Majorstake Ltd v Curtis

This appeal raises a short point of statutory construction of some importance to landlords and tenants of residential flats. As with many issues of this kind, the question is easier to pose than to resolve.

The appellant, Majorstake Ltd (“the landlord”), is the freehold owner of Boydell Court, St John’s Wood Park which comprises two blocks of flats known as Block A and Block B. The respondent, Mr. Curtis (“the tenant”), is the lessee of Flat 77 on the 7^th floor of Block B under an underlease dated 22^nd July 1957 for a term of 51 years from 25^th March 1957. The contractual term of the lease expires on 25^th March 2008.

One of the many purposes of the Leasehold Reform, Housing and Urban Development Act 1993 (“the Act”) was to confer on tenants of flats held under long leases the right to obtain a new lease on payment of a premium determined in accordance with the terms of the Act, subject to certain provisions to safeguard the interests of landlords in redeveloping their property. The right to obtain a new lease is exercised by the service on the landlord of a notice under section 42 of the Act. On receipt of such a notice the landlord is required to give a counter-notice under section 45 stating that he admits (or does not admit, as the case may be) that the tenant has a right to acquire a new lease but (if it be the case) that he intends to make an application to the court for an order under section 47 declaring that the right to acquire a new lease shall not be exercisable by the tenant by reason of his intention to redevelop any premises in which the flat is contained.

On 16^th September 2003 the tenant gave notice to the landlord under section 42 that he claimed the right under the Act to acquire a new lease. It is not disputed that he was a qualifying tenant within the meaning of the Act and was therefore entitled to do so. On 21^st November 2003 the landlord served a notice under section 45 admitting the tenant’s right to claim a new lease but stating that it intended to apply for an order under section 47 that that right should not be exercisable on the grounds that it intended to redevelop premises in which the flat was contained. It is accepted that the notice complied with the requirements of section 45(2)(c).

On 20^th January 2004 the landlord issued proceedings in the Central London County Court seeking a declaration that the tenant’s right to acquire a new lease of Flat 77 Boydell Court was not exercisable by reason of the landlord’s intention to redevelop premises in which the flat was contained. In his witness statement filed in support of the application one of the landlord’s directors, Mr. Elan Sasha, explained that it was the landlord’s intention to redevelop Flat 77 by combining it with the next door flat, Flat 79, to create a single large apartment containing four reception rooms, four bedrooms, kitchen and three bathrooms. The plan was to reduce each flat back to a shell by removing all internal walls and then to fit new windows, erect new internal partition walls, rewire, replumb, fit new bathrooms, lay new floors and redecorate. By the time the application came to be heard, however, the landlord’s plans had changed. In a witness statement dated 1^st December 2004 another director, Mr. Sony Douer, explained that it was now the intention to combine Flat 77 with Flat 74, which is situated immediately below it and is owned by a subsidiary of the landlord, to form a single duplex apartment. That also involved reducing each flat back to a shell, installing a staircase between the two levels, renewing the windows, reordering the interior on both levels and replacing the services. It was accepted before the judge that the landlord had established its intention to redevelop Flat 77 as part of the plan described by Mr. Douer, that it involved substantial works of construction in both flats and that it was necessary for the landlord to obtain possession of both flats to enable it to carry out the proposed works. The main issue between the parties, and the only issue that arises on this appeal, is whether that part of Block B which comprises Flats 77 and 74 constitutes “any premises in which [Flat 77] is contained” within the meaning of section 47(2)(b).

Section 47 therefore lies at the heart of this appeal. It forms part of Chapter II of the Act which comprises those sections dealing with the right of certain tenants holding under long leases as defined by the Act to acquire new leases of their flats giving them an additional term of 90 years. Chapter I, to which it will be necessary to refer in a moment, is concerned with the right of collective enfranchisement by which, subject to certain conditions, tenants may obtain the transfer of the freehold of the building containing their flats to a nominated third party to be held on their behalf.

Section 47 itself provides as follows:

The landlord’s case before the judge, as before this court, was that the expression “any premises” is perfectly general and is therefore apt to refer to the whole or any part of a building. (Indeed, before us Mr. Brock Q.C. initially submitted that it was apt to refer to two or more flats in widely separated parts of a building, but in the context of the expression as a whole which carries with it the idea of containment, I do not think that they can extend quite as far as that.) The tenant’s case was that, having regard to the context in which it is found, the expression taken as a whole was intended to refer to the building as a whole or a self-contained part of it.

The judge preferred the tenant’s construction for three reasons. His first and main reason was that he believed it to be consistent with the use of the same expression elsewhere in the Act. He noted that under section 41(4) the tenant is entitled to obtain from persons with a superior interest in his flat information about a number of matters likely to affect his claim, including the service of a claim under section 13 to purchase the freehold of the building containing his flat pursuant to the collective enfranchisement provisions of Chapter I. Section 41(4) provides as follows:

The judge noted that subsection (4)(a) also contains the expression “any premises containing the tenant’s flat” and refers back to section 13. That led him back into Chapter I and to section 3(1)(a) which provides that the provisions of that chapter apply to premises which consist of a self-contained building or part of a building. In the light of that he concluded that the expression “any premises in which the [tenant’s] flat is contained” in section 47 was to be construed in the same way and thus meant the building as a whole or any self-contained part of it, in this case Block B.

The judge’s second reason for preferring the tenant’s case was his instinctive feeling that the landlord’s proposals were simply a device to frustrate the tenant’s lawful expectation. He pointed out that the landlord did not need possession of Flat 77 to carry out work to Flat 74 as such (although it could not, of course, carry out the particular work it had in mind without obtaining possession of Flat 77) and it appears that he thought that the landlord was taking advantage of the fact that it had control of Flat 74 to enable it to obtain possession of Flat 77. He noted that the landlord was not seeking to redevelop a large part of the property, such as the whole of Block B, which he thought might properly be considered to represent the interest of the landlord which the statute was intended to protect, but was merely seeking to do work on two individual flats in relation to which he thought the legislation was intended to protect the tenant’s interest. His third reason was that if Flats 77 and 74 could be regarded as premises in which Flat 77 was contained, he did not think that Flat 74 was a substantial part of those premises within the meaning of section 47(2)(b).

I am afraid I do not find the judge’s reasons persuasive. Although I agree that it is essential to construe section 47 in the context of the Act as a whole, it is also necessary to bear in mind that Chapter I is dealing with a fundamentally different subject matter from that covered by Chapter II. That does not necessarily mean that Chapter I can be excluded from consideration altogether, but it does mean that it is necessary to read Chapter I with that distinction clearly in mind.

The word “premises” is used in a variety of different senses in the Act. Sometimes it is used to refer to a single flat (as in the definition of a flat in section 101 (c)), sometimes to a part of a building in which a flat is contained and sometimes to the building as a whole. In the light of that it was common ground that it is impossible to attribute any single clear meaning to the word and that it is necessary to examine the context in which it occurs in each case to understand the particular sense in which it is used. In my view the judge failed to take sufficient account of the fact that, although the same phrase is used in Chapters I and II, the two chapters are dealing with fundamentally different matters and that the only purpose of s.41(4) is to enable the tenant to obtain information about any claim for collective enfranchisement that may have been made in respect of the building as a whole. He needs that information, both because he may wish to take advantage of any such claim himself and because a pending collective enfranchisement claim under the provisions of Chapter I operates to suspend the tenant’s right to obtain a new lease under the provisions of Chapter II. Since a collective enfranchisement claim can only relate to a self-contained building or part of a building, the expression “premises containing the tenant’s flat” in section 41(4)(a) must inevitably refer to a self-contained building or part of a building, but there is no reason why it should be construed in the same way in sections 45 or 47 which are concerned with the acquisition by the tenant of a new lease in an individual flat.

The judge’s second reason, as he himself recognised, was little more than an instinctive reaction to the landlord’s case. A finely-tuned instinct may often be a valuable guide in resolving legal problems, but when dealing with legislation of some technical complexity such as that with which this case is concerned, it is no substitute for careful analysis.

His third reason was really no more than an extension of his first, since it assumes that the purpose of sections 45 and 47 is to preserve the landlord’s right to redevelop the building as a whole, or at any rate a substantial part of it, rather than to allow him to redevelop different parts of it on a piecemeal basis. However, that is the very question at issue in this case. Moreover, the suggestion that it would not be right to say that Flat 74 is a substantial part of the premises misses the point. The premises for the purposes of the argument are Flats 74 and 77 viewed together. The landlord clearly did intend to carry out works of construction on substantially the whole of those premises.

As I have already observed, the word “premises” is used in a different senses in different parts of the Act and its meaning in any particular instance depends to a large extent on the context in which it is found. The expression we have to construe is “any premises in which the [tenant’s] flat is contained” in the context of section 47 and it is important to read it as a whole when seeking to identify its correct meaning. On behalf of the landlord Mr. Brock emphasised the use of the word “any” and submitted that the expression as a whole naturally refers to the building as a whole or any part of it which includes the tenant’s flat. This became known as the “red line” argument since it involves the proposition that, provided he can satisfy the court that he has a genuine and realistic intention of doing so, the landlord can redevelop any part of the building containing the tenant’s flat which is capable of being identified by a continuous line drawn on a three-dimensional plan of the building. Mr. Denehan for the tenant accepted before us that the expression could not be restricted to the building as a whole or even a self-contained part of it, but he submitted that on its ordinary and natural meaning it referred to an existing recognisable unit, such as a whole floor of the building, which contains the flat in question.

Chapter II of the Act seeks to strike a balance between the interests of tenants holding under existing long leases to acquire new leases and thereby obtain additional security of tenure and the interests of landlords in developing and renovating their property. In striking this balance Parliament has hedged about the landlord’s right to obtain possession of an individual flat with the various requirements set out in section 47(2). These include a requirement that the existing lease be due to terminate within five years of the date of the tenant’s notice, a requirement that the landlord intends to demolish or reconstruct or carry out substantial works of construction on the whole or a substantial part of the premises in question and a requirement that the landlord be unable to undertake that work without obtaining possession of the tenant’s flat. Two things follow from these. The first is that the tenant is not at risk of losing his right to acquire a new lease until his original lease has almost come to an end in any event. The second is that the landlord must be able to satisfy the court that he has a genuine and realistic intention of carrying out redevelopment work of a substantial kind for which he needs to obtain possession of the tenant’s flat. Section 47(2) appears to have been closely modelled on section 30(1)(f) of the Landlord and Tenant Act 1954 and the court can therefore be expected to apply broadly the same principles when considering an application under section 47 as it does when considering opposition by a landlord to the grant of a new tenancy under section 30(1)(f). The statute itself reflects Parliament’s decision as to how the balance between the interests of landlords and tenants is to be held.

Once it is accepted, as it was on both sides, that the premises to which section 47(2)(b) refers may be something less than the whole building or a self-contained part of it, I find it difficult to limit its meaning by reference to some notional unit within the building. If the landlord wishes to redevelop two adjacent flats to create a single larger unit, as the landlord in this case originally had in mind in the case of Flats 77 and 79, I think he can properly say that the balance of interests reflected in sections 42 to 47 favours redevelopment rather than the grant of a new tenancy. That would certainly be the case if he intended to redevelop the whole building and I can see no reason why it should be otherwise just because the redevelopment will extend to only part of it. As far as the individual tenant is concerned, the effect, of course, is the same. The expression “any premises in which the flat is contained” is very broad. If Parliament had intended the tenant’s right to acquire a new lease to be defeated only in cases where the landlord intended to redevelop the building as a whole, or a self-contained part of it, that could easily have been achieved by a provision similar to that found in section 3(1)(a). Similarly, if Parliament had intended that the premises in question should represent an existing recognisable section of the building, it would have been possible to define “premises” for the purposes of section 47 in appropriate terms. In section 101(1), for example, a flat is defined as a separate set of premises, whether or not on the same floor. The concept of a set of premises is one which could have been employed to identify an existing recognisable part of the building.

In my view the landlord’s argument gains some additional force from the use of the words “any premises” rather than “the premises”. Sections 40 and 41 recognise that there may be a number of persons with superior interests in the tenant’s flat and that the premises in which their respective interests subsist may not all be the same. Moreover, since any new lease acquired under these provisions will be for an additional term of 90 years, the relevant landlord for the purposes of Chapter II must hold either the freehold or an interest in the flat for a term of sufficient length to enable him to grant the tenant a new lease of that duration. In order to accommodate these circumstances section 40 provides that the landlord for the purposes of Chapter II is the person who holds the first superior interest capable of granting a term of 90 years. If in sections 45(2)(c) and 47 Parliament had used the words “the premises”, it might have been possible to argue that it intended to refer to the whole of the premises in which the relevant landlord’s interest subsists, whether that was the whole building or some part of it. The question would then be whether those premises “contained” the flat (they would not do so if they did not extend beyond the flat itself) and whether the landlord intended to redevelop the whole or a substantial part of them. In many cases a landlord whose interest subsists in only part of the building would not be capable of carrying out work of sufficient substance to fall within section 47(2)(b).

The concept of containment underpinned Mr. Denehan’s submission that the expression “any premises in which the [tenant’s] flat is contained” is intended to refer to some existing recognisable unit within the building such as a whole floor or part of a floor which is physically separated in some way (for example by the stairwell or lift shaft) from the remainder. The concept of containment is undoubtedly an essential element. It is what prevents the landlord from redeveloping the flat on its own since the premises in which the flat is contained must by definition be larger than the flat itself. (It also, in my view, prevents him from treating two separated parts of the building as the relevant premises for this purpose since in this context the word “premises” connotes one undivided part of the building as a whole.) The expression “premises in which the flat is contained” is clearly capable of referring to a whole floor of a building, but the difficulty with his argument is that it is just as capable of referring to part of a floor which includes the flat in question, even though it would not otherwise be recognised as a defined unit in any ordinary sense.

Another difficulty I have with Mr. Denehan’s submission is that it is liable to produce arbitrary consequences. For example, a landlord who intends to combine two or more adjacent flats in the manner originally proposed by the landlord in the present case would be in a position to do so if together they represented the whole floor on one side of the staircase or lift shaft, but not if they represented only part (albeit the larger part) of that section of the floor. Similarly, a landlord who wished to combine flats on two adjacent floors, as the landlord in the present case now intends, would be able to do so if the redevelopment work involved a substantial part of the floor on which the tenant’s flat was situated, but not if it involved only the tenant’s flat and perhaps a few others on the same floor. In the absence of any clear indication in the language of section 47(2)(b) that this is what Parliament intended I find it difficult to accept that his construction is correct.

The demise in the present case extends to half the depth of the floor and ceiling joists between Flat 77 and the flats above and below it as well as the internal and external walls between those levels. However, it was pointed out in the course of argument that in some cases it might not extend beyond the internal plaster coating of the walls and ceiling of the flat, leaving the brick and timber structure in the possession of the landlord. In those circumstances could the landlord maintain that the solid structure surrounding the flat represented “premises in which the flat is contained” within the meaning of section 47(2)(b)? If so, it would provide a strong argument against the landlord’s construction since it would effectively enable him in those circumstances to obtain possession of a single flat for the purposes of redevelopment.

In my judgment such an argument seeks to press the linguistic analysis of section 47(2)(b) too far. Although it could be said in such a case that the flat demised was “contained in” the premises represented by the solid structure, it was clearly not the intention of Parliament in my view to draw a distinction of that kind. The definition of a flat as “a separate set of premises” which is “constructed or adapted for use for the purposes of a dwelling” does not easily lend itself to a distinction of that kind and in my view to treat the solid structure as separate from the flat itself would be to frustrate the will of Parliament rather than to give effect to it.

The so-called “box room” argument presents greater difficulty. Like the previous argument it depends on whether for the purposes of section 47(2)(b) the flat can properly be regarded as being “contained in” any premises which comprise the flat itself and some other insignificant part of the building such as an adjacent box room accessible from the common parts and currently in the landlord’s possession. Again, it is said that, if it can, it would enable the landlord to obtain possession of an individual flat for the purposes of refurbishment and would thus suggest that the construction being advanced on behalf of the landlord in this case cannot be correct.

It seems to me that, if the landlord’s argument is accepted, it must also be accepted that the flat and the box room together do constitute premises in which the flat is contained and that the landlord would be able to defeat the tenant’s right to acquire a new tenancy if he could satisfy the court that he intended to carry out substantial works of construction on the whole or a substantial part of those premises taken as a whole. However, although I doubt whether Parliament had a situation of this kind in mind, I do not think it conflicts with the purposes of the statute in the way that the previous argument does. Moreover it should be noted that the argument gains much of its force from the use of an insignificant additional area represented by a box room. It would not appear as attractive if the landlord’s intention were to incorporate into the existing flat an adjacent area of much more significant proportions, although it is difficult to see where the difference of principle might lie. Moreover, it is unlikely that the expression “substantial works of construction” in section 47(2)(b) will be construed any more liberally that the corresponding expression “substantial work of construction” in section 30(1)(f) of the Landlord and Tenant Act 1954 and therefore, taking the example of the box room, if the landlord intends to carry out major work of that kind on a substantial part of the premises the existing flat will inevitably be the primary focus of attention. I do not think that it would be contrary to the purposes of the legislation to allow a landlord to obtain possession in order to carry out a redevelopment of that kind. I strongly suspect, however, that opportunities to take advantage of the legislation in this way will in practice be few and far between.

Mr. Denehan submitted that each case would depend on its own facts and to some extent that must be true, but I do not think that this is one of those cases in which the court is concerned with the meaning of an ordinary word or expression of the kind that was considered in Cozens v Brutus [1973] A.C. 854. Both landlords and tenants need to be able to organise their affairs on the basis of a clear understanding of the effect of these provisions and for that purpose it is necessary to decide what Parliament meant by the expression under consideration in this case.

In the end I am unable to accept Mr. Denehan’s submission because I do not think that it is possible to identify some irreducible minimum unit that constitutes “premises in which the [tenant’s] flat is contained” for the purposes of sections 45 and 47 of the Act. The creation of a single maisonette out of two vertically adjacent flats does not seem to me to run counter to the purposes of the statute and that part of the building which comprises the two flats can properly be described as premises containing Flat 77 just as much as can the seventh floor of the building on which it is situated. In the end I am unable to find either in the policy or the language of the Act any indication that a part of the building which comprises two adjacent flats does not constitute premises in which each is contained, whether the two flats be vertically or horizontally contiguous.

In these circumstances I have come to the conclusion that the landlord’s construction is correct and that the appeal should be allowed.

The relevant facts and the statutory background to this appeal are set out in the judgment of Moore-Bick LJ, and I gratefully adopt what he says. I agree with his conclusion and with his reasoning, but, because the point at issue is not easy, of some potential general significance, and has resulted in a divergence of judicial views, I shall give my reasons in my own words.

The issue in this appeal turns on the meaning of the expression (“the expression”) “any premises in which the [tenant’s] flat is contained” in sections 45(2)(c), 47(2)(a) and 47(2)(b) of the Leasehold Reform, Housing and Urban Development Act 1993 (“the 1993 Act”). Any section referred to in this judgment is in the 1993 Act unless the contrary is stated.

Does the expression refer to any unit of property provided that (a) it includes the whole of the flat and (b) it also extends to other property, as the appellant landlord contends? Or is it restricted to an existing recognisable unit of property which cannot be characterised as artificially assembled for the purposes of enabling the landlord to rely on section 47, as the respondent tenant argues? Or is it limited to the premises defined in section 3, i.e. the block of flats in which the flat concerned is situated, or a vertically severable part of that block, as the judge held?

The third of these possibilities (which Mr Denehan, on behalf of the respondent, realistically did not pursue) must, in my view, be rejected. Considering the wording of sections 45(2) and 47(2), as a matter of ordinary language, that construction involves an artificially restricted interpretation of the expression. It is also not entirely easy to reconcile with the inclusion of the word “any” in the expression. Additionally, it is not, despite the judge’s view to the contrary, supported by the provisions of section 41(4)(a). I accept that those provisions contain words very similar to the expression, and that the “premises” there referred to would normally be limited to property falling within section 3. However, that is because the ambit of the relevant words is expressly limited by the reference to section 13 in paragraph (i) of section 41(4)(a), which takes one to section 3. Further, adopting the section 3 definition for the purposes of sections 45 and 47 produces anomalies: it would, for instance, exclude a building with one flat (e.g. the familiar shop with flat above) from the ambit of the expression.

So the choice is between the very broad meaning propounded by the appellant, and the somewhat subjective meaning favoured by the respondent. I have come to the conclusion that, while it is not without its difficulties and problems, the appellant’s interpretation is to be preferred.

The appellant’s construction seems to me to accord (subject to one possible point) with the natural meaning of the expression, and to give rise to a simple test which is easy to apply, albeit possibly rather generous to landlords. There should rarely, if ever, be any difficulty in deciding whether premises include the whole of a flat, and some other property.

However, the appellant’s construction may appear, at least to some people, not to involve an entirely natural use of language. A combination of two flats, one above the other, in a large block such as Boydell Court, can fairly be said to constitute “premises” in which each flat is “contained” particularly in the eyes of a conveyancer, but some might regard it as a slightly artificial use of ordinary language.

At least to my mind, there does not appear to be any artificiality in the appellant’s construction; if there is, it is certainly not particularly striking. The word “premises” particularly when preceded by the word “any”, has a naturally very wide meaning. As for “contained”, the appellant’s case does not, in my judgment, attribute an artificial meaning to it. In particular, the notion that the appellant’s case is weakened by the use of the word “contained”, rather than “included”, in the expression does not seem to me to undermine, to any extent, the appellant’s case. On the contrary; the two words appear, at least to me, to be virtually synonymous. Any attack on this basis is also weakened by the fairly poor drafting of the 1993 Act, and by the fact that it can be said to be a conveyancing statute, so a reading which would seem natural to conveyancers is relatively easy to accept.

The appellant’s construction is, however, not quite as simple as it seems. If a landlord owns two flats which are not contiguous (e.g. two flats on unadjacent floors), the combination of the two flats could be within the wide meaning of “premises”, particularly to a conveyancer. (A transfer or lease of two such flats might well describe them together as “premises”). Yet it seems to me that Mr Brock QC, who appeared for the appellant with Miss Windsor, was probably right ultimately to concede that two flats would have to be contiguous before they could fall within the expression. I do not believe that this weakens the appellant’s case, because the word “contained” in the expression can fairly be said to be capable of carrying with it the notion of the “premises” being a single piece of property which is greater than, and includes, the flat concerned.

There are, it is fair to say, arguably unattractive consequences of the appellant’s construction. If the expression is given the meaning for which the appellant contends, it would enable a landlord to deprive a tenant of his right under Chapter II of Part I of the 1993 Act (“Chapter II”), by establishing an intention to redevelop the tenant’s flat plus, say, an adjoining box-room. I think that it must follow from the appellant’s construction that a landlord could do this. Unattractive though this may appear to some people, it does not deter me from adopting the appellant’s construction.

The purpose of section 47, as I see it, is that Chapter II should not be invoked by a tenant of a flat to defeat a fairly imminent prospect of redevelopment, in the same way as section 30(1)(f) of the Landlord and Tenant Act 1954, from which the language of section 47(2)(b) is plainly lifted. Where the relevant flat is the sole subject of a projected redevelopment, there is no need for section 47: the tenant can redevelop the flat himself (subject to the terms of the lease permitting, but if the landlord wants the redevelopment, and it is inherent in section 47 that he does, then he can relax any covenant which would preclude it). However, once the proposed redevelopment extends to the flat together with other property, even if the other property is not very extensive, the tenant could not carry it out: only the landlord could do so, but he would need possession. The fact that this means that a landlord can sometimes formulate redevelopment proposals partly (or even solely) for the purpose of defeating the exercise of a right by a tenant under Chapter II is unattractive. However, it has always been a relatively well-known feature of section 30(1)(f) of the 1954 Act, and it has not been abrogated by the legislature despite the 1954 Act having been amended from time to time over the past 40 years.

A more unattractive possible result of the appellant’s construction which was canvassed in argument is that, if a flat demised to the tenant only extended to the inside of the plaster of the outer walls, premises which included only the flat and the outer walls of the flat would fall within the expression. Strictly speaking, it is unnecessary to decide whether this would be right, but, as one of my reasons for rejecting the respondent’s construction is that it is unclear and uncertain in its effect, the point ought to be addressed.

I am of the view that the point is not right, and does not follow from the appellant’s construction. If, as is common ground, the flat alone is not within the expression, then it appears to me that something more than its brick or stone envelope must be added before one is in the ambit of the expression. Tempting though it is, I do not think that this conclusion can be justified, as Mr Brock suggested, by recourse to de minimis. I consider, however, that the conclusion can be justified on the basis that the “premises in which the [tenant’s] flat is contained” must contain some additional space, and not just some wall, as well as the flat itself. In other words, for property to be within the expression, it must contain something which can fairly be called “premises” in addition to the “flat” itself.

(Another possible answer to the two conundrums discussed in the preceding paragraphs is that the common assumption before us that the expression cannot include the flat on its own is wrong. That assumption seems to be correct as a matter of ordinary English, but it does not appear to me to be a gross violation of language to hold that the flat on its own is included in the expression. Some support for such a reading might be said to be found in section 11(9), which is mentioned below. I doubt such a reading of the expression is correct, but, as it would involve trespassing into an area which was not argued, and going behind the agreed basis upon which the appeal and the case below were argued, it is inappropriate to say any more about it).

I find the respondent’s construction less satisfactory, both linguistically and in practical terms, than that of the appellant. I accept that such a construction could be possible if it was required by the context. However, it does seem to me to involve cutting down the naturally wide meaning of a statutory expression in the absence of any express words of limitation, in circumstances where there is no need to do so, and where there is no indication in the statute as to how it is to be done. Further, it also involves cutting down the meaning of the expression in a way which, in my opinion, could often lead to subjective, almost arbitrary, assessments, and therefore to uncertainty.

It does not strike me as a particularly natural or consistent use of language to conclude that two flats (excluding the common corridor, lift, and staircase) on a floor of a block with two flats per floor could be within the expression, when, in the same block, two flats, one above the other, could not. I also wonder whether, on the respondent’s case, two flats, one above the other, in a converted house with one flat per floor, would be within the expression. Further, I presume that the question whether two horizontally adjacent flats in a block with more than two flats per floor could be within the expression, as interpreted by the respondent, would depend on the lay-out, or even the construction, of the block.

In most cases at any rate, any part of a block of flats, which is greater than a flat and less than the block, and which the landlord wishes to redevelop, could be characterised as not being an existing recognisable unit. It could also be seen as being artificially assembled for the purposes of redevelopment. The concept of an existing recognisable unit would not merely be a judicial addition to, or implication into, the 1993 Act; it would also, as just explained, give rise to uncertainties in its application, being somewhat vague and rather subjective. Further, given that the premises within the expression are envisaged as being property which the landlord wishes to redevelop, I am not clear how the concept of artificiality assists.

I turn to see whether any assistance is to be found elsewhere in the 1993 Act. As already mentioned, the word “premises” is generally a word of very wide meaning, especially if it is preceded by the word “any”. Further, every time the word “premises” is used in the 1993 Act, it appears to me that the use is consistent with that wide meaning. Of course, in most cases, the wide meaning is cut down by express words of limitation, but that does not mean that the word itself does not have a wide meaning. On the contrary. The term “a man of property” only applies to a male with assets, but that does not alter the fact that the word “man” in that term encompasses, in its normal way, all males.

The judge thought that section 3 in Chapter I of Part I of the 1993 Act (“Chapter I”) cast doubt on the appellant’s case, particularly as it is referred to (via section 13) in section 41(4)(a) in Chapter II. I do not agree. It is true that sections 3 and 41 include similar expressions to that in sections 45(2) and 47(2), namely “flats contained in the premises…” and “any premises containing the flat” respectively. However, neither the provisions of section 3 nor those of section 41 suggest “premises” has a narrow meaning when used in the 1993 Act, even where the word is being used to apply to a unit of property containing a flat. On the contrary; those sections are quite consistent with “premises” being a word with a wide meaning, which is then narrowed by excluding any “premises” which do not satisfy all the three conditions in section 3(1) and that in section 3(2).

Particularly in those circumstances, it is hard to see any justification for presuming that the word was intended to be similarly confined in different sections of the 1993 Act, which contain no express narrowing conditions as in section 3, and no cross-reference (like section 41(4)(a), through the mention of section 13) to section 3. On the contrary: I consider that section 3 rather supports the appellant’s interpretation, in that it shows that, when the draftsman wished to narrow the meaning of the word “premises” for a particular statutory purpose, he did so by means of clear express words.

(It is arguable that the judge’s reliance on section 41(4)(a) was misconceived in any event. That is because it may well be that section 41(4)(a) would apply to a section 13 notice which sought to enfranchise property (including the flat in question) even if the property did not comply with the requirements of section 3: I think that such a notice would still be a section 13 notice for the purposes of section 41(4)(a).)

I see the force of the respondent’s point that one should not focus too much on the word “premises”, but on the expression as a whole, perhaps particularly the word “contained”. Even on that basis, it is difficult, at least in my opinion, to see how the word “premises” in the expression can legitimately and practically be limited in its ambit, save, perhaps, to the whole of the building in which the flat is situated. However, on that basis, quite apart from the fact that it is difficult to see the function of the word “any”, there is still the point that one is not only implying words into the expression, but implying words very similar in terms and concept to words which the draftsman has expressed in section 3.

That point is reinforced when one considers the definition of “flat” in section 101(1); it is defined as “a separate set of premises…which forms part of a building…”. Two points arise here. First, the draftsman had it in mind that a flat mentioned in the 1993 Act would be in a “building”; yet he did not require the redevelopment in section 47(2) to be of the whole or a substantial part of the building containing the flat. That seems to me to render it less appropriate (or more inappropriate) than it might otherwise have been to limit the “premises” in the expression to the building or a “natural” part of the building. Secondly, in section 101 the draftsman also had in mind the notion of a “set of premises” which he did not seek to introduce into the expression. Yet the respondent’s interpretation of “premises” in the expression gets close to invoking the notion of a “set of premises”.

The contention that the expression could apply to any piece of property provided that it includes the whole of the relevant flat also derives a little support, in my opinion, from section 11. This section is concerned with the right of a tenant of a flat to make enquiries of his landlord for the purposes of Chapter I. Section 11(1) entitles the tenant to be told of anyone who has an interest in “the relevant premises”, which are defined in section 11(9) as being “the building in which the flat is contained” or, where the ownership of the building is split, “that part of that building”. So, however the ownership of the building is split, and however “artificial” the part which includes the flat, that part is “relevant premises” “in which the flat is contained”.

The problem thrown up by this appeal must, of course, be resolved by reference to the particular wording of the expression in its statutory context. Accordingly, it may be unhelpful, even dangerous, to look at decisions on other statutory provisions. Nonetheless, I find it instructive to consider the reasoning of the House of Lords in Maunsell’s case, which was concerned with the meaning of “premises” in section 18(5) of the Rent Act 1968 (“the 1968 Act”). That section afforded in certain circumstances protection to a sub-tenant of a “dwellinghouse… form[ing] part of premises which have been let as a whole on a superior letting”. The House was divided on the issue of whether “premises” in that context applied to a farm with buildings (including the “dwellinghouse”, which in the Rent Act context, of course, can extend to a flat or even a room).

Lord Wilberforce, giving the main speech for the majority (who considered that a farm was not “premises” for this purpose), rejected at 389 the contention that the word “premises” should be “given a universal meaning” in the light of the legislative and case law history of section 18(5) of the 1968 Act. Lord Reid at 382 to 383 accepted that the word should be “given its ordinary meaning unless there is sufficient reason to give it in this particular case a secondary or limited meaning”; he then went on to hold, for similar reasons to those of Lord Wilberforce, that there was sufficient reason to give the word what he accepted was a “limited meaning” in that case. Viscount Dilhorne thought that “premises” naturally meant buildings rather than farmland with buildings (see at 384), a view with which Lord Wilberforce, at 386, had some sympathy.

In a powerful dissenting speech, written with Lord Diplock who agreed with it, Lord Simon of Glaisdale at 390 to 391 invoked the “golden rule” of statutory construction, namely that in technical legislation, a word is normally to be given its appropriately technical meaning (if it is has one), or (if it does not have one) its ordinary meaning. There was no good reason, in their view, for departing from that rule in the case before them, so that “premises” in their opinion meant the property comprised in the headlease, whether it was open land, buildings, parts of buildings or a combination.

It appears to me that the differing views in Maunsell’s case not only show the differences of judicial opinion that can be produced by the use of the word “premises” (which Lord Reid at 382 called a “difficult word”) in a statute. They also tend to support the conclusion I have reached in this case. The adoption of a naturally wide meaning of “premises” is (subject to one point) supported by the minority. It is also supported by Lord Reid (at least in the absence of good reason to depart from it). Given that Viscount Dilhorne and Lord Wilberforce thought it naturally applied to buildings, their reasoning and conclusions were, to put it at its lowest, not inconsistent with the appellant’s case here.

A point I ought to mention is that Lord Simon held that the technical meaning of “premises” was the property (of whatever nature) comprised in the headlease. That was because section 18(5) of the 1968 Act was in terms concerned with the whole of the property demised to the superior landlord. In this case, rightly in my view, neither party raised the possibility of the expression meaning the property comprised in the interest of the landlord (although it was canvassed briefly by Moore-Bick LJ in argument). The use of the word “any” and the absence of any reference to the property comprised in the landlord’s interest in sections 45 and 47 serve to demonstrate how this case differs from Maunsell’s case in this connection.

While it can be said with force that those distinctions assist the appellant, such differences serve to emphasise the dangers of construing a word in a statute in reliance on judicial decisions on the meaning of the same word in a different statute. Nonetheless, at least the general statements of principle and the opinions as to prima facie meaning of “premises” in the speeches of Lord Reid and of Lord Simon (with which Lord Diplock agreed) are, to my mind, of some, albeit limited, help in this difficult case.

In the end, canons of construction and judicial decisions (even of the House of Lords) on the meaning of similar words in other statutes cannot conclusively determine the outcome of this appeal. We have to decide the meaning to be ascribed to the composite expression “any premises in which the [tenant’s] flat is contained” in sections 45 and 47, in the light of its linguistic and purposive context. For the reasons I have given, which are substantially the same as those of Moore-Bick LJ, with whose judgment I entirely agree, it seems to me that the appellant’s construction is to be preferred to that of the respondent or to that adopted by the judge.

In these circumstances, for my part, I would allow the appeal.

I gratefully adopt, without repeating, Moore-Bick LJ’s account of the facts and circumstances of this appeal. I agree with much of his analysis and that of Neuberger LJ in his judgment. I regret, however, that I do not agree with all of it, in particular with regard to the facts of the present case.

We are concerned with “an intention to redevelop any premises in which the tenant’s flat is contained” – see section 47(1) of the 1993 Act. We are therefore concerned with the concept of premises which contain a flat in the context of an intention to redevelop those premises. The three elements which contribute to this concept all reappear in section 47(2). The enquiry has to look at the flat and premises which might contain it as they are at the time of the landlord’s application.

The word “premises” is used in a number of different contexts in this statute, and no help is to be found from other places for its meaning and extent in section 47. I agree with Moore-Bick LJ for the reasons which he gives that the judge was wrong to derive help from Chapter 1 of the 1993 Act. I also note that the expression in section 47 is “any premises”.

I agree that, depending on the facts, “any premises in which the tenant’s flat is contained” may apply to premises consisting of less than a whole building or block. I am prepared to suppose that the full 7^th floor of this block of flats would qualify.

However, where do you stop? The appellant landlords’ case in effect means that “premises in which the tenant’s flat is contained” may comprise the flat together with any premises contiguous with it around which the landlord chooses to delineate a three dimensional parcel. Thus the flat plus an adjacent box-room, or, if the demise of the flat includes the plaster but not the underlying walls, the flat plus the walls would each comprise premises in which the tenant’s flat is contained. I do not think so. In the context of redevelopment, these artificial units are scarcely premises, but they are clearly, in my view, not premises in which the flat is contained. The premises have to contain the flat and have to be an existing recognisable unit which may sensibly be said to do that, not some artificial unit, put together simply to achieve the statutory requirement, which would not otherwise be recognised as an existing containing unit. I do not regard this as reading words into the statute. It is rather construing words which express the concept of premises which may properly be said to contain a flat in the context of an intention to redevelop those premises. The statutory expression is not “any premises in which the tenant’s flat is included”.

I understand that each floor of this block of flats has seven or eight flats. The 7^th floor would, I think, be premises in which Flat 77 is contained. Whether part of the 7^th floor, including Flat 77, would also be premises in which Flat 77 is contained would be debateable. But, in my judgment, it overstrains the statutory language to say that Flat 77 on floor 7 and Flat 74 on floor 6 are together and alone premises in which Flat 77 is contained. Together they are not a recognisable existing unit which contains Flat 77. This was Mr Denehan’s essential submission, which I would accept.

I accept that my construction of section 47 of the 1993 Act leaves rough edges at the factual margins. Two adjacent flats on a floor comprising 8 flats with no identifiable physical separation such as a lift lobby would be an example. I would be inclined to say that two such flats did not come within section 47 because they were not, in the context of redevelopment, together a recognisable existing unit. But imprecise statutory language will often result in problems of this kind; and equivalent problems arise, I think, with the appellant’s construction, as is illustrated by the difficulties identified, and to my mind not satisfactorily resolved, in paragraphs 22 to 25 of Moore-Bick LJ’s judgment and paragraphs 35 to 42 of Neuberger LJ’s judgment.

For these reasons, I would dismiss this appeal.