Irish Court of Appeal

 

THE COURT OF APPEAL

CIVIL

[approved]

[no redaction needed]

Court of Appeal Record Number: 2023/157

Court of Appeal Neutral Citation: [2024] IECA 199

High Court Record Number: [2023] IEHC 25

 

Whelan J.

Costello J.

Pilkington J.

 

BETWEEN/

 

 

RGRE GRAFTON LIMITED

 

 

PLAINTIFF/

APPELLANT

 

 

- AND –

 

 

BEWLEY'S CAFÉ GRAFTON STREET LIMITED & BEWLEY'S LIMITED

 

DEFENDANTS/

RESPONDENTS

 

 

JUDGMENT of Ms. Justice Costello delivered on the 31^st day of July 2024 

Introduction  

1.             In Bewley's Café on Grafton Street in Dublin there are six stained glass windows which were created by the well-known stained glass artist Harry Clarke in 1928 [1].  The issue in this appeal is: who owns the Harry Clarke windows?  The appellant, who holds the landlord's interest in the premises, says that it is the owner of all six windows.  The respondents say that they were tenant's fixtures and thus were the property of the first named respondent who holds the tenant's interest in the premises and now belong to the second named respondent pursuant to a Deed of Assignment from the first to the second named respondent.  The High Court (McDonald J.) held that that appellant was the owner of four of the windows on the western wall, referred to as the Four Orders windows, as they depicted four orders of classical architecture, and the second named respondent was the owner of the remaining two, which were referred to as the Swan Yard windows, as they were set in the openings which overlooked this yard.  The appellant has appealed the order in relation to the two Swan Yard windows and the respondents have cross-appealed in relation to the Four Orders windows. 

2.             The central plank of the appellant's claim is that the works comprise windows which form part of the leased premises and on that basis they are, as a matter of law, the property of the landlord. 

3.             The respondents contend that the works are not windows and say that they are decorative and ornamental panels and are not part of the fabric of the building.  They maintain that the works have been in the ownership of the tenant at all material times and that they are tenant's fixtures and that, accordingly, they cannot be the property of the landlord. They contend that there were sashes glazed with clear glass in the window frames which functioned as the windows of the building and the sashes containing the stained glass panels were mounted in frames set immediately inside in what was described as a double fenestration system. They say that a minute of the board of the first named respondent dated 9 March 1928 shows that it paid for the Harry Clarke windows which reinforces the respondents' case that they are tenant's fixtures and that the second named respondent is now the owner of all six works.

4.             The contemporary evidence was incomplete, sparse and frequently inconclusive. Expert witnesses on both sides, and in turn, the judge, were required to infer what the original arrangement may have been from this evidence and the evidence from the state of the building and the windows in the 1970s, 1980s and 1990s when it was accepted that the original arrangement had been altered to a considerable but contested degree. It thus is vital to distinguish between direct evidence, accepted common fact, and inferences from direct evidence or commonly agreed fact in determining the issues presenting in this appeal.

5.             There was little dispute between the parties as to the applicable legal principles, which I shall consider later in this judgment. During the appeal the debate largely turned upon the evidence which could prove or disprove the theories of the respective parties as to the original configuration of the building and the approach of the judge to this evidence.

The facts    

6.             The facts have been set out in very considerable detail in the judgement of the High Court and I do not propose to repeat McDonald J.'s excellent statement of the facts and description of the café and the works in dispute. This judgment should be read in conjunction with that of the High Court.

7.             The major difficulty in this case is the incomplete nature of the evidence and the resulting difficulty in resolving the legal issues presenting in the case, a problem of which the trial judge was acutely aware. Of necessity the High Court, and this Court on appeal, is required to draw inferences from this incomplete picture. It need not be emphasised that great care must be taken in such a task.

8.             On appeal, the appellant focused its case to the issue whether the works were windows and thus part of the fabric of the building. If they were originally windows, it said, they could not have been tenant's fixtures. If they were not originally tenant's fixtures, the changes which occurred in the café in the intervening years did not transform them into tenant's fixtures. The respondents' case in the High Court was that the works were installed as and remain to this day, tenant's fixtures.

9.              I therefore propose to start by considering the evidence regarding the construction of the café, such contemporary documents as exist, the lease of the café and the subsequent changes to the café with a view to determining what the evidence available establishes and what inferences may be drawn from this evidence as each side's case turns upon what they contend was the legal position when the windows/works were placed in the building.

Assembling the site and constructing the building

10.         Between 14 December 1925 and 28 July 1986 Ernest Bewley acquired by assignment the leasehold interests in the premises at 78 and 79 Grafton Street and 2, 3 and 4 Johnson's Court.  He thus established a site upon which he commissioned the construction of the building of the café.  As was observed by the High Court, and not contested by the parties, it appears to have also been Ernest Bewley's intention to let the café premises to Bewley's Oriental Cafés Ltd., a company then owned by members of the Bewley family.  At the time Ernest Bewley was not only the first landlord, but he was also a director and chairman of Bewley's Oriental Cafés Ltd. (Since that time it has undergone a change of name and of ownership and is now known as Bewley's Café Grafton Street Ltd., the first named respondent). 

11.         It is common case that the premises was developed by Ernest Bewley. As McDonald J. makes clear, he intended that the café would be comparable to contemporary continental oriental cafés.  The Grafton Street façade is very elaborately decorated in the Egyptian Revival style with a bronze balustrade. The Johnson's Court façade is also decorated, though in a less elaborate design. The interior was in keeping with the exterior: there was evidence that Joshua Clarke & Sons, the firm founded by Harry Clarke's father and carried on by him at the time, supplied stained glass panels as well as other clear glass for use in both the glazing and the decoration of the premises. As the developer of the building, Ernest Bewley was responsible for commissioning what was and remains to this day, a striking building.

Harry Clarke's correspondence    

12.         As described by the trial judge, some of the correspondence written by Harry Clarke in relation to the project survives, but it is incomplete.  It comprises solely material emanating from Harry Clarke and his family firm.  Most importantly, the responses from the two firms of architects involved in the development are not available and the architects' instructions to Harry Clarke and the correspondence from their client (or clients) to them are unavailable.  The trial judge was unable to form any reliable view as to whether the stained glass works were commissioned by one firm or the other and he also noted the apparent confusion on the part of Harry Clarke as to the identity of his client. 

13.         The earliest relevant correspondence is dated 7 January 1927 from Harry Clarke to McDonnell & Dixon Architects, in which proposals for "stained glass and leaded work" were enclosed.  On 26 January 1927 the firm's order book described an entry for 49 panels in the front shop of Bewley's Café, Grafton St., with sheet glass borders, antique and cathedral (types of clear glass), and opalescent glass.  This established that the firm provided glass, including stained glass, in connection with the development of the premises in addition to the six disputed windows.  In particular, it supplied leadlights for a lantern which the trial judge held referred to a roof light to be installed in the café.  The order book for 30 March 1927 referred to the supply of glazing to the premises including the glazing of a sash under the stairs and the supply of white cathedral glass and two large side windows.

14.         On 4 April 1927 the firm wrote to McDonnell & Dixon sending "suggestions for dealing with the large openings at back of Café" and proposing that the details "would not be the same in any of the six windows".  This is the first reference to what were to become the six disputed windows.

15.         The suggestions put forward on 4 April 1927 were not accepted because on 13 April 1927 Harry Clarke wrote to McDonnell & Dixon saying:

"I enclose herewith amended design for stained glass at above.

 

Regarding the filling of the windows looking on Swan Lane with a different scheme of glass, I am convinced that this would not be successful, and think that if you and Mr. Bewley decide, that the expenditure would be too great with this present suggestion, that a simple scheme might be evolved for the six windows in question."  

16.         The suggestion of a single scheme for all six works was rejected. Joshua Clarke & Sons wrote again to McDonnell & Dixon on 20 April 1927 enclosing "our designs with alterations in accordance with your suggestion".  Two of these designs remain in the Bewley's archive, the design for the Ionic Order work and also for the Swan Yard works.  As the trial judge observed, "The drawings are largely consistent with what can be seen today in the finished stained-glass work".  The designs are divided into sashes or panels consistent with what can be seen in the café today, though they do not show the opening mechanism which was subsequently installed.

17.         On 6 May 1927 the firm wrote to the second firm of architects, Messrs. Millar & Symes, in relation to the commission.  The letter is headed "Mr. Bewley's New Café Grafton St" and states:

"In accordance with the verbal instructions received we are about to proceed with the stained glass windows at the café for the above account in the sum of... £390...

 

As the full sized drawing will entail considerable work it will be some time before it will be ready for inspection. We would be greatly obliged if you will kindly confirm this commission."  

As the trial judge observes, there is nothing to explain why this correspondence is addressed to Millar & Symes as opposed to McDonnell & Dixon.  The firm wrote again on 7 June 1927 to Millar & Symes seeking confirmation of their instructions.  This appears to have been confirmed as the order book for 7 June 1927 shows two entries for Messrs. Millar & Symes, entries number 1536 and 1536A.  Number 1536 is:

"Make Supply Fix Three windows back of [the] Café 

Stained Glass for Messrs Bewleys Café, Grafton St as Est Proposal No. 3

6^th May £210."

18.         Order book entry no. 1536A reads:

"2 Windows of Side of Café No. 4 Proposal 

Same Manner as No. 3 £180

1 extra window No.3 [£]70."

19.         There is a gap in the record until 6 September 1927 when Harry Clarke wrote to "Bewley Esq." at Westmoreland Street, Dublin, in the context of Civic Week from 17 to 25 September 1927 as follows:

"Dear Sir,

I beg to send herewith a copy of the matter I proposed using with your kind permission small show cards, which you were kind enough to consent to hand in your Café.

I would be greatly obliged if you will let me know that this has your approval.

Thanking you for past favours.

 

Yours faithfully,

[Signed] Harry Clarke

 

Copy of Notice.

 

The Stained glass windows for our New Premises in Grafton Street are being designed and made by J. Clarke & Sons, at whose Studios, 6/7 North Frederick Street, these and other Works of Art may be seen in [the] process of being made."

20.         Work on the stained glass proceeded. On 7 October 1927 Harry Clarke wrote to T. Dixon Esq. of McDonnell & Dixon updating him in relation to the progress where he confirmed that "[w]e are working on the stained glass windows" for Bewley's Café. 

21.         On 9 February 1928 Harry Clarke wrote to McDonnell & Dixon referring to an error in his estimate of 20 April 1927:

"In these Proposals I estimated for three (3) windows at £210, whereas there are in all four windows, the mistake happened through my going to the building in person when in process of erection, and I think you will agree with me, and I hope Mr. Bewley also, that this work at £70. [p]er window which I have done is reasonable, in the cost to date the work on the four windows is £233. so that even if you and Mr. Bewley consent to allowing me additional £70. [t]he margin of profit will be very narrow indeed."

22.         The trial judge observed that this letter very clearly indicates that Harry Clarke had visited the premises while in the course of construction and that it demonstrated his involvement in the process at a relatively early stage.  He remained personally involved in the fixing of the works, as is clear from his letter of 2 March 1928 to Mr. Dixon of McDonnell & Dixon:

"I visited the above on last Wednesday and saw [the] final window of the first set erected in position and it seemed to me, and I hope you and Mr. Bewley will

agree the work is satisfactory."  

23.         He also sought confirmation regarding the payment of the additional £70 for the additional fourth window and asked:

"...if this can be settled I would ask you to kindly issue certificate for the four windows and the balance due £66.15.0 on the leaded work."

24.         The "leaded work" is separate to the six stained glass windows and is a reference to work carried out in respect of the construction of the premises and was for the account of the landlord, Ernest Bewley.  On 13 March 1928 Harry Clarke sent statements of accounts repeating his request for payment of the two items.  He wrote again on 30 April 1928 to Messrs. McDonnell & Dixon seeking payment for the balance of account for leaded lights at Messrs. Bewley's Café, Grafton Street in the sum of £66.15 and also:

"To Designing making supplying and fixing stained glass for 3. windows at back of ground floor Café at Messrs. Bewley's premises, Grafton St. as per Est 20^th April 1927. £210

 

To Designing making supplying and fixing stained glass for 2 windows at side of Café as per Est. 20^th April 1927. £180.   

 

To Designing making supplying and fixing one extra  window for windows at back [of] ground floor Café not included in above. £70. 

_____________ 

Total: £526.15.0

25.         It is unclear when the two Swan Yard stained glass windows were fixed though it appears to have been by the end of April 1928.  Payment in respect of these windows was sought on 30 April 1928. Thus, the six disputed windows were installed by spring 1928.

26.          There appears to have been some delay in discharging the account because on 7 June 1928 Harry Clarke wrote to Charles Bewley Esq., Messrs. Bewley's Café, Westmoreland Street, Dublin, as follows:

"Dear Mr. Bewley,

 

At the request of W. Dixon, Esq., Architect. I have pleasure in enclosing herewith our account for the Stained Glass work at your New Café in Grafton Street.  I would be greatly obliged for your cheque, and understand from Mr. Dixon's Office that issue of certificate is unnecessary."

He was mistaken as to Mr. Bewley's given name, as there was evidence that Charles Bewley was at that time in Tasmania and the reference ought to have been to Ernest Bewley. The statement of account furnished to McDonnell & Dixon on 30 April 1928 was enclosed in the letter. 

27.         On 22 June £200 was paid in cash on the account and Harry Clarke sought the balance of £326.15 from Charles Bewley on 7 August 1928.  On 21 September 1928 he wrote to Charles Bewley Esq. stating:

"Dear Mr. Bewley,

I am greatly obliged for your cheque, £326.15, in settlement of our account, and have pleasure in enclosing herewith receipt."

Unfortunately, the receipt does not survive. 

28.         The last letter is dated 5 October 1928 and it is from Harry Clarke to Charles Bewley Esq., Westmoreland Street.

"Dear Mr. Bewley,

I hope you will accept these two designs for your window as a souvenir of our pleasant business relations".   

The framed drawings were enclosed. 

 The board minutes of the company from 9 March 1928

29.         An important record from 9 March 1928 upon which both parties relied is the board minutes of the company held on that date at 78/79, Grafton Street.  Mr. Ernest Bewley was in the chair and also present were Mr. T.W. Bewley and Mr. G.A. Overend, the secretary.  I set out the minute in full.

"The notice convening the meeting was duly read.  The minutes of meeting held on 30^th June 1927, were duly read and signed. 

 

A letter from Mr. Ernest Bewley to his Fellow Directors was read, referring to expenditure on 78-79 Grafton Street, Dublin.

 

Mr. G.A. Overend, Solicitor, attended and went through the proposed lease of the Grafton Street and Johnson's Court premises, with the Directors, and explained the various provisions therein.  The question was then raised as to the date from which the lease should run, and as to the rent which should be paid, and Mr. Overend explained that in as far as the Landlord was concerned, the premises were finished before April 1927, and that the work done thereafter consisted of fittings necessary to adapt the premises for use as a café, and which expenditure would in due course appear in the Company's accounts, as a Liability of its own.  He further explained that Mr. A.W. Shea of Messrs. Dockrells, had reported that the true basis in which the rent should be calculated was pto on the total cost to the Landlord, plus such ground rent as the Landlord is liable for.  Mr. Ernest Bewley then stated that having gone into the figures, he had ascertained that the total cost to him as landlord was about £45,000.0.0 paid up of purchase prices paid by him for the various buildings, and the contractor's accounts, and that over and above this figure, he had spent, or authorised the expenditure on the company's behalf, of a sum of £14,900.0.0 largely comprising the cost of ovens, tables, chairs, counters and other café fixtures and fittings.  The Board then resolved, in view of this explanation, that the rent of £3,500.0.0, and the date of commencement of same, namely 1st April 1927, seemed reasonable, and approved of the lease and counterpart, and authorised the seal of the company to be affixed thereon."  

 

Ernest Bewley

20.4.27[sic]"

30.         Mr. Overend, the secretary of the company, appears to have acted as solicitor for both parties.  Secondly, only two directors were present, Mr. Ernest Bewley and Mr. T.W. Bewley.  Thirdly, the letter from Mr. Ernest Bewley to his fellow directors referred to in the minutes has not survived nor the report of Mr. Shea of Messrs. Dockrells.    

31.         The minutes record that Mr. Overend explained that from Mr. Ernest Bewley's perspective (1) the premises were finished before April 1927, (2) that the work done thereafter "consisted of fittings necessary to adapt the premises for use as a café", (3) the company would be liable for this expenditure, (4) this expenditure would in due course appear in the company's accounts, (5) Mr. Ernest Bewley said that the total cost to him as landlord was about £45,000 made up of the purchase price of the site and "the contractor's accounts", (6) Mr. Bewley had spent or authorised the expenditure on the company's behalf, of the sum of £14,900, (7) the sum of £14,900 comprised "largely" the cost of ovens, tables, chairs, counters and "other café fixtures and fittings". 

32.         The implications of all of these elements in the minutes will be considered further in this judgment.

The lease of 1928

33.         On the same day, 9 March 1928, Ernest Bewley entered into a lease with the company for a term of 21 years commencing on 1 April 1927 for a rent of £3,500 per annum.  The tenant covenanted to "well and sufficiently uphold maintain repair and keep the interior and exterior of the said premises...in good and substantial repair and condition and...at the expiration or sooner determination of the term hereby granted deliver up to the [landlord] together with all fixtures...".  The lease makes no reference to tenant's fixtures and does not refer to the Harry Clarke stained glass windows.   

Real Evidence

34.         The Harry Clarke windows were designed to fill six opes in the café to the rear of the premises.  There are six rectangular window frames each designed to take four sashes. They are removable by simply unscrewing the sashes without damaging the building as appears from a video filmed in 1998.

35.          The sashes are hinged at the bottom to allow the windows to open inwards to ventilate the café, by what is known as a hopper mechanism.  There is a photograph of the windows opening inwards in this fashion which was admitted into evidence which confirms that the windows originally provided ventilation for the café.  The window frames and sashes were fixed with a complex brass mechanism which allowed the three upper sashes to be opened in unison. The trial judge said that "the works were fitted with a brass hopper mechanism which allowed the top three sashes to be opened by manipulating a brass rod which ran down the length of one side of each of the works. This appears to have been quite a sophisticated mechanism. It comprised a vertical brass rod to the side (which was used to control the mechanism) which was, in turn, connected to the hinged rods on each side of the window frame by means of a slender horizontal brass rod. By manipulating the handle on the side rod, the three sashes would open inwards in tandem." The lowest sash of each window opened separately in the original arrangement. The hopper mechanism was removed at some time prior to 1998 and, while it no longer operated to open the sashes, pieces of it still exist and clearly illustrate the mechanism.  It was not disputed that this formed part of the original scheme for the windows. 

36.         The existing window frames for the Four Orders windows are flush against the bullnose bricks forming the opes of the four windows.  The plasterwork is both old and undisturbed. This can be clearly seen in close up in a video taken in 1998 which records the removal of all six window sashes for renovation works.  Mr. Slattery, the expert called on behalf of the appellant, gave evidence that this indicated that the window frames are original and have not been disturbed since they were originally installed.  The trial judge accepted this interpretation of the evidence from the video and held that there was nothing to suggest that the window frames in which the individual sashes of the Four Orders works sit were ever moved and that they appear to be in their original position.  The Four Orders stained glass sashes sit in these original window frames.

37.          Mr. Slattery gave evidence that the outer side of the frames of the Four Orders works had been painted and carved in a manner that suggested that they were intended to act as part of the exterior of the premises.  The trial judge noted that this evidence was not contested.  Further, Mr. Slattery opined upon viewing the video from 1998 that there was evidence of weathering on the exterior facing side of the frames and the sashes of the Four Orders works. The trial judge concurred stating "One can see from the video that there are marks on the external side of the frames of the sashes which Mr. Slattery said were consistent with weathering".  

38.         There were originally two opes facing onto Swan Yard.  One has been blocked up for many years, certainly since before 1986. There was no evidence as to why this change was affected. No window frame in respect of this opening survives.

39.          The other window frame contains sashes glazed with clear glass.  Mr. Brian O'Connell, who gave evidence on behalf of the respondents, examined the frame, the sashes and the surrounding plasterwork in detail.  He was of the view that the mortar joints between the brickwork shows the original construction.  His view was that if the window frame had been replaced, he would expect to see the mortar cut back and he would also expect to see a new mortar joint.  He examined the joint on the inside of the window frame where the plaster reveal meets the timber window frame, and he gave evidence that there was no indication that the plaster reveal had ever been replaced.  The trial judge accepted that the window frame was the original frame inserted into the window during construction.

40.         Mr. O'Connell said that the clear glass sashes in the Swan Yard window frame were also original and therefore must have been installed at the same time as the window frame in 1927. On the basis that the frame was glazed with clear glass sashes which he said were also original, he theorised that the stained glass sashes which were "fixed" to the building in 1928 must have been installed other than in the original frames. In 1986 the two Swan Yard windows were fixed to internal window frames and were back lit. One of them was in front of the blocked up opening and the other mirrored the remaining external actual window. He concluded that this arrangement must reflect the original arrangement which he described as double fenestration.

41.         However, he could not explain how this system would operate to ventilate the café, which all witnesses agreed was "a significant requirement". Most significantly there were no nail or screw marks or other evidence on the Swan Yard timber frames or sashes to indicate that a mechanism existed either to open the clear glazed window sashes or to connect the clear glass sashes and window with any internal equivalent containing the Harry Clarke stained glass sashes. The absence of this evidence suggests that there was no hopper mechanism on the clear glass sashes equivalent to the mechanism which existed on all of the stained glass sashes. Mr. O'Connell could not explain how the Swan Yard windows could ventilate the café if there was no mirror hopper mechanism which worked in tandem with the stained glass mechanism. The absence of this physical evidence of a mechanism for opening the sashes when coupled with the inability to explain how the opes could ventilate the café if the double fenestration postulated by Mr. O'Connell existed is significant in the resolution of the issues in this appeal.

42.         Mr. O'Connell gave evidence which was not contested that the clear glass sashes in the Swan Yard window and the stained glass Swan Yard sashes were the same size. This meant that the two sets of sashes were interchangeable.

Subsequent events affecting the stained glass windows

43.         On 27 November 1927 the café opened. This was prior to the completion and installation of the stained glass windows. It is common case that six windows must have been glazed at the time, though there is no direct evidence to that affect. The stained glass windows were installed in March and April 1928 and they provided ventilation and light to the café thereafter.

44.         A report in the Irish Times dated 11 January 1941 records "that a series of very beautiful stained glass windows was being removed and replaced by the ordinary embossed glass....They were being removed to a place of absolute safety, and will be restored to their proper places when all risk of damage through aerial activity is past..."

At an unknown date the stained glass was returned to the café, presumably after the end of the war in 1945.

45.         At some point between 1946 and 1972, the windows were boxed in from the outside and were back lit by artificial light. They no longer lit the room by permitting daylight to pass, nor did they provide any ventilation. In 1986 Campbell Catering acquired the shares of the first named respondent. One of its directors and a shareholder, Mr. Campbell gave evidence that the remaining Swan Yard window was protected on the outside by a wire grille and that through the grille he could see "some kind of opaque panel". This evidence was not challenged and was accepted by the trial judge.

46.         At the time of the change of ownership of the company, it was necessary to carry out repairs to the building. In particular it was necessary to install a new fire escape stairs in the southwest corner of the café. This required the repositioning of the Swan Yard windows from their then existing location to a newly constructed internal wall which screened the new staircase. The Swan Yard window which had not been blocked up was, at this point in time, "separately glazed" and the "panels of the Harry Clarke artwork were held in wooden sub-frames which were in turn attached by screwed bottom mounted hinges to an independent frame. That frame was mounted to the wall to the inside of the external window" according to the architect retained by the first named respondent, Mr. Horan. He gave evidence that both the sashes and the frame of the Swan Yard windows were moved to the new internal wall dividing the new emergency staircase from the café.

47.         In 1987 the premises were the subject of a sale and leaseback agreement whereby the appellant's predecessor in title acquired the landlord's interest and the first named respondent remained as tenant under a new lease. The implications (if any) of the terms of the sale and lease to the ownership of the six stained glass windows will be considered later in this judgment.

48.         In 1998 works were carried out to the café and to facilitate the refurbishment the sashes of all six windows were removed from their frames. This was recorded in a video which was admitted into evidence and the trial judge made findings of fact based upon his own viewing of the video and the evidence of Mr. Slattery commenting on what was shown on the video as discussed above.

Summary of the decision of the High Court in relation to the status of the Harry Clarke works

49.         There is little to be gained in a detailed exposition of the decision of the High Court, given the nature of the appeal and cross-appeal. The issues will emerge during the discussion of the arguments in the appeal and I shall refer to trial judge's analysis as required in due course. For the purposes of placing the appeal in context the following facts were described or found by McDonald J.:

1.             The Harry Clarke works were commissioned in the context of the development by Ernest Bewley of a new café to be occupied by the family company then known as Bewley's Oriental Cafes Ltd.

2.             The building was complete in 1927 and the café opened for business on 27 November 1927. "Given that there must have been window frames in place when the café opened in November 1927, it seems likely that, when it came to the installation of the stained-glass works in 1928, those frames were retained in place and fitted with the sashes containing the stained glass" (para. 67).

3.             It is unclear from the surviving correspondence whether the works were commissioned by the landlord or the tenant.

4.             It is unclear from the correspondence whether the account was paid by Ernest Bewley on his own behalf or on behalf of the company or by the company directly.

5.             It would be "unsafe" to take the minutes of the board of the company of 9 March 1928, on its own, as sufficient evidence to ground a finding of fact that, on the balance of probabilities, the company paid for the Harry Clarke works.

6.             The six stained glass windows were installed in March and April 1928. They were capable of weathering the building.

7.             They originally provided light and ventilation to the café. The windows opened by means of the hopper mechanism previously described. They were designed to be functional, not purely ornamental.

8.             The sashes may be removed from their frames without causing any damage to the frames or the building, with care and relatively easily.

9.             The original window frames are still in situ in the Four Orders window frames and one of the Swan Yard window frames. The second Swan Yard window was blocked up prior to 1972.

10.         The six stained glass windows were removed for safe keeping from January 1941 until an unspecified date, presumed by McDonald J. to be around the end of the war. The windows were replaced during this period by ordinary, embossed or opaque, glass. It is not possible to conclude from the report in the Irish Times that there was no second layer of windows, as posited by Mr. O'Connell for the respondents.

11.         At some time between 1946 and 1972 all of the windows were boxed in (or in the case of one of the Swan Yard windows, blocked up) so that they no longer ventilated the café nor permitted natural light to illuminate the café.

12.         The stained glass panels were back lit by artificial light.

13.         In 1986 the remaining window overlooking Swan Yard was observed to be glazed with clear glass and protected by a wire grille.

14.         In 1987 the Swan Yard stained glass works were in two frames, one in front of the window glazed with clear glass and the other in front of the blocked up window opening. The first frame was mounted to the wall to the inside of the external window with a space between the clear glass window and the Harry Clarke works. There was thus a double fenestration arrangement in the case of one of the Swan Yard windows with a clear glass window in the external wall and a space between it and the Swan Yard work created by Harry Clarke.

15.         The Swan Yard works were moved in their frames to a newly erected internal wall and back lit so as to still "read" like windows.

16.         The 1998 video showed that the Four Orders works were flush against the bullnose brick of the western façade of the café and therefore there would have been insufficient space for an external layer of fenestration between the Four Orders works and the façade. There was no evidence that the window frames were ever moved and they "appear to be in their original position".

17.         The outside of the Four Orders works was painted and carved in a manner that suggested that they were intended to act as part of the exterior of the premises. There was evidence of weathering on the external side of the Four Orders works which "supports the view that the works were exposed to the elements for some time" (para. 147).

18.         Mr. O'Connell for the respondents could not explain how the double fenestration could operate to ventilate the café, nor point to any reference to such an arrangement in any relevant literature, nor had he ever encountered such systems.

19.         There was no physical evidence of any connecting rod system in respect of any of the stained glass windows and a putative second layer of windows, including the clear glass window overlooking Swan Yard, as envisaged by Mr. O'Connell's double fenestration theory.

50.         Based on these facts the High Court reached the following conclusions in respect of the Four Orders windows.  On the balance of probabilities, the Four Orders works were installed in 1928 in the position shown in the 1998 video and they operated as windows at that time admitting light and ventilation to the café and also weathering the café from the elements. They therefore formed part of the external skin of the building at that time (para. 150).  The Four Orders works operated in the same way as conventional windows and on their installation in the western façade for the building they constituted windows. They became part and parcel of the building and they would not be considered to be fixtures. They replaced the original windows installed in 1927 in substitution for those windows, not as an addition to the original structure (paras. 167-168).  The removal of the sashes would amount to the removal of the windows, as, without the sashes, there would be only an empty window frame, so the fact that the sashes are readily removable does not mean that they are tenant's fixtures and not part of the windows of the building.  As part and parcel of the premises, the Four Orders windows passed under the sale of the premises in 1987 to Royal Insurance plc as purchaser as an inherent part of the premises sold. The appellant, as its successor in title is the owner of the Four Orders windows, subject to the 1987 lease to the first named respondent (para. 202).

51.         The trial judge reached a different conclusion regarding the Swan Yard windows. He acknowledged that "it appears that the original intention was that the Swan Yard works would be placed in the window openings overlooking Swan Yard."  However, he noted that there was no physical or other evidence that they were sited in the external wall of the café building overlooking Swan Yard.  He noted that there "...is evidence which establishes that...a double fenestration system [as postulated by Mr. O'Connell] was in place in 1987".  This was because Mr. Horan and Mr. Campbell gave direct evidence that one of the Swan Yard works was mounted to the wall on the interior side of the clear glass window in the exterior wall overlooking Swan Yard.  He accepted the evidence of Mr. O'Connell that the frame of the window in the external wall of the café overlooking Swan Yard is the original window in that location and that there was nothing to indicate that there was movement of a window in that location.  He expressly held that he accepted Mr. O'Connell's evidence which he described as being measured, authoritative and very helpful.  At para. 154 of his judgment he held:

"...[I]t seems to me that the evidence can only establish that there was no movement of the window frame. It is clear from the evidence that it would be movement of the window frame that would likely require a cutting of the surrounding mortar and the render/plaster which should be observable today.  There is no basis to think that movement of the individual sashes sitting in such window frames would require any intrusion into the surrounding mortar or render. Having seen the 1998 video showing the way in which the stained-glass sashes of the Harry Clarke works could be removed from the frames without any damage to the window reveals or surrounds, one could not exclude the possibility that the Swan Yard works were originally installed in the window frames overlooking Swan Yard and that, at some, unknown point thereafter, the sashes containing the stained glass were removed from the window frames and transferred to mock windows in the locations observed by Mr. Horan and Mr. Campbell in 1987."

52.         McDonald J. held that the appellant, as the plaintiff in the proceedings, bore the burden of proving on the balance of probabilities that the clear glass window found in 1987 in the external window frame was not part of the original installation and that the Swan Yard works were in fact fitted in the window openings in the Swan Yard wall in 1928.  He adverted to the significant gaps in the evidence.  In particular, he emphasised that the correspondence provides no detail in relation to what was involved in the installation of any of the stained glass works. 

53.         The appellant urged the High Court that the Swan Yard works were designed to open inwards and provide ventilation and that they were fitted with the hopper mechanism to facilitate this.  The trial judge acknowledged that this was potentially a significant factor as "[t]he question arises as to whether it is plausible to think that a different approach might have been taken by the parties in 1928 in respect of the Swan Yard works, on the one hand, and the Four Orders works, on the other."   He acknowledged the "obvious difficulty" that could arise with ventilation in the event that a double layer of windows had been employed.  He observed that "[a]t minimum, the existence of those practical difficulties suggest that such a system would have been seen as an unattractive solution." He then proceeded, as he acknowledged himself, to speculate in paras. 160 - 162 in the following terms:

"160. So, why then might the parties have decided to put a double layer of windows in place at Swan Yard but not in the western wall? If there is no possible reason why they might have decided to do so, that would support the plaintiff's case that it is implausible to think that the double layer of windows observed in 1987 was adopted in 1928. But, in my view, there is at least one possible reason why the parties may have decided to take such a course..... I have already drawn attention to the fact that the Swan Yard window openings overlook a rather dark blind alley which is unlikely to be much frequented by the public. In those circumstances, stained-glass windows in those openings would not only be at risk from a security perspective but would also be at risk from vandalism. As I have previously mentioned, a stone thrower could cause irreparable damage to the stained glass without much risk of the act of vandalism being seen by any passer-by. In addition, there may have been a concern about the risk of damage to the stained glass from the sacks of flour being hoisted to the bakery on the upper floor. There may even have been a concern about the ingress of flour dust from the sacks of flour. There was certainly time for any concerns of that nature to have manifested themselves before the Swan Yard works were installed. It should be recalled in this context that, by the time the Swan Yard works came to be installed in March or April 1928, the café had already been operating for a number of months. For the reasons previously explained, these considerations were wholly absent in the case of the Four Orders works but they could have existed in the case of the Swan Yard works. The retention of the existing glass windows at the Swan Yard side of the building may have been perceived to provide some measure of protection to the stained-glass works. Of course, that purpose might also have been secured by putting a security grille or security bars in place. However, such a solution may have been more likely than the retention of the clear glass window to cause the type of intrusive shadow highlighted by counsel for the plaintiff in the course of his cross-examination of Mr. O'Connell (as discussed in para. 125 above).

 

161. As I have already noted, such a double fenestration system would likely have created practical difficulties for the ventilation function of the Swan Yard works if not wholly impeded that function. On one view, that supports the plaintiff's case that it is implausible to think that such a system was employed in 1928. However, the fact remains that, in 1987, the Swan Yard works were not providing any ventilation function to the café at that time. If the ventilation function was not considered necessary at that time, why would the position have been any different in 1928? Again, the fact that there was a gap of almost a year between the opening of the café and the installation of the Swan Yard works in 1928 may be relevant. That gap in time would have given some opportunity to assess whether ventilation was important at this location of the café or whether there were other factors potentially in play that outweighed the need for it there (such as the issue of security previously discussed or indeed the ingress of flour dust from the haulage of the sacks of flour to the bakery above). As discussed in para. 144 above, such considerations were entirely absent in the case of the Four Orders works and provide a plausible explanation for why different approaches may have been taken at the two locations in issue.

 

162. I have not lost sight of the fact that, in the photograph discussed in para. 12 above, the hopper mechanism can be seen to be fitted to the Swan Yard works at that time. I have previously observed that the photograph would appear to have been taken in the 1970s or early 1980s. It might be thought that this would suggest that the Swan Yard works may have been functioning as conventional windows at that time admitting ventilation to the café. It may well have been the case that they did provide some level of ventilation prior to being boxed in (as described by Mr. Campbell). However, I do not believe that the presence of the hopper mechanism is sufficient to infer that the Swan Yard works functioned as a direct source of ventilation either at the time of the photograph or in 1928. As the 1998 video demonstrates, the hopper mechanism was still fixed to the Four Orders works in 1998 notwithstanding that, as the evidence of Mr. Campbell makes clear, they were no longer functioning as a source of ventilation at that time. Moreover, based on the opinion of Dr. Caron and Ms. Costigan, the hopper mechanism was likely taken into account by Harry Clarke in his design of the stained-glass works. Dr. Caron went so far as to express the view that the mechanism is part of the fabric of the works and that it is a shame that it has been removed. It is hardly surprising in those circumstances that the mechanism would not be removed even in circumstances where the Swan Yard works provided either no ventilation at all or only provided limited or compromised ventilation."

54.         In para. 163, having acknowledged that the factors discussed involved speculation, he emphasised the evidence as to what was observed in 1987 involved no element of speculation.  He acknowledged the reality was that there was no evidence before the court which provided "...a reliable basis on which to conclude, on the balance of probabilities, that the double window arrangement observed in 1987 was not also in place in 1928 when the Swan Yard works were installed."  He acknowledged that it was possible that the arrangement was not installed until sometime after 1928, as argued by the appellant, but said "[b]ut there is no evidence which confirms this.  It is, at least, equally possible that the system was installed at the time of the original installation of the Swan Yard works in 1928."  He then proceeded:

"While I have been able to conclude that it is implausible that such a system was used in the context of the Four Orders works, the factors discussed in paras. 160 to 161 suggest that the employment of such a system on the Swan Yard wall cannot be dismissed as implausible.  In those circumstances, I do not believe that it is possible to hold that the [appellant's] postulated version of events is the more probable. There is no sufficient basis in the evidence to do so. It follows that the [appellant] has failed to establish on the balance of probabilities that, in 1928, the Swan Yard works were installed in the window openings in the wall overlooking Swan Yard."

55.         McDonald J. concluded that the Swan Yard works were installed in 1928 as part of a double layer of windows of the kind observed by Mr. Horan and Mr. Campbell in 1987.  He held that the clear glass windows occupied the external window openings and therefore the removal of the Swan Yard works would not have interfered in any real way with the convenient use of the café premises.  The café would still have been weather tight and the building would still have been complete, with no gap left in its external shell.  In those circumstances "where the Swan Yard works were installed parallel to existing windows, they are properly classified as additions to the structure or fabric of the café. They fall within the ambit of things which, in the words of Scrutton L.J. in Boswell v. Crucible Steel, "... have been brought into the house and affixed ... after the structure has been completed."  On that basis, they are classically in the nature of fixtures. In contrast to the Four Orders works (which replaced the existing windows) they cannot be considered to be part and parcel of the fabric of the café."  

56.         McDonald J. then proceeded to consider whether anything that had occurred in the years since 1928 affected the status of the stained glass works.  He held that the temporary removal of the works during the Second World War and later in 1998 made no difference either to the Four Orders works or the Swan Yard works.  Likewise, the removal of the Swan Yard works to a new internal wall separating the café from the fire escape in 1987 did not change their status. 

57.         He noted that by 1972 the five openings were boxed in as observed by Mr. Larry Ebbs who commenced his work maintaining the stained glass in that year.  That boxing was replaced in 1998 and this in turn was replaced in 2016 by a Perspex screen which remained in place up to the date of trial.  He concluded that there was insufficient detail available to satisfy him on the balance of probabilities that the various measures which were put in place had the effect of converting the status of the Four Orders work to fixtures.  In relation to the Perspex sheeting, he held that on its own it could not reasonably be considered to be an appropriate or acceptable outer shell for the café and therefore the Four Orders works continued to form part and parcel of the external skin or shell of the café premises.  As such they were part and parcel of the leased premises and did not constitute fixtures. 

58.         He concluded that the Swan Yard works were tenant's fixtures as opposed to landlord's fixtures.  He considered the issue on the basis that the Swan Yard works were installed in 1928 as part of a double fenestration arrangement parallel to - and to the inside of - the existing clear glass windows on the Swan Yard side of the café.  Unlike the clear glass windows, they were not a necessary element of the external shell of the premises.  McDonald J. therefore concluded that they were not required to serve any purpose of the landlord and that "[i]n those circumstances, it is difficult to see that the Swan Yard works could be said to have been installed with a view to improving the landlord's interest in the café premises."  He therefore concluded that the Swan Yard works must have been installed "to further the interests of the tenant in the operation of its café business" and should be considered to constitute tenant's fixtures. 

59.         This conclusion in turn required him to consider s.17 of Deasy's Act and whether the first named respondent as the tenant was entitled to remove the windows under the provisions of the section.  Five conditions must be satisfied before a tenant can rely on this provision.  He had identified these in para. 23 of his judgment as follows:

"23. ...First, the items in question must have been affixed by the tenant to the premises at the tenant's expense. Second, the items must be for the purposes of the tenant's trade or for ornament in relation to the tenant's occupation of the premises. Third, the items must be capable of being removed without substantial damage either to themselves or the premises in which they have been installed. Fourth, the items must not have been installed pursuant to an obligation to do so owed to the landlord. Fifth, the objects must not have been installed in breach of the tenant's agreement with the landlord. In addition, it should be noted that the language of the section plainly envisages that the parties are free to reach a contrary agreement in which case, the section will not apply." 

60.         At para. 188 of the judgment McDonald J. concluded that four of these conditions were met because the Swan Yard works were installed as ornamentation in relation to the tenant's occupation of the building and for the purposes of the café trade operated by it; there was no evidence that the tenant was obliged by the landlord to install them; there was nothing in the 1928 lease which prohibited their installation and having regard to the 1998 video, it was clear that the sashes can be removed without substantial damage either to themselves or to the premises.

61.         The High Court then considered whether the works were affixed to the premises at the tenant's expense.  McDonald J. noted that there was no direct evidence that the tenant paid for the Swan Yard works but that the minutes of the board meeting of 9 March 1928 was evidence that the tenant would be responsible "for defraying the costs of the café fixtures and fittings".  He referred to his earlier expressed view that it would be unsafe to take the board minute on its own as sufficient evidence to ground a finding that, on the balance of probabilities, the tenant paid for the Harry Clarke works.  However, he was of the view that there was other material available which supports the view that they constitute tenant's fixtures and that this changed the position.  He said he was "now of the view that the Swan Yard works were installed solely for the benefit of the tenant and that they are tenant's fixtures."  He explained his reasons as follows:

190. ...In the first place, while the minute did not mention stained-glass, it dealt expressly with café fixtures and made clear that these were for the tenant's account. There is no reason to suppose that the stained-glass fixtures comprising the Swan Yard works would be treated as an exception to the arrangement recorded in the minute. Secondly, although there is no inventory of what was included in the sum of £14,900 (which was for the tenant's account), there is no reason to suppose that such a large sum (in 1928 values) did not include the sum of £180 due to Harry Clarke for the Swan Yard works. Thirdly, given that the works are fixtures for the benefit of the café trade carried on by the tenant, it would make no sense that Ernest Bewley would pay for them on his own behalf rather than on behalf of the tenant." 

62.         At para. 192 of the judgment McDonald J. concluded that having regard to the terms of the minute and the fact that the Swan Yard works constitute fixtures, the bill for the Swan Yard works was for the account of the tenant and therefore was an expense of the tenant and accordingly the final condition in s.17 of Deasy's Act was satisfied in respect of the Swan Yard works.

63.         The trial judge rejected the appellant's argument that the obligation to deliver up possession at the end of the 21 year term of "all fixtures" displaced the provisions of s.17 of Deasy's Act.  The obligation to give possession of the fixtures to the landlord at the end of the 21 year term was never triggered.  McDonald J. held that in those circumstances, where that obligation was never triggered and where the landlord and tenant relationship is now governed by a lease of 1987, he could not see any basis upon which the provision of the 1928 lease could be said to affect the entitlement of the tenant to rely on s.17 of Deasy's Act.

64.         In paras. 197 to 207 the trial judge considered the potential impact of transactions entered into in 1987 by way of sale and leaseback and he concluded that the 1987 transactions did not alter his conclusions that the Swan Yard works are fixtures owned by the first named respondent nor the conclusion that the Four Orders works comprise part and parcel of the café premises leased to the first named respondent under the 1987 lease.  As the arguments advanced in the High Court were not pursued in the appeal it is not necessary to set them out at this point.

65.         He accordingly held that the appellant was the owner of the Four Orders works and was entitled to an order setting aside the purported transfer of those works from the first named respondent to the second named respondent.  The respondents had succeeded in establishing that the Swan Yard works constituted tenant's fixtures and the transfer of ownership of those works from the first named respondent to the second named respondent could not be impugned by the appellant and therefore the second named respondent was entitled to a declaration that it is the owner of the Swan Yard works.  He made no order as to costs.

The appeal and cross-appeal

66.         As has been previously set out, the appellant appealed the findings of the High Court in relation to the ownership of the Swan Yard works and the respondents cross-appealed in relation to the findings in respect of the Four Orders windows.  

The legal principles relevant to the appeal

Appellate review of the trial judge's findings of fact   

67.         The parties agreed on the principles to be adopted by an appellate court in relation to the factual findings and expert evidence.  In Emerald Isle Assurances & Investments Ltd. v. Dorgan [2016] IECA 12 at para. 31 Ryan P. summarised the principles as follows:

"(a)   Were the findings of fact made by the trial judge supported by credible evidence? If so, the appellate court is bound by the findings, however voluminous and apparently weighty the testimony against them.

(b)     Did the inferences of fact depend on oral evidence of recollection of fact?  If so, the appeal court should be slow to substitute its own different inference.

(c)     In regard to inferences from circumstantial evidence, an appellate court is in as good a position as the trial judge in that regard.  Did the judge draw erroneous inferences?"

68.         In Cloonan v. HSE [2022] IECA 129 at para. 114 Power J. speaking for this Court held:

"This Court is required to consider: whether the findings made were supported by credible evidence; whether they were based on a reasoned conclusion; whether there were any significant and material errors in the way in which the trial court reached its findings; and whether the trial judge engaged sufficiently with essential parts of the evidence. If an issue of non-engagement is raised, it is not established by identifying other parts of the evidence that might support a different conclusion. Rather, what must be shown is a failure to deal with something that went to the very core or essential validity of the trial court's findings."

And at para. 155 of her judgment she said:

"It is well established that an appellate court may have greater scope when reviewing findings made on the basis of expert evidence than when doing so on non-expert oral testimony.

She summarised the position at para. 160 as follows:

"The relevant principles on the scope of appellate review in respect of findings based on expert evidence may be summarised thus:

(i)      that an important part of the trial court's assessment of expert evidence is the application of logic and common sense to the views of the expert;

(ii)     that, on review, such assessment may be significantly more amenable to analysis based on the logic of the positions adopted by the competing experts;

(iii)    that where a finding of fact is of significant materiality to the overall conclusion of a case, it must be accompanied by reasons which set out, clearly, the basis for the finding without which an appellate court will be unable to determine whether the finding is sustainable;

(iv)    that 'speculative leaps' unsupported by credible evidence are not sustainable;

(vi)    that a bare statement by the trial court that it prefers the evidence of one expert over another will not be sufficient;

(vi)    that some explanation, however brief, is required to show the basis for such preference; and

(vii)   that this necessarily involves engagement with an analysis of the competing views of the experts."

69.         In James Elliott Construction Ltd. v. Irish Asphalt Ltd. [2011] IEHC 269 at para. 12 Charleton J., then a judge of the High Court, observed:

"Every expert witness has to be evaluated on the basis of sound reasoning. An expert witness is, however, no different to any other witness simply because he or she is entitled to express technical opinions; all of us are subject to human frailty: exaggerated respect based solely on a witness having apparent mastery of arcane knowledge is not an appropriate approach by any court to the assessment of expert testimony. Every judge has to attempt to apply common sense and logic to the views of an expert as well as attempting a shrewd assessment as to reliability."

70.         These are the principles this Court must apply in assessing the evidence from the expert witnesses and the inferences drawn by the trial judge from this evidence and from the extant documentary evidence and the real evidence when determining the arguments raised on the appeal and cross-appeal.  

 

Whether an object is a fixture; whether it is a landlord's fixture or a tenant's fixture

71.         Crucial to the Court's determination is the issue of whether the stained glass works are part and parcel of the building or are fixtures; and if the latter, whose fixtures. As was set out in the respondent's written legal submissions to the High Court and accepted by McDonald J., an object which is brought onto the land may be classified under one of three headings: (a) a chattel (i.e. an item of tangible, moveable property); (b) a fixture (a chattel that is affixed to land or premises); or (c) an item that, because of its role, is properly classified not as a chattel or as a fixture but as part and parcel of the land.  Neither party contended that the stained glass windows were chattels; the issue was whether they were fixtures (as contended by the respondents) or part and parcel of the land (as contended by the appellant).  Where an item is an integral element of the fabric of the leased premises it is treated as part and parcel of the premises.  This is to be contrasted with the position of a fixture. 

72.         Professor Wylie in Irish Landlord & Tenant Law (4^th Edition) 2022 at para. [9.02] stated:

"The terminology in this subject is often confusing and this should be borne in mind in reading the authorities. In particular, the word 'fixture' is often used in different senses. In its most general sense a fixture refers to some chattel which has been affixed or attached to land. As Holmes J put it in Earl of Antrim v Dobbs:

'It is essential to the legal idea of a fixture that it should be let into the lands, or firmly fixed to some fabric having its foundation in the ground.'

So far as a landlord and tenant are concerned the question obviously arises as to the ownership of a chattel so affixed, but it is crucial to distinguish between different situations to all of which the term 'fixture' may be applied. One is where the landlord, before granting the lease, at the time of granting it or during the lease's currency, installs on the demised premises chattels or fitments. These remain the landlord's property and, in so far as they are fixed to the premises, they may be referred to as 'landlord's fixtures'. The converse situation is where the tenant installs chattels or fitments on the demised premises. In so far as these become so attached or annexed to the premises as to constitute 'fixtures', they may become the property of the landlord, in which case again they may be referred to as "landlord's fixtures", or be caught by a clause which deals with landlord's fixtures. On the other hand, despite becoming 'fixtures,' the items in question may remain the property of the tenant, in which case they are usually referred to as 'tenant's fixtures.' The source of much confusion is that, in the context of the relation of landlord and tenant, the word 'fixtures' on its own is often taken to mean the last category only." 

What is a window?

73.         The issue of whether a "window" which is both affixed to the building and easily removable is to be regarded as part of the land or a fixture was considered in the case of Climie v. Wood (1869) L.R. 4 Exch 328. Willes J. expressly said that windows were an example of an element that should be regarded as part of the fabric of the building.  At pp. 329-330 of the report, he said:

"There is no doubt that sometimes things annexed to land remain chattels as much after they have been annexed as they were before. The case of pictures hung on a wall for the purpose of being more conveniently seen may be mentioned by way of illustration. On the other hand, things may be made so completely a part of the land, as being essential to its convenient use, that even a tenant could not remove them. An example of this class of chattel may be found in doors or windows. Lastly, things may be annexed to land, for the purposes of trade or of domestic convenience or ornament, in so permanent a manner as really to form a part of the land; and yet the tenant who has erected them is entitled to remove them during his term, or, it may be, within a reasonable time after its expiration." (Emphasis added).

74.         But this begs the question: what is a window? The question of what is a "window" is a mixed question of law and fact. In Holiday Fellowship Ltd. v. Hereford [1959] 1 W.L.R. 211 Lord Evershed M.R. ascribed the following meaning to "windows":

"I wish to be a little careful in the meaning that I am attaching to the word "windows". I have described the bays or similar structures, which are, at any rate to judge from the photograph, made of the same material as the walls, in the ordinary sense of that term. We are not concerned with any question of repair to the brick or stone structures containing the actual windows. For the purposes of this case and of the question raised in the originating summons, I take "windows" to mean, and to be confined to, the glass panels and the wooden framework and apparatus in which the glass is placed..."  (Emphasis added).

75.         The question was considered again in Easton v. Isted [1903] 1 Ch 405. The case concerned the windows in a conservatory which overlooked a neighbouring property and an agreement by the owner to pay one shilling a year as an acknowledgement of being allowed to overlook the defendant's property.  Subsequently, the conservatory was converted into a passage and the glazed side was bricked up leaving the glazed roof functioning as a skylight to the passage.  Some years later the defendant built a wall on his boundary which carried above the skylight thereby obstructing access of light to the plaintiff's passage.  Joyce J. rejected the suggestion that there was a distinction between the original glazed portion of the vertical side of the conservatory and the remaining sloping roof.  He held:

"A window is not less a window because it is not capable of being opened, nor is it less a window because it is not fixed in a vertical plane.  I think the glazed top was just as much a window as the fixed portions of the vertical side, and, inasmuch as it received light over the defendant's property, I think it overlooked the property in the sense in which that term is used in the agreement."

The decision and reasoning was upheld on appeal.

76.         In Boswell v. Crucible Steel Company [1925] 1 K.B. 119, the Court of Appeal of England and Wales considered whether plate glass which formed part of the walls of a building were landlord fixtures within the meaning of a repairing covenant (or were part of the building).  The court held that "[t]he skin of the house" consisted to a large extent of glass.  That being so, Bankes L.J. held that it was "impossible to say that windows such as these, forming part of the original structure of the house, are landlord's fixtures.  The County Court Judge was wrong in law in holding that these windows were not part of the walls of the house."

77.         Scrutton L.J. said that the greater part of the outer sides of the building consisted of plate glass windows so that the premises are practically enclosed in a wall of glass.  He held that the expression "landlord's fixtures" "cannot include a thing which forms part of the original structure of the building.  It must be regarded as confined to things which have been brought into the house and affixed to the freehold after the structure is completed."  Atkin L.J. held that the windows could not be regarded as landlord's fixtures

"for the simple reason that they are not fixtures at all in the sense in which that term is generally understood.  A fixture, as that term is used in connection with a house, means something which has been affixed to the freehold as accessory to the house.  It does not include things which were made part of the house itself in the course of its construction.  And the expression, "landlord's fixtures" as I understand it, covers all those chattels which have been so affixed by way of addition to the original structure, and were so affixed either by the landlord or, if by the tenant, under circumstances where they were not removable by him.  As these windows were part of the original structure, representing the walls of the house, so that without them there would be nothing that could be described as a warehouse at all, they cannot come under the head of landlord's fixtures."   

78.         Thus, a window may perform a variety of functions: it may illuminate the building; it may weather the building; and it may ventilate the building.  However, it is possible to have a window which does not ventilate the building and it is possible to have an internal window which does not weather the building.  It is also possible to have darkened windows which do not illuminate the building.  The fact that the sashes of a window may be readily easily removed and replaced, does not deprive them of the characteristic of a window. Each case must be assessed by reference to the particular facts of the case.  If the purported window is essential to the building, then it should probably be considered legally to be a window and part and parcel of the premises and should not be treated as a fixture.

Discussion

The Four Orders Windows

79.         In light of these principles and the findings of the trial judge, I turn now to consider the arguments advanced on the appeal.  It should be noted that many of the issues which were canvassed in the High Court did not feature in the appeal.  The essential issue was whether the stained glass windows were windows in the sense of being part and parcel of the premises or whether they were tenant's fixtures which belonged to the first named respondent originally and which were removable by the first named respondent under the provisions of s.17 of Deasy's Act. 

80.         It must be acknowledged that the trial judge had a most difficult task in trying to resolve the issues presenting in the proceedings in circumstances where there was so little evidence and, as a result, both witnesses and the court were left to conjecture and inference to try to bridge the gaps in that evidence.  The difficult task is to identify the secure uncontroversial evidence and then to reach inferences supported by and based on that evidence while refraining from impermissible speculation divorced from the evidence.

81.         The starting point must be the period 1927 to 1928 when the premises were built, and the Harry Clarke windows were commissioned and installed and the first named respondent commenced to trade as a cafe.  The appellant and the first named respondent each assert that they were originally and have always remained the owner of the stained glass windows.  Neither party argued on appeal that the ownership has changed by reason of any of the events affecting the premises in the intervening period, other than the first named respondent's assertion that it has assigned the ownership of the windows to the second named respondent in December 2020.  Each point to events from the 1940s, 1970s, 1980s, 1990s and 2000s and seek to draw certain conclusions which, they say, show who was the owner of all six windows in 1928.  They each contend that the trial judge erred in concluding that the windows were not all the property of one or other party.

The evidence from the premises

82.         In the 1920s Ernest Bewley personally assembled a site between Grafton Street, Johnson's Court and Swan Lane.  He developed the site as an oriental café, not a generic premises.  The drawings and photographs of the façade present an elaborate ornamental building in keeping with his vision of a continental style café.  The façade on the side at Johnson's Court, though plainer, is in keeping with the overall vision for the building.  Ernest Bewley paid for the oriental façade, and he paid for the stained glass panels and lantern lights used elsewhere in the building as the developer of the premises. There was clearly one coherent design for the premises as a whole and the six stained glass windows at the heart of this case were entirely in keeping with that overall design. Certainly, if the trial judge was correct in concluding that the Four Orders windows were part and parcel of the premises, it is difficult to see why he concluded that the Swan Yard windows did not serve any interest of the landlord as they complimented and completed the vision for the café just as the Four Orders windows did. If the landlord had not commissioned and installed all of the stained glass in the café he potentially put at risk the artistic integrity and harmony of the overall design of the premises. Of course, as a director of the tenant as well as the landlord, it was open to Mr. Bewley to ensure that any fixture installed by the tenant did not compromise his vision as landlord for the premises, but it would nonetheless be surprising if the landlord was responsible for the entire artistic design of the café save for two windows overlooking Swan Yard.

83.         There were six opes in the sitting area to the rear of the building, though one has been completely blocked up.  The stained glass windows produced by Harry Clarke's company were designed to go into these six opes.  The original frames for the windows are still in situ in the five remaining opes and have not been disturbed since they were installed. This is established by the evidence of Mr. O'Connell in respect of the remaining Swan Yard frame, the video from 1998, and the evidence of Mr. Slattery in respect of the Four Orders frames, and which was accepted by the trial judge. The trial judge also found as a fact that the frames for the Four Orders windows are flush against the bullnose brick of the four openings and there is no space for an outer layer of window to be inserted between the existing frames and the bricks. It follows that the five frames remaining in situ were in situ in 1927 and held the sashes which operated as the external windows to the building. These weathered and lit the building.

84.         The sashes containing the stained glass panels were and are easily removable from the window frames.  This can be seen from the video and it is confirmed by the fact that the sashes were removed in 1941 and restored after the end of the Second World War. They were similarly easily removed and restored in 1998. The outer side of the Four Orders sashes are painted and carved in a manner which the trial judge accepted suggested that they were intended to act as part of the exterior of the premises. The trial judge also found as a fact, based upon, inter alia his own viewing of the video, that there was some evidence of weathering on the exterior facing side of the frames and sashes, and, by implication, rejected the evidence of Mr. Larry Ebbs that the condition of the stained glass panels is inconsistent with them having been exposed to the elements. Thus, the physical evidence suggests that the Four Orders sashes were designed to form part of the exterior of the building and were not designed to be placed behind an outer layer of clear glass windows and that they were exposed to the elements. This leads to the inference that they were designed to and functioned for a period of time as the outer windows of the building.

85.         The Four Orders windows were designed to open by means of the hopper mechanism. This was an inherent part of the design and is clearly an original feature. The surviving mechanism shows that the windows were designed to open inwards, and that they actually were operated in this matter, as appears from the photograph of the café described by the trial judge. This means that the windows also originally ventilated the café space, which, it was accepted by all the witnesses, was a significant requirement. It also reinforces the argument that the windows were originally designed to be part of the weathering of the building as the inward opening of the sashes allowed for ventilation while protecting the interior from ingress of rain.

86.         While there was no physical evidence that the Four Orders sashes, which in the 1998 video were mounted in the original frames, had been there from 1928, the issue is whether, on the balance of probabilities they were always in situ in the original frames. In answering this question, the absence of evidence to the contrary is significant. There was no physical evidence that there were ever any other frames which could have held the sashes. It was not possible to insert outer frames between the existing frames and the bullnose bricks of the opening.  There was no evidence of any interior frames, which no longer existed by 1972, which could have held the stained glass sashes with the existing, original frames then holding clear glass sashes. In addition, there is no evidence of any physical connection between the existing stained glass sashes and the hypothetical sashes with clear glass to which the stained glass sashes were linked in order to facilitate the coordinated opening of the two layers of windows. Furthermore, it is difficult to conceive why the arrangement posited by the respondents through Mr. O'Connell would have been adopted in 1927.  His theory requires the following sequence of events:

1. a double fenestration design is devised;

2. temporary sashes glazed with clear glass are inserted to weather the premises in 1927;

3. the stained glass sashes are installed in 1928 inside the clear glass sashes, mounted on an internal frame;

4. the hopper opening mechanism for the stained glass sashes is linked to the external clear glass sashes so that they can open in tandem;

5. that the original 1928 double fenestration arrangement comprising inner frames in which the stained glass sashes were mounted inside the outer frames glazed with clear glass sashes is at a later time be replaced by:-

(1) removing the inner frames in which the stained glass sashes had until that point in time been held,

(2) removing the outer clear glass sashes,

(3) inserting the stained glass sashes in the outer frame

(4) boxing the now exposed stained glass in, and

(5) lighting it from behind.

The more probable scenario, it seems to me, is that posited by McDonald J. at para. 67 of his judgment:

"Given that there must have been window frames in place when the café opened in November 1927, it seems likely that, when it came to the installation of the stained-glass works in 1928, those frames were retained in place and fitted with the sashes containing the stained glass."

As the task of the court is to determine the issues of fact on the balance of probabilities, I am satisfied that this conclusion is far more probable- and therefore correct- when contrasted with the alternative suggested by the respondents.

87.         It is also important to note that ventilating the premises with a double fenestration arrangement would be more difficult than ventilating the premises if the stained glass sashes operated by the hopper mechanism were the only exterior window. While the hopper mechanism for one of the sashes survives, there is no evidence of a double hopper mechanism in relation to the Four Orders windows or any other indication that the opening of the Four Orders windows could be coordinated with the opening of another layer of windows. Given the importance of ventilation, this is a significant omission.

The Swan Yard Windows

88.         The physical evidence in relation to the Swan Yard windows was different. Witnesses were only able to give evidence concerning the arrangement of these two windows from 1987. The crucial question is whether and, if so, to what extent this configuration differed to the original configuration in 1928. At an unknown date between 1946 and 1972 one of the opes was concreted over. So it is clear that there was at least some alteration to the original configuration. Mr. Ebbs, a stained glass worker, did not give oral evidence but his witness statement was admitted into evidence. He said that he had worked on the maintenance of Bewley's stained glass collection since 1972. He said that from 1972 the stained glass windows were always boxed in either by blockwork or "some other panelling" and that "they have never been exposed to the weather or let in natural light". The reference to blockwork can only be to the Swan Yard ope which has been blocked up. This confirms that his evidence includes the Swan Yard windows and is not confined to the Four Orders windows. Mr. Ebbs said that he removed the stained glass panels a number of times but he does not say that he saw a clear glass window behind one set of panels when this occurred. It would be unusual to describe the existing clear glass window overlooking Swan Yard as "panelling" and it would not prevent the stained glass sashes mounted immediately in front of it from letting in natural light. This evidence suggests that there was no glass window behind the Swan Yard stained glass panel when Mr. Ebbs worked on the panels, or certainly, that he can recollect. This evidence was not analysed by the trial judge and in fairness to him, perhaps not too much weight should be placed on it as Mr. Ebbs was not cross-examined on his evidence. However, it would tend to support the view that the configuration present in 1987 may not have reflected the original arrangement.

89.         The issue was whether the arrangement in relation to the clear glazing and the stained glass sashes reflected an alteration or continuation of the arrangement since 1928. In 1987 the stained glass sashes of the two windows were mounted directly in front of the remaining and the former opening on separate frames. The trial judge accepted the clear evidence of Mr. O'Connell that the frame in the remaining ope facing Swan Yard was the original frame inserted when the building was constructed and it had not been disturbed.

90.         Mr. O'Connell also gave evidence that the sashes containing clear glass were identical in size to the sashes in the Swan Yard window.  I understand this to mean that the Harry Clarke Swan Yard window sashes could be inserted into the original frame overlooking Swan Yard in place of the existing clear glass sashes, precisely as described by the trial judge in para. 67 quoted above. 

91.         At present there are thick metal bars in front of the remaining window overlooking Swan Yard but previously, in 1986, there were no such bars and the window was protected by a wire grille. There was uncontested evidence from Dr. David Caron, a stained glass expert who gave evidence on behalf of the appellant, that during the 1920s stained glass windows in churches were regularly protected by such grilles against breakage and that this method had been employed to protect Harry Clarke windows elsewhere. He also said that the grilles diffused the light and did not give rise to shadowing on the stained glass when in direct sunlight. On the other hand, Mr. O'Connell gave evidence that the glazing bars of the clear glass window, even though coordinated with the stained glass panels mounted inside, would cause a shadow on the stained glass sashes.

92.         As with the Four Orders windows, the Swan Yard windows opened inwards to provide ventilation for the café by means of the hopper mechanism previously described.  Importantly, however, there is no evidence of a hopper mechanism (or any equivalent mechanism) ever having been fixed to the clear glass sashes in the remaining Swan Yard window or the corresponding façade of the stained glass sashes linking the two sashes, and Mr. O'Connell could not explain how the double fenestration system he believed originally existed could have operated.

93.         Mr. Slattery gave evidence that the exterior side of the Four Orders sashes were designed and constructed in a manner to provide a weather function. The trial judge accepted this evidence. Mr. Slattery also gave evidence to the same effect in relation to the Swan Yard stained glass sashes and the trial judge did not advert to this evidence or explain why he accepted it in respect of the Four Orders sashes but rejected it in relation to the Swan Yard sashes. This evidence supports the argument that the sashes were designed to be placed on the exterior of the building and to perform a weathering function and requires to be considered in determining whether the appellant has established on the balance of probabilities that the Swan Yard works performed a weathering function and were part and parcel of the premises from 1928.

94.         The evidence of Mr. Campbell was that in 1986 the Swan Yard sashes were not on the exterior of the building, one ope was blocked up and the other was glazed with clear glass.  The stained glass sashes were in a frame directly in front of the ope (or former ope) and back lit by artificial light.  He gave no evidence that there was any linkage between the clear glass sashes and the stained glass sashes.

95.         Inferences from evidence must be drawn with care.  The court must consider the possibilities in the context of all of the evidence.  When assessing the evidence of expert witnesses, "'speculative leaps' unsupported by credible evidence are not sustainable".  So, it is important to consider the evidence of Mr. O'Connell in the context of the evidence of Mr. Ebbs and Mr. Slattery, as discussed above, and the contemporaneous documents, as well as the physical (or real) evidence from the premises and his conclusions based on the real evidence. 

Evidence from contemporary documents

1. Correspondence

96.         The correspondence from Harry Clarke and his firm shows that he visited the site while it was under construction.  This is in keeping with the evidence that he was always keen to view and understand the premises in which his works would be placed.  This, of course, only makes sense if they were in fact to be windows in the buildings and not merely decorative panels which could be removed and placed elsewhere. For instance, there was no evidence that he inspected the premises to ascertain where the decorative panels which were commissioned in 1927 were to be installed.

97.         He did so before the construction phase of the building works had concluded and the fit out had started.  This tends to support the inference that his client, in respect of all six windows was the developer/landlord and not the tenant.

98.          His correspondence usually describes the works as windows, not panels. While this is not decisive, it is indicative of what he regarded to be his commission.  He did not refer to the six works as panels. In the card he drafted to be handed out to the public during civic week, he described them as "[s]tained glass windows". While he undoubtedly regarded them as artworks, that does not detract from the fact that he consistently distinguished between the stained glass windows on the one hand and stained glass panels on the other.

99.         Harry Clarke regarded the commission as a single commission for six stained glass windows from one client. There was never any suggestion that he had two clients, one for the four westerly opes and one for the two openings onto Swan Yard. The first letter referring to suggestions for dealing with the large openings at the back of the café is dated 4 April 1927 and proposed a general scheme for all six windows. The letter predates any interest of the first named respondent in the premises at the time (though the lease subsequently granted on 9 March 1928 ran from 1 April 1927). The letter follows earlier correspondence in relation to the development and general glazing of the building and should be seen as part of the continuum of the relationship with the developer client. There is no suggestion that he now had two clients, one for the building in general and one for some or all of the windows in the café. Ernest Bewley was the client in relation to the other glazing in the café building and there is nothing to suggest that Harry Clarke believed that he had a different client in respect of the six windows or indeed in respect of two of the six windows.  He sought payment for all six windows without distinction from, firstly, the architects and then directly from Mr. Bewley. He did so at the same time as he also sought payment for items which were clearly for the account of the developer, i.e. Mr. Bewley.

100.      It must be acknowledged that Harry Clarke was not always entirely clear as to the identity of his client as he mistakenly referred to Ernest Bewley as Charles Bewley in circumstances where Charles Bewley was at that time living in Tasmania.  Also, as Ernest Bewley was a director of the limited company, the references to Mr. Bewley's involvement in and of itself is inconclusive that Harry Clarke was not also dealing with the company, the soon to be tenant, as opposed to Mr. Bewley qua developer of the premises. But it is clear that he never gave any indication that he had two clients, which is the implication of the decision of the High Court.

101.     Further, if there was a double fenestration arrangement for the two Swan Yard opes, either planned from the very beginning or in response to the experience of operating the café from November 1927 as the trial judge concluded, Harry Clarke in all likelihood would have provided the external clear glass glazing which was to be paired with his stained glass works. This is because he was very involved in every facet of the design, craft and fixing of his windows. Such an arrangement had the potential to impact negatively on his work and it is abundantly clear that he was very proud of the six windows. It is therefore significant, in my view, that there is no evidence that his firm was ever engaged to supply these outer glass windows and, more significantly, no such item is included in any of the accounts furnished to the two firms of architects or Ernest Bewley.

102.     It is perhaps also worth noting that Ms. Lucy Costigan gave evidence that there was no evidence in the Harry Clarke archive of a double fenestration system as hypothesised by Mr. O'Connell, while Dr. Caron stated that he had never seen such an arrangement amongst the thousands of stained glass windows he had viewed.

2. The lease 

103.     On 9 March 1928 Mr. Bewley as landlord entered into a lease with the first named respondent as tenant for a term of 21 years running from 1 April 1927.  The lease did not identify any tenant's fixtures and nor did it make any reference to stained glass windows. It provided that, on the determination of the lease, the tenant should deliver up possession to the landlord "together with all fixtures". If, as the trial judge concluded, the Four Orders windows were part and parcel of the premises and were an expense to be paid by Ernest Bewley as the landlord, but the Swan Yard works were tenant's fixtures because they were to serve the purposes of the tenant but not the landlord, it might have been expected that this unusual arrangement would have been spelled out in the lease and the Swan Yard works identified as tenant's fixtures to distinguish them from the Four Orders windows, which were not. In this regard it is relevant that the trial judge based his conclusion that the Swan Yard works were tenant's fixtures in part on the experience of the tenant operating the café between November 1927 when it went into possession and commenced trading, and March/April 1928 when the Swan Yard works were installed.

3. The minute of 9 March 1928

104.     The board of Bewley's Oriental Cafés Limited met on 9 March 1928 to approve the lease from Ernest Bewley to the company. Ernest Bewley took the chair. The meeting was attended by Mr. T.W. Bewley and Mr. G.A. Overend, solicitor, in his capacity as company secretary. The minutes record that a letter had been received from Ernest Bewley to his fellow directors referring to expenditure on 78-79 Grafton Street and then continued:-

"Mr. G.A. Overend, Solicitor, attended, and went through the proposed lease of the Grafton Street and Johnson's Court premises, with the Directors, and explained the various provisions therein.  The question was then raised as to the date from which the lease should run, and as to the rent which should be paid, and Mr. Overend explained that in as far as the Landlord was concerned, the premises were finished before April 1927, and that the work done thereafter consisted of fittings necessary to adapt the premises for use as a café, and which expenditure would in due course appear in the Company's accounts, as a Liability of its own.  He further explained that Mr. A.W. Shea of Messrs. Dockrells, had reported that the true basis in which the rent should be calculated was pto on the total cost to the Landlord, plus such ground rent as the Landlord is liable for.  Mr. Ernest Bewley then stated that having gone into the figures, he had ascertained that the total cost to him as landlord was about £45,000.0.0 made up of purchase prices paid by him for the various buildings, and the contractor's accounts, and that over and above this figure, he had spent, or authorised the expenditure on the company's behalf, of a sum of £14,900.0.0 largely comprising the cost of ovens, tables, chairs, counters and other café fixtures and fittings.  The Board then resolved, in view of this explanation, that the rent of £3,500.0.0, and the date of commencement of same, namely 1^st April 1927, seemed reasonable, and approved of the lease and counterpart, and authorised the seal of the company to be affixed thereon." 

105.     Objections taken by the appellant in the High Court to the admissibility of the minute were not maintained on appeal. Several points may be made in respect of this minute. Ernest Bewley was present at the meeting in two capacities: as the chairman of and shareholder in the company and as the owner and developer of the building to be leased by him to the family company. The key decision to be taken at the meeting was the date of the term of the lease and the rent payable. It was not primarily concerned with allocating the costs and expenses between the landlord and the tenant. Mr. Overend said (and Mr. Bewley did not dissent) that from the landlord's perspective "the premises were finished before April 1927, and that the work done thereafter consisted of fittings necessary to adapt the premises for use as a café, and which expenditure would in due course appear in the Company's accounts, as a Liability of its own." If this were to be construed as one would a certificate of completion by an architect, then the building would be deemed to be completed "before April 1927" which was before Harry Clarke's firm had been commissioned to design works for the six openings in the café, but in my judgment this would be incorrect, and it should not be construed in this manner.

106.     On 26 March 1927 Harry Clarke submitted proposals for additional work at the premises. On 27 April 1927 his firm submitted an account for various items including "sash under stairs" estimated on 28 March 1927, plus "2 doors similar [to] Gallery panels" and "2 windows [to the] side [of the] Large café", which were highly unlikely to have been furnished "before April 1927" and yet they were clearly for the account of the developer and not the proposed tenant. Furthermore, the minute says that the work done thereafter "consisted of fittings necessary to adapt the premises for use as a café" would be a liability of the company. It does not state that all work carried out thereafter would be the tenant's liability and it does not state that the landlord's works had entirely ceased. The minute elaborated on the expenses incurred by Mr. Bewley as landlord and continued that "he had spent, or authorised the expenditure on the company's behalf, of a sum of £14,900.0.0 largely comprising the cost of ovens, tables, chairs, counters and other café fixtures and fittings." This does not expressly include the commission of six large stained glass windows for the café. It is to be remembered that the windows were actually being installed around the time of the meeting and the total cost (£526.15.0) was a not inconsiderable expense. Further, if the trial judge was correct in concluding that the two Swan Yard works were tenant's fixtures but the Four Orders were not, then the failure of Mr. Bewley to mention the cost of the two Swan Yard works (£180) as an expense of the tenant, if it was in fact a tenant liability, which he had incurred on its behalf rather than for his own account, may be surprising.

107.     McDonald J., having reviewed the minute and considered the evidence of Ms. Costigan, concluded that it was not possible, based upon the minute, to conclude that the tenant paid for the six stained glass windows. At para. 79 he said that there was no evidence specifically identifying the ultimate payer and continued:

"...For all we know, the attendees of the meeting may have overlooked the stained-glass entirely. It may also be the case that the issue of the stained glass was the subject of a later board meeting or it may have been the subject of subsequent correspondence or discussions between Ernest Bewley and the company. Equally, the bill for the stained glass may have been settled by Ernest Bewley personally. In my view, it would be unsafe to take this one minute of the board of the tenant company, on its own, as sufficient evidence to ground a finding that, on the balance of probabilities, the tenant paid for the Harry Clarke works."

Conclusions based upon the evidence

108.     The trial judge very carefully analysed the position in relation to the Four Orders windows and concluded, on the balance of probabilities that these windows were installed in 1928 in the position shown in the 1998 video, replacing the presumed temporary glazing which had previously weathered the building since at least November 1927, and that they operated as windows at that time admitting light and ventilation to the café and also weathering the building from the elements. He therefore concluded that they formed part of the external skin of the café building at that time. He rejected the theory that there was a double fenestration arrangement and thus that the windows were in fact tenant fixtures.

109.     In light of the physical evidence from the building set out in paras. 88 - 103 I am quite satisfied that he was correct to reach this conclusion and I would reject the cross-appeal in relation to this conclusion. The frames in which the windows sit are the original frames. The stained glass sashes could easily have been inserted into the frames in 1928 in place of the temporary glazing securing the building, as the trial judge held likely occurred (para. 67 quoted above). The sashes were designed to face the exterior and showed signs of weathering, despite the fact that they have not been exposed to the elements since 1972. The windows were designed to open inwards to ventilate the café and it was never contended that they did not do so in the original arrangement. All of which suggests on the balance of probabilities that the Four Orders windows were designed and functioned as the exterior windows of the building.

110.      The contrary suggestion of double fenestration is simply neither supported by the evidence nor proved on the balance of probabilities by the respondents.  It is to be recalled that in assessing the evidence of the expert witnesses this court (and the court of first instance) is required to apply common sense and logic to the views of the expert.  Common sense and logic leads to the rejection of Mr. O'Connell's evidence in relation to the Four Orders windows.  First, and most importantly, there was no space to insert an outer layer of sashes glazed with clear glass between the frames holding the stained glass sashes and the bullnose bricks. Second, there was no physical evidence to support the existence of an interior arrangement, now entirely lost, which held the stained glass sashes in front of the existing frames which were glazed with clear glass. Third, the theory- which is only a theory- postulates an unnecessary, cumbersome means of mounting the stained glass without establishing any basis why this elaborate approach would have been adopted, and fourth, the theory does not explain how the windows would or could have operated to ventilate the café- a function they undoubtedly originally performed- particularly in the total absence of any physical evidence of any connection between the stained glass sashes and any other sashes or mechanism (other than the hopper mechanism) which would have been required to enable the windows to ventilate the café as intended.

111.     The Four Orders sashes constituted windows as defined in the case law in my judgment. In 1928 they weathered the building and without the sashes in place the building would not have been weathered. They lit and ventilated the room. The fact that the sashes were easily removable did not deprive them of their character as windows. They come within the meaning of windows in Holiday Fellowship and they were essential to the premises' convenient use as in Climie v. Wood. While Boswell was concerned with landlord's rather than tenant's fixtures, the reasons why the windows in that case were held not to be landlord's fixtures apply equally to this case to show that the Four Orders windows cannot be tenant's fixtures: they were part of the original structure of the building and were not "brought in" and "affixed" after the structure had been completed. The fact that there was, by inference, temporary glazing which was replaced by the Four Orders sashes does not alter this conclusion, in my judgment.

112.     Furthermore, I agree with McDonald J. that the fact that a sash window may be easily removed for maintenance or repair does not deprive it of the characteristic of a window or entitle a tenant to remove it as a tenant's fixture. To hold otherwise would potentially leave many buildings open to the elements, and I do not believe this to be the law.

113.     I should add that removing the Four Orders sashes, even if done without damaging the window frames, would, in my judgment, result in substantial damage to the freehold within the meaning of s.17 of Deasy's Act. It would leave the landlord with an incomplete premises which was not weatherproof or secure. The boxing in of the Four Orders windows described by McDonald J. was neither designed as, nor intended to function as the exterior windows of the building. In my view, this would cause substantial damage to the building and it is no answer to say that you could replace the existing windows with others. That would be to permit the tenant to first damage the building and then repair it, which is not what the section permits.

114.     I turn now to consider the Swan Yard windows. In my judgment the trial judge was correct to hold, and this court should not interfere with the finding of fact, that the window frame in the remaining window facing Swan Yard was the window frame originally inserted during the construction of the premises. This is clearly established by Mr. O'Connell's evidence.

115.     The Harry Clarke windows were not completed until March/April 1928 yet the tenant commenced trading as a café in November 1927.  It is therefore reasonable to infer that there was glazing in all six window frames by that date.  In the case of the Four Orders, it is probable that the sashes containing the temporary glazing were replaced and the sashes containing stained glass were inserted, as I have explained.  The Swan Yard stained glass sashes are equally capable of being inserted into the original Swan Yard window frame, according to Mr. O'Connell, whose evidence in this regard was not contested.   

116.     The exterior of Swan Yard sashes was designed and constructed in a manner to provide a weathering function. The trial judge did not advert to this important piece of the jigsaw, which is all the more surprising as he accepted the same evidence regarding the Four Orders sashes and relied upon it in reaching his conclusion as to the probable mounting of those stained glass sashes in 1928.

117.     The Swan Yard windows were designed so that they could ventilate the café.  The four sashes could open inwards and they were fitted with the same hopper mechanism as the Four Orders windows.  The High Court recorded that all witnesses agreed that the ventilation was a significant issue in the café as the room was heated by two open fires, the patrons smoked on the premises and the café operated a boiler which produced steam in the large room.

118.      There was no evidence of any attachment to the frame with the clear glazing which would enable the sashes to be opened either in tandem or in conjunction with the double fenestration arrangement posited by Mr. O'Connell.

119.     The trial judge held that the evidence was such that he could not exclude the possibility that the Swan Yard windows were originally installed in the window frames overlooking Swan Yard and that at some unknown point thereafter the stained glass sashes were removed from the window frames and transferred to mock windows in the locations observed in 1987. So far, I would agree with his judgment.

120.      He proceeded to observe that the physical evidence of the existence of a clear glass window at Swan Yard in 1987 is plainly inconsistent with the case that in 1928 the stained glass window was placed in the window opening now containing the clear glass window and that it was for the appellant to show that this "was not part of the original installation".  Undoubtedly, the appellant bore the burden of proof in respect of its case.  However, in my judgment, the trial judge erred in concluding that this meant the appellant was required to show that the arrangement in 1987 was not the original arrangement.  The appellant was required to prove, on the balance of probabilities, what the situation was in 1927/1928; not to disprove that the position in 1987 applied in 1928. It was beyond dispute that there had been changes to the arrangement of all six windows by 1987. It was therefore vital not to project backwards as such: the task was to consider the evidence from the 1920s and then to assess whether the evidence from later decades assisted in or altered the court's inferences considering all of the available evidence. Accordingly, the trial judge asked himself the wrong question in my view. 

121.     Looking at the evidence from 1927 and 1928 it is clear that Harry Clarke believed that he had one commission from one client to design six windows.  I would not agree with the trial judge's observation that this did not always mean that Harry Clarke "always had external windows in mind".  As I have said, he consistently distinguished between stained glass panels on the one hand and stained glass windows on the other and he never referred to the Swan Yard works as panels or in any way distinguished them from the Four Orders works. He sought payment for all six windows without distinction from the architects who were responsible for the development of the building and then from Mr. (Charles) Bewley. There is nothing in the Harry Clarke correspondence to suggest any difference in the commissioning or fixing of the six windows or that there was any redesigning of the manner in which the Swan Yard windows were to be fixed in the building, as is implicit in at least part of the trial judge's reasoning. 

122.     Secondly there is no evidence to suggest that Ernest Bewley personally was liable for and paid all of the accounts for all of the glass from Harry Clarke's firm with the exception of the Swan Yard works, which were, in the trial judge's view, for the account of the tenant. As I have already indicated, I do not believe that the minute of 9 March 1928 supports such an inference. If Ernest Bewley paid for the Four Orders windows, as the trial judge found, it is difficult to understand why he would have divided the obligation to pay for the set of six stained glass windows between himself and the company.

123.     The answer cannot be that the Swan Yard windows served the tenant's purpose, rather than the landlord's, while, per contra, the Four Orders windows served the landlord's purpose and he did not leave it to the tenant to complete this last, and significant aspect of the overall design of the café. As I have already explained, Ernest Bewley clearly had a specific vision for the entire building and it simply makes no sense that he would pay such attention to securing a harmonious design for the entire premises, with the exception of the two Swan Yard windows. If he also commissioned the two Swan Yard windows as part of the overall design of the building, it would be truly remarkable if he decided that these two elements alone were to be paid for by the tenant and were to be tenant's fixtures.

124.     Thirdly, all six windows were designed to ventilate the café and all six were fitted with the hopper mechanism.  There was no evidence that it was either necessary to treat the Swan Yard windows differently from the Four Orders windows or that in fact any decision to treat them differently was ever reached, whether at the time of commissioning the works or in the interval between the opening of the café to trade in November 1927 and the "fixing" of the Swan Yard windows in April 1928. In addition, there is no explanation as to why the hopper mechanism would be maintained on the Swan Yard windows if, by the time they came to be installed, those windows were not intended to ventilate the café, as the mechanism would be redundant.

125.     The trial judge then asked himself whether there was no possible reason why they might not be treated differently and then speculated as to possible reasons for the different treatment between the two sets of windows.  In my judgment, this involved asking the incorrect question and answering it by reference to speculation and thus was unjustified and erroneous.

126.     McDonald J. first speculated that as the windows faced onto Swan Yard they were at a greater risk of vandalism or posed a greater security risk to the premises than the Four Orders windows. The respondents contend that this was in fact inconsistent with the evidence which showed that the Four Orders yard was accessible from a schoolyard and thus there was no distinction between the two locations as regards risk of damage or security.  Furthermore, there was no evidence that this was a concern at all at the time or that it was a greater concern in respect of Swan Yard rather than the Four Orders yard.  Secondly, he failed to have regard to the evidence that wire grilles were employed to protect stained glass windows against vandalism and to protect them generally in the 1920s and that such a grille was observed to be in situ in 1987 by Mr. Campbell.  The trial judge held that it was improbable that the grille was an original feature because of the risk of shadows been cast by the grille onto the stain glass window, a matter with which Harry Clarke was known to be concerned.  However, this failed to have regard to the evidence of Dr. Caron to the effect that this was the method used at the time and that the grilles did not cast a shadow as they diffused the light. If it was a commonly employed device, as the uncontroverted evidence established, it is reasonable to infer that it had been used without objection to protect other Harry Clarke windows.  Finally, an outer layer of clear glass is not an effective protection against accidental or deliberate damage.  The wire grille would be at least as effective in protecting the windows from damage and securing the premises and there was evidence that such a grille was in situ in 1987, though admittedly there is no actual evidence that this was so from the 1920s.

127.     So for all of these reasons I do not think it was correct to infer that the hypothetical risk of damage to the windows in Swan Yard could support an inference that the windows overlooking Swan Yard were treated differently to those on the western façade.

128.     The second possible reason advanced by the trial judge for the difference in treatment between the Four Orders windows and the Swan Yard windows was the risk of damage to the windows from sacks of flour being hoisted to the bakery above the café or from flour dust. It is important to note that there was no evidence whatsoever in relation to sacks of flour being hoisted or of any flour dust permeating the café as a result.  Mr. Campbell, a witness as to fact, said that he was speculating why one of the two Swan Yard windows was blocked up.  There was no evidence at all that the hoisting of these sacks of flour in front of one of the windows was such a cause of concern in 1927/8 that it was deemed necessary to protect the two Swan Yard windows from damage by utilising a double fenestration configuration in 1928 when they came to be installed.  If the risk of damage from flour sacks and/or flour dust was met by the installation of the double fenestration arrangement, it begs the question why subsequently the window directly beneath the hoist was bricked up entirely. In addition, this hypothetical requires that there was a change in design between the commissioning of the works in 1927 and the fixing of the stained glass windows in 1928 and there is not a shred of evidence to suggest that this occurred. At the very least one would expect that the additional work would have been undertaken by Harry Clarke's firm and that he would have sought payment for the extra work arising from a redesign, but there is no such evidence.

129.     Thirdly, the trial judge held that the presence of the hopper mechanism was not sufficient for him to infer that the Swan Yard windows functioned as a direct source of ventilation either at the time of the photograph of the café from the 1970s or 1980s or in 1928.  The trial judge acknowledged that the Swan Yard windows were fitted with the hopper mechanism, and they were designed to open inwards and to provide ventilation.  He noted that there were obvious difficulties with ventilation if the double fenestration arrangement postulated by Mr. O'Connell and observed in situ in 1987 was employed.  The trial judge said that the fact remained that in 1987 the Swan Yard works were not providing any ventilation function to the café at that time. He said "[i]f the ventilation function was not considered necessary at that time, why would the position have been any different in 1928?".  He speculated that in the intervening period between the opening of the café and the installation of the Swan Yard works in April 1928 there may have been a reassessment of the need for ventilation from these two windows at this location in the café (or whether there were other factors potentially in play that outweighed the need for it there such as the issue of security previously discussed or indeed the ingress of flour dust from the haulage of sacks of flour to the bakery above).  

130.     I have already said why I do not believe it was open to the court to speculate on the issue of security or the question of flour dust from the haulage of sacks of flour to the bakery.  It is also important to note that the conclusion of the trial judge necessarily involved a review or revisiting of the original plan in light of the experience while operating the café.  This is because the design of the windows to provide ventilation would be otiose if it had been the original intention to install a double fenestration arrangement in the Swan Yard openings.  The trial judge himself posited concerns that arose in the light of experience operating the café (at least in part) and therefore his conclusion requires that the client (be it the landlord or the tenant) change their mind in relation to the arrangement and redesigned the mechanism in the six months between 1927 and the installation of the Swan Yard windows in April 1928.  There is no scintilla of evidence to support the contention that this occurred and it would be most surprising if there had been no mention of it at all in Harry Clarke's correspondence.  Furthermore, I do not agree with the trial judge's suggestion that if the Swan Yard windows (or indeed the Four Orders windows) were no longer required to ventilate the café in 1987, that they were likewise not so required in 1928.  This ignores the fundamental difference that in 1927 the café was heated with two open fires and that this was no longer the case by 1987.  Thus, the requirements for ventilation in 1987 were not comparable to those in 1927/1928.

131.     It follows, in my view, that the trial judge failed to place sufficient weight on the need to ventilate the café, the fact the Swan Yard windows were designed to open inwards and help to ventilate the space, they were fixed with the same hopper mechanism to facilitate the coordinated opening of the sashes which would be otiose and finally, if the windows were no longer required to ventilate the café, it was not necessary to redesign them by installing a novel, complex double fenestration mechanism. The alternative was simply to shut them.  

132.     At para. 163 the trial judge concluded:

"The reality is that there is no evidence before the court which provides a reliable basis on which to conclude, on the balance of probabilities, that the double window arrangement observed in 1987 was not also in place in 1928 when the Swan Yard works were installed. ...I do not believe that it is possible to hold that the plaintiff's postulated version of events is more probable...  It follows that the plaintiff has failed to establish on the balance of probabilities that, in 1928, the Swan Yard works were installed in the window openings in the wall overlooking Swan Yard."

133.     For the reasons I have set out, I would disagree with this conclusion of the trial judge.  The appellant was not required to prove that the window arrangement observed in 1987 was not also in place in 1928 when the Swan Yard works were installed.  The appellant was required to prove on the balance of probabilities whether the Swan Yard works were windows in 1928.  This, in turn, requires the appellant to prove, on the balance of probabilities, that the stained glass sashes were installed in the original outer frame in 1928 (and by implication that the double window arrangement observed in 1987 was therefore a later configuration).  I am of the opinion that the appellant has proved this to be the case on the balance of probabilities because:

(1)     At all times Harry Clarke and the architects treated this as one commission, not two separate commissions.

(2)     There is nothing to indicate that there were two commissioning clients.

(3)     All six windows were designed to ventilate the café by opening inwards and all six were provided with a hopper mechanism designed to open the windows in tandem.

(4)     A double fenestration arrangement would, at the very least, significantly reduce the ventilation function of the Swan Yard windows.

(5)     There was no physical evidence of any mechanism ever applying to the existing clear glass sash in the Swan Yard window to facilitate the opening of these sashes in tandem with the stained glass sashes.

(6)     It was the practice in the 1920s to protect stained glass windows with a wire grille which suggests that other Harry Clarke windows were so protected without his objection.

(7)     There was evidence that a wire grille was protecting the opening overlooking Swan Yard in 1987.

(8)     If the double fenestration arrangement was designed to protect the stained glass windows from possible danger from flour sacks or flour dust, it does not explain why subsequently one window was bricked up and the stained glass mounted as a faux window in front of the original opening.

(9)     An outer layer of clear glass windows would not provide very effective protection against either accidental or deliberate breakage and would provide no better security than a wire grille.

         

In my judgment, if one considers the combined effect of these points and then consider the

improbability of the decision of the High Court of providing one arrangement for the Four

Order windows on the one hand and a separate arrangement with a different owner for the

Swan Yard windows on the other hand, the trial judge's conclusions are improbable, and,

in my view, insufficient to justify a rejection of the case made out by the appellant by a

combination of direct evidence and reasonable inference.  This leads me to conclude that,

contrary to the judgment of the High Court, the appellant has on the balance of

probabilities established that in 1928 the Swan Yard windows were installed in the external

openings in the wall overlooking Swan Yard as the original external windows of the

building and there was no double fenestration arrangement.  It follows that in my judgment

there was no distinction between the Four Orders windows and the Swan Yard windows

and just as the Four Orders windows were windows as a matter of law, so too are the Swan

Yard windows as they are also part and parcel of the premises.  It follows that I also do not

agree that they were tenant's fixtures or that the first named respondent paid for the two

windows.

The sale and leaseback of 1987

134.     The question then is whether the sale and leaseback of the premises in 1987 alters in any way the title to the six windows. I agree with the High Court that the Four Orders windows as part and parcel of the building in 1987 passed under the sale of the property in the normal way without the requirement to refer specifically that they formed part of the sale. However, the position in 1987 with regards to the Swan Yard windows was different. The evidence is that by then the Swan Yard works were not part of the exterior of the building, they did not weather it, nor did they ventilate or illuminate the room with natural light. The stained glass sashes were mounted in frames in front of either a blocked up former window or a window glazed with clear glass. The issue is whether they were still windows or whether they had lost the character of windows.

135.     I agree with the trial judge that the fact the sashes for all six windows were removed for safe keeping during the war for a number of years did not deprive them of the character of windows. Nor did the Four Orders windows lose their status as windows when they were removed for several months for restoration in 1998. In each case the removal was temporary and the clear intention was to restore the windows as the windows of the building in due course. The passage of time, even though lengthy in the case of the war, did not alter this (particularly as the reason for the latter removal continued throughout that period and cannot be said to have been in any way indicative of a decision by the landlord to permanently alter its premises). Can the same be said in respect of the Swan Yard windows in 1987?

136.     In respect of the blocked up window the answer to this question must be yes. I cannot agree with the submissions of counsel for the appellant that the frame with the stained glass sashes mounted in front of the blocked up window nonetheless remained in law a window in those circumstances. It could no longer perform any of the functions of a window identified in the case law and considerable works to the premises would be required to enable the frame and sashes to be inserted into the ope again. In my judgment by 1987 the Swan Yard work mounted in front of the closed opening was no longer a window. However, that did not mean that it had become the property of the tenant. That case was never made in either evidence or argument. It follows that the frame with the stained glass sashes must be regarded as an element which was no longer part and parcel of the building, which was still the property of the landlord but which was nonetheless affixed "during the currency of the lease" to the premises. As such, it may best be regarded as a landlord's fixture. That being so, it too passed under the sale of the property without the need to specify that it was included in the sale in the same way in which fixtures in situ have always passed unless expressly excluded from the sale.

137.     The special conditions of sale do not alter this conclusion. The fact that "purchased chattels" are a defined term in the contract merely adds to the matters included in the sale. The definition cannot exclude from the sale matters which are otherwise included. If it was the intention of the parties that the sale would not include the Swan Yard works these would have to be expressly excluded from the sale to prevent them from passing to the purchaser. This is not what the contract provided. It follows that, despite the change of legal character of this Swan Yard window, title to the window passed under the 1987 sale.

138.     The position in relation to the Swan Yard work which was mounted in front of the clear glazed window is less clear. There was no evidence when the double fenestration arrangement commenced, whether it was intended to be temporary, or how long the clear glass glazing was in place, so it is not possible to conclude with certainty that by 1987 the second Swan Yard window had also lost the legal character of a window. I prefer not to reach any definite conclusion in relation to this issue and, happily, it is not necessary to do so in order to determine this appeal. As I have explained, regardless of whether the Swan Yard windows were part and parcel of the building (and thus retained the character of windows) or whether they became landlord's fixtures, they passed under the sale because they were not expressly excluded from the sale.

139.     In light of these conclusions it is not necessary to consider the other issues raised in the appeal and cross-appeal or any of the other matters discussed in the judgment of the High Court.

Conclusions

140.     The trial judge was correct to conclude that the Four Orders windows were windows as they were part and parcel of the premises. They weathered, lit and ventilated the café. In law, they were and remained the property of the original landlord and his successors in title. They passed under the sale in 1987 as part and parcel of the building and the appellant, as the successor in title is the owner of the Four Orders windows.

141.     The trial judge erred in concluding that the appellant had failed to prove on the balance of probabilities that the Swan Yard windows also were part and parcel of the building and in concluding that, instead, that they were tenant's fixtures and therefore the property of the second named respondent on foot of an assignment from the first named respondent. In my judgment the appellant proved on the balance of probabilities, having regard to the totality of the evidence, that the Swan Yard works were part and parcel of the premises and thus were the property of the original landlord and his successors in title. The trial judge's conclusion was based upon impermissible speculation which must be rejected in light of the entirety of the evidence and the inferences which it is reasonable to draw from such evidence.

142.     These conclusions determine the ownership of the six windows and it follows that I would allow the appeal of the appellant in relation to the trial judge's ruling as to the ownership of the Swan Yard works and I would reject the cross-appeal of the respondents in relation to the trial judge's conclusions as to the ownership of the Four Orders work.

143.     As the appellant has been wholly successful on appeal, my provisional view is that it should be entitled to the costs of both the appeal and the trial, to be adjudicated in default of agreement. If any party wishes to contend for a different order as to costs, its solicitors must contact the Office of the Court of Appeal within ten days of the delivery of this judgment. In that event, both parties should furnish written submissions of no more than 2000 words, concerning the costs of the High Court proceedings and the appeal, to be exchanged by 10 October 2024.

144.     Pilkington J. has authorised me to indicate her agreement with this judgment.

 

Result:     Appeal Dismissed