looks like Mallow, the second-largest house on the plat looks like Sherborne. The evidence of the plat indicates that the plan-template common to Mallow and Sherborne was in Munster at the start of the plantation. Its execution in Mallow suggests to me that it was designed as an ideal for Munster, just as the plat shows us Munster’s idealised plantation topography.

So, was Spenser happy?

Professor Herron’s critique of my ‘Home truths’ paper deals two substantive issues. One is Rycote’s relationship with Mallow. I have argued in this response that the similarities between the two residences are too superficial for the former to be regarded as a model for the latter. I believe that Mallow Castle must still be seen as a very close relative of Walter Raleigh’s house in Sherborne, and that the two structures, combined with the drawings of houses by Arthur Robins, reflect an original architectural vision for the Munster plantation, paralleling the topographical vision represented on the Robins plat.

The second issue is Kilcolman as a building and a locus. For me, Spenser lived in a modest house in a relatively isolated location, conscious that it was a modest residence by comparison with that of his neighbour, Sir Thomas Norris, and fully aware that it was vulnerable. For Professor Herron, Kilcolman had grandeur as well as accessibility, and it therefore deserves to be regarded as something more than a ‘bare raggedy stump in the middle of nowhere, next to a forsaken bog’.

I recognise from Professor Herron’s critique that I left a possible hostage to fortune in ‘Home truths’ when I made a throwaway comment about Spenser’s state of mind as he went to bed at night. But I am not inclined to retreat from the general sentiment. During Spenser’s time in Kilcolman both Norris and Raleigh started to build new houses to replace old castles, which is something that he did not do, possibly because he did not have the resources. He knew the new house in Mallow to visit and probably knew of Sherborne. Norris’s house was bigger and more modern than Kilcolman, and it had an enviable pedigree: it was designed by, or was derived from a design by, an architect in the Renaissance sense of that word. I suspect that Spenser knew this. It is possible that he saw the plan from which Mallow was built; Norris or Raleigh might even have owned, and showed him, one of the foreign-published manuals of architecture by which the ‘star-shaped’ principle of Renaissance fortification was disseminated. Given what they built, and given the possibility (discussed in ‘Home truths’) that the Robins plat was actually in Youghal, Norris and Raleigh must have talked about architecture and architectural drawings, certainly to each other but perhaps also in Spenser’s company. Professor Herron writes that Kilcolman’s native architecture, ‘outside of’ Spenser’s new half-timbered addition, was ‘far grander than any half-timbered secretary’s hovel in London or in Kent’. Maybe so, but was that the competition? Would Norris or Raleigh have been impressed by Spenser’s small, half-timbered, ‘parlour’/great chamber?

Perhaps I was wrong about the effect of the narrow, spiralling, stone-built stairs on Spenser’s mood when he went to bed at night. Perhaps he was satisfied that his alterations to Kilcolman succeeded in giving it an Elizabethan ambience. Still, I imagine that his mood was a little wistful whenever he returned to Kilcolman having visited Thomas Norris in his new mansion. And that his sleep was sometimes disturbed by noises in the night…


Tadhg O’Keeffe




Fig. 1:  External dimensions of a sample of sixty towers in Co. Cork; Kilcolman is indicated by the black dot.




Fig. 2:     Lohort Castle, Co. Cork.  A row of small, square, gutter holes directly above the uppermost flat-headed window represents the level of the original parapet.



Fig. 3:     Castlemartyr Castle, Co. Cork.  The three chimneys were inserted into the hall in the 1610s or 1620s, an alteration which necessitated removing the original roof and designing a new roof. The tower-house at the far-end of the hall was taller than the tower at Kilcolman.



Fig. 4:     Kentwell Hall, Suffolk, in the early nineteenth century (undated print, engraved by John Hawksworth).


Fig. 5:     (a) Plan of the ground floor of Mallow Castle showing with dotted lines the alignments of angled faces of the towers at the front of the house; (b) an unscaled ‘star-shaped’ Renaissance palazzo published in Le fortificationi di Buonaiuto Lorini, Venice 1609: 167; (c) Simon Basil’s plan of Sherborne, c.1600, redrawn.


Fig. 6:     Tixall gatehouse, Staffordshire (after Francis Grose, The Antiquities of England and Wales 5, London 1784: 44).



[1] I would cite these and other regionally-important later medieval castles in support of my assertion in ‘Home truths’ that Tudor culture had not penetrated Irish court circles very deeply by the end of sixteenth century, notwithstanding the English contacts of the Old English earls.

[2] Tadhg O’Keeffe, ‘Kilcolman Castle, Co. Cork: a new interpretation of Edmund Spenser’s residence in plantation Munster’, International Journal of Historical Archaeology, Vol. 21 (2017): 223-39, at 231-2.

[3] Charles Coulson, ‘Hierarchism in conventual crenellation. An essay in the sociology and metaphysics of medieval fortification’, Medieval Archaeology 26 (1982): 69-100; for an alternative view see Colin Platt, ‘Patterns in licences to crenellate’, The Castle Studies Group Journal 23 (2009-10): 232-33.

[4] ‘Spenserian ambitions’, footnote 11.

[5] Tadhg O’Keeffe, ‘Lohort Castle: medieval architecture, medievalist imagination’, Journal of the Cork Historical and Archaeological Society 118 (2013): 60-70.

[6] ‘Kilcolman Castle, Co. Cork: a new interpretation’.

[7] Nicholas Cooper, Houses of the Gentry, 1480-1680 (London and New Haven: Yale University Press, 1999), Fig. 22. For Basil see Howard Colvin, A Biographical Dictionary of British Architects 1600-1840, 3rd edition (New Haven and London: Yale University Press, 1995): 108.

[8] It must be noted that Ander Gomme and Alison Maguire wrote that the plan of Sherborne was produced by Basil in order to add the pentagonal turrets to an existing rectangular block of 1592-94; see Design and Plan in the Country House: from Castle Donjons to Palladian Boxes (New Haven and London: Yale University Press, 2008): 59-60. They offer no proof, however, and Basil’s drawing suggests that the corner towers were original.

[9] A curious feature of Mallow is the placement of toilets in the angle beyond the stairs. The arrangement has parallels in later, Georgian, architecture.


  • Call us 6 months ago

    Neither did those who "decommissioned" towers by partial demolition meticulously cut off the upper floors to leave such horizontal levels; upper rooms did not physically fall off towers, leaving attractive horizontal levels.

    Link / Reply

You must log in to comment.


Cite as:

Tadhg O’Keeffe, "Don’t worry, be happy: reading Herron reviewing O’Keeffe making Spenser envy Norris," Spenser Review 49.2.5 (Spring-Summer 2019). Accessed April 14th, 2024.
Not logged in or