England and Wales High Court (Chancery Division) Decisions

You are here: BAILII >> Databases >> England and Wales High Court (Chancery Division) Decisions >> PW & Co v Milton Gate Investments Ltd [2003] EWHC 1994 (Ch) (08 August 2003)
URL: http://www.bailii.org/ew/cases/EWHC/Ch/2003/1994.html
Cite as: [2004] Ch 142, [2004] 3 EGLR 103, [2004] L & TR 8, [2003] EWHC 1994 (Ch), [2004] 2 WLR 443

[New search] [Printable RTF version] [Buy ICLR report: [2004] Ch 142] [Buy ICLR report: [2004] 2 WLR 443] [Help]

		Neutral Citation Number: [2003] EWHC 1994 (Ch)

IN THE HIGH COURT OF JUSTICE
CHANCERY DIVISION

Royal Courts of Justice
Strand, London
WC2A 2LL

8th August 2003

B e f o r e :

Mr Justice Neuberger
____________________

Between:

	PW & Co
	V
	Milton Gate Investments Ltd

____________________

HTML VERSION OF JUDGMENT
____________________

MR JUSTICE NEUBERGER:

Introduction

This case involves two connected disputes, which raise a number of points of difficulty. The first dispute is between PW & Co. ("PW), in its capacity as former tenant of a substantial office building, and Milton Gate Investments Limited ("Milton Gate") its former landlord. The second dispute is between Milton Gate and PW's former subtenants of parts of the building, who are two companies in the BT plc group, under seven underleases ("the Underleases").

The disputes have their origin in the fact that, when the lease of the building ("the Headlease") was granted to PW, and when PW granted the Underleases, the parties involved, and their respective advisers, were all under an arguable misconception as to the effect on a subtenancy of the exercise by the head-tenant of a right to determine the head-tenancy. Because it is so fundamental to an understanding of the issues in this case, it is appropriate to explain the point, albeit in fairly summary terms, before turning to the facts, and the issues between the parties.

In Pennell -v- Payne [1995] QB 192, the Court of Appeal had to determine what Simon Brown LJ (who gave the only reasoned judgment) described at 195C as "a pure question of law", which he then summarised in these terms:

"[W]here a head-tenant serves an upwards notice to quite, i.e. a notice upon his landlord, is the landlord thereupon entitled to possession against a subtenant irrespective of whether that subtenancy was granted within the terms of the Headlease?"

Two points should be made about that formulation. First, it is implicit that, in the absence of the notice, the subtenancy in question would, as between head-tenant and subtenant, have otherwise continued. Secondly, even if the effect of the notice described by Simon Brown LJ would be to determine the subtenancy in common law, it may be that the subtenant would be entitled to remain in possession as against the landlord, because statute entitled him to do so.

The only case where the question as formulated by Simon Brown LJ had previously been decided was in Brown -v- Wilson (1949) 208 LTJo. 144. In light of observations in earlier cases, Hilbery J concluded that, in the case envisaged by Simon Brown LJ, the subtenant was entitled to remain in possession: in other words, that the notice by the head-tenant determining the head-tenancy did not put an end to the subtenancy.

In Pennell, the Court of Appeal unanimously took the opposite view, and overruled the decision in Brown -v- Wilson. The decision in Pennell was considered by the House of Lords in Barrett -v- Morgan [2000] 2 AC 264, where Lord Millett (who gave the only reasoned speech) said that he had "no doubt that this case was correctly decided" (see at 274A).

The decision in Brown -v- Wilson had stood for some 45 years, and (as was pointed out by Simon Brown LJ at [1995] QB 201A-B) it had been assumed to represent the law, in various textbooks (some of which expressed doubts as to its correctness). Accordingly, it is perhaps not very surprising that those involved with the drafting of the Headlease and Underleases in the present case were under an arguable misapprehension as to the law.

There is one other point of law which should be mentioned at this stage. The Headlease and Underleases were, at least potentially, within the ambit of Part III of the Landlord and Tenant Act 1954 ("the 1954 Act"), which affords protection to tenants in occupation of business premises (unless their tenancies have been lawfully excluded from such protection by a court order). Accordingly, even if the effect of a notice served by a head-tenant to determine the head-tenancy puts an end to a subtenancy as a matter of extra-statutory law, a subtenant who has the benefit of the 1954 Act would nonetheless be entitled to remain in occupation of the premises sublet to him because his tenancy would continue pursuant to the provisions of that Act.

The apparent misconception on the part of those involved in the conveyancing documentation in the present case has led to a wide ranging series of well presented arguments, involving fundamental questions of property law, including the impact of the Human Rights Act 1998 ("the 1998 Act"), questions relating to the permissible ambit of estoppel by deed and estoppel by convention, and the circumstances in which such an estoppel can be avoided, generally and in a tripartite situation.

With those introductory observations, I turn to the relevant facts.

The facts

I shall begin by setting out the relevant provision of the Headlease, of the Underleases, and of the licences pursuant to which they were granted. I shall then summarise the evidence relating to the negotiation of the grant of the Headlease and the Underleases (which can be dealt with pretty shortly). Finally, I shall describe the relevant events which took place from the time that "Milton Gate" was proposing to acquire the freehold reversion to the Headlease (which must be explained in a little more detail).

The Headlease

Until late 1994, the freehold of "Milton Gate", 1 Moor Lane, London EC2 ("the premises"), was owned by City of London Real Property Company Limited ("CLRP"), a member of the Land Securities plc Group. By the Headlease, dated 29th November 1990, CLRP let the premises to PW the chartered accountants, then known as Price Waterhouse. The Headlease was for a term of 25 years from 24th June 1990 at an annual rent (payable in advance on the usual quarter days) of £6,616,000 (subject to upwards review every five years).

The centrally relevant provision of the Headlease was clause 5(6), which was in the following terms:

"The Lessee [that is, PW] may determine this Lease on the 24th day of June 2002 by giving to the Lessors [that is CLRP] not less than twelve months prior written notice and on the expiration of such notice this present Lease shall determine subject to any Permitted Underleases but without prejudice to any claim by either party against the other in respect of any antecedent breach of any covenant... and in the event of the Lessee serving such notice the Lessee shall upon the 24th day of June 2002 pay to the Lessors a sum equivalent to nine months yearly rent payable at that date Provided That such sum will not be payable if on the expiration of such notice 75% of the lettable office area of the ... premises is underlet in accordance with the terms of this Lease and such Permitted Underlease or Underleases have an unexpired term of at least 5 years (emphasis added).

I shall refer to the sum equivalent to "nine months yearly rent payable" as "the Penalty".

Clause 3(25) of the Headlease contained detailed provisions which limited and controlled the circumstances in which, and the basis upon which, PW could underlet. By virtue of clause 3(25)(2)(e)(i), a "Permitted Underlease" was a subtenancy which was required to:

"Contain on the part of the Underlessees (inter alia) the Stipulated Covenants and provisions for rent reviews on the same dates and in the manner (mutatis mutandis) specified [herein] and [to] reserve no less than the open market rent then reasonably obtainable for the subject premises..."

Clause 3(25)(2)(e)(ii) defined a "Permitted Underlessee" as a person "who shall have previously to the execution of the Permitted Underlease entered into the Stipulated Covenants directly with the Lessors".

The "Stipulated Covenants" referred to in clauses 3(25)(2)(e)(i) and (ii) were defined in clause 3(25)(1)(iii) as including obligations to:

"Observe and form the covenants and conditions on the Lessee's part herein contained so far as they relate to or affect the subject premises (save... the covenant to pay the rents hereby reserved ... ) ..."

Clause 3(25) also required PW not to underlet the whole or any part of the premises save by a Permitted Underlease to a Permitted Underlessee, and not to effect any such underletting without first obtaining the consent of the Lessors, which consent was not to be unreasonably withheld.

The Headlease was at all times vested in PW. Towards the end of 1994, CLRP transferred the freehold reversion to the Headlease to three Swedish companies, to which I will refer as "Allmanna". On 19th June 2000, the freehold reversion was transferred by Allmanna to Milton Gate, who retain the freehold of the premises to this day.

The Underleases

PW granted a total of seven Underleases, each of which was in respect of a single floor of the premises. Between them, they involved subletting effectively the whole of the premises. Four of those Underleases were entered into on 3rd March 1994, and they extended to the ground, first, second, and third floors of the premises. Each of these four Underleases was granted to BT, for a term expiring on 28th September 2007. A fifth Underlease was granted by PW to BT on 28th April 1995; it related to the fourth floor, and was for a term expiring on 20th June 2015. I shall refer to these five Underleases as "the BT Underleases" and the parts of the premises thereby demised as "the lower five floors".

Each of the Underleases granted to BT on 3rd March 1994 had a clause, which conferred on BT the right to determine, by giving at least 12 months notice, on 29th September 1998 or 17th June 2002, but, in the event of it exercising such right, BT had to pay a penalty to PW. In the Underlease which I have seen (that of the third floor), the penalty was to be twice the annual rent in the event of determination on 29th September 1998, and the annual rent in the event of determination on 17th June 2002.

Two further Underleases ("the MLEL Underleases") for terms expiring on 20th June 2015, were granted by PW on 10th January 1995; they extended to the fifth and sixth floors ("the upper two floors"). These two Underleases were granted to Merrill Lynch (Europe) Limited ("MLEL"), who assigned them to BT on 22nd March 2002.

In each of these seven Underleases, PW were "the Lessors", which expression was defined to include "where the context admits... the estate owner for the time being entitled to the reversion immediately expectant on the determination of the term hereof '.

All the seven Underleases reserved rents which were subject to upwards only reviews. Because the Underleases were each in respect of part only of the premises, they contained service charge provisions, which included a sinking fund (i.e. a fund into which the lessor could require the lessee to make periodical payments which in due course could be used for defraying the cost of substantial services). However, it was only open to the Lessors under the Underleases to implement the sinking fund from 18th June 2002.

In the normal way, each of the Underleases was granted pursuant to a written licence executed by the freeholder (CLRP or Allmanna), the head-tenant (M), and the proposed subtenant (BT or MLEL). With one or two variations (not material for present purposes) the licences to underlet ("the Licences"):

i) provided that it was "accepted by the parties that such sub-lettings will create Permitted Underleases as such expression is defined by the Lease";

ii) stated that the Licence was "supplemental to" the Headlease;

iii) contained, in clause 11, a covenant by the proposed subtenant (i.e. BT or MLEL) direct with "the Landlord" (i.e. CLRP or Allmanna) substantially in these terms:

"that the subtenant will as from the date of the subtenancy and for so long as the subtenant (here meaning [BT or MLEL] only) is liable to the lessor thereunder [i.e. PW] perform and observe the covenants (other than as to payments of rent) on the part of the lessee contained in the [Head]lease in so far as the same are not inconsistent with the lessee's covenants to be contained in the subtenancy hereby permitted..."

The parties' beliefs at the time of the negotiations

The negotiations for the grant of the Headlease between the respective advisers to CLRP and PW were under way by early 1990, and the discussions relating to PW's right to break, which eventually became clause 5(6) of the Headlease ("Clause 5(6)"), took place between April and June 1990. I believe that it is unnecessary to describe the contents of the negotiations, because they throw no more light on the understanding of each of the parties than one can gather from the terms of the Headlease.

In addition, there is evidence, in the form of witness statements, as to the state of mind of the individuals at CLRP and PW and their respective solicitors, who were responsible for negotiating the terms of the Headlease. The members of both negotiating teams, and their respective clients, believed that, in the event of PW exercising its right to break under clause 5(6), any subtenancy granted by PW, at least provided that it was a Permitted Underlease, would become enforceable and binding as between the subtenant and the freeholder.

The correspondence and witness statements disclosed in relation to the grant of the BT Underleases, and the associated Licences, reveal a very similar state of mind on the part of each of the parties involved. It is fair to say that there is nothing in the correspondence leading up to the grant of the BT Underleases which unequivocally indicates that BT believed that their Underleases would survive the exercise by PW of its right to determine the Headlease. The negotiations which led to the inclusion in each of the BT Underleases of the right of the lessor thereunder to introduce a sinking fund for service charges only after 18th June 2002, arose from BT's desire to avoid a sinking fund, PW's comparative indifference on the issue, and CLRP's desire to have a sinking fund. At first sight, this aspect of the negotiations therefore suggests that BT communicated, albeit by implication, the fact that it shared the assumption that its Underleases would survive the determination of the Headlease pursuant to clause 5(6), with effect from 24th June 2002. However, the agreement that the sinking fund provision only applied from 18th June 2002, could be justified on the basis that, given that the BT Underleases were likely to be within the 1954 Act, BT would become the direct tenants of the head-landlord, even if their Underleases otherwise determined pursuant to the general and extra-statutory law. Indeed, there was reference, in the correspondence concerning the sinking fund, to the Underleases continuing under the 1954 Act.

The negotiations for the MLEL Underleases involved not only a different proposed subtenant, MLEL, but also a different freeholder, because Allmanna had by then acquired the freehold reversion to the Headlease. The correspondence leading up to the MLEL Underleases reveals a virtually identical position to that indicated by the correspondence leading up to the grant of the BT Underleases.

As in the case of the grant of the Headlease, there is evidence as to the state of mind of the parties involved in the grant of the BT Underleases and the associated Licences. They all believed that the BT Underleases would survive the exercise by PW of its right to break the Headlease. While I do not have evidence from MLEL in this connection, I think it is a fair assumption that they and their advisers had the same view at the time of the negotiations for the grant of the MLEL Underleases.

Events after the grant of the Underleases

By March 2000, Milton Gate was considering purchasing the freehold of the premises. An internal memorandum produced by its solicitors, dated 10 April 2000, indicates that Milton Gate were told, and believed, that, in the event of PW exercising its right under clause 5(6), the Underleases would determine on 20 June 2002, subject to any rights which BT and MLEL or their successors ("the under-tenants") might have pursuant to the 1954 Act. The correspondence shows that, in that event, Milton Gate envisaged serving notices on the under-tenants "to renegotiate the terms" of the Underleases. Given Milton Gate's subsequent statements, it seems clear that, from April 2000 onwards, it and its advisers were of the view that this indeed represented the position.

As already mentioned, Milton Gate acquired the freehold reversion from Allmanna on 19th June 2000.

On 6th October 2000, PW served a notice (a " Notice") exercising its right to determine the Headlease pursuant to clause 5(6). The effect of this Notice was, it is accepted on all sides, to put an end to the Headlease on 24th June 2002. At the time of service of the Notice, PW believed that the Underleases would continue beyond 24th June 2002, notwithstanding the determination of the Headlease (unless BT exercised their respective rights to determine the Underleases).

In his witness statement, Mr Roderick Menzies, PW's Estates Manager, said that, from 1998, PW "always knew that [it] would exercise the break clause as [PW] had no occupational requirement for [the premises]. He said that he started considering "the ramifications of exercising the break and.... the terms of the [Head]leases in more detail" in early 2000. As a result, he "understood that if [PW exercised its right under clause 5(6)], the sub-tenancies would continue" (unless the subtenants exercised their right to break) because of the terms of clause 5(6). It appears that this understanding was based on advice, because he went on to say in his evidence that the effect of certain reinstatement provisions in the Headlease was "also explained to me".

Mr Menzies also stated in his evidence that PW's rent under the Headlease, which had been increased to about £6.95m per annum on review in 1995, was kept the same on the 2000 review by an agreement reached in March 2001. In March 2001, the rent under the BT Underleases was reviewed to a level such that the aggregate rents under all seven Underleases was about £6.735m per annum.

At a meeting on 15th November 2000, following a conversation between solicitors a week or so earlier, "Milton Gate" told PW that, in light of the decision in Pennell, the effect of the Notice would be to determine not merely the Headlease, but also the Underleases. PW was taken by surprise by this contention and took issue with it. Milton Gate's position was subsequently communicated to BT at a meeting, and was discussed between their respective solicitors, shortly afterwards on 25th April 2001.

During 2001, there were three sets of independent negotiations. First, PW was suggesting to Milton Gate that the Headlease should be surrendered (on the basis that this would ensure that the Underleases would not determine on 24th June 2002). Secondly, Milton Gate and BT were discussing the possibility of BT taking a new 15-year overriding lease of the whole of the premises. Thirdly, there were negotiations between BT (who remained in occupation of the lower five floors) and MLEL, with a view to BT taking an assignment of the two MLEL Underleases.

The third set of negotiations proved fruitful and in January 2002, BT agreed to acquire the two MLEL Underleases, which were duly assigned to BT in March 2002. The second set of negotiations, between Milton Gate and BT for the grant of an overriding lease, continued more actively into 2002. During the negotiations with BT, "Milton Gate" was maintaining its position that the Underleases would all expire on 24th June 2002, and BT were contending that they would not. The overriding lease which was being negotiated would have been for a term expiring in 2017, at an initial rent of over £7.21m per annum, significantly above both the then passing rent and the then market rental value of the premises. BT were anxious to remain in the premises, and "Milton Gate" was putting pressure on them by threatening to serve hostile notices under section 25 of the 1954 Act, relying on an intention to redevelop, which could have led to BT having to vacate, probably some time in 2003. These threats caused BT "serious concern" as they wished to stay in the premises. If the premises had been let on the overriding lease at an annual rent of more than £200,000 above the passing rent under the Headlease and for an unbreakable term running to 2017 it would have considerably increased the value of Milton Gate's freehold reversion, by millions of pounds. Indeed, an increase of £10m or even £15m has been mentioned in the evidence.

In April 2002, BT's wish to remain in the premises, as evidenced by their negotiations with Milton Gate for the grant of an overriding 15-year lease, and by their acquisition of the two MLEL Underleases, changed. BT decided that the premises were surplus to their requirements, and that they wished to vacate. Accordingly, on or about 16th April 2002, BT ended the negotiations for the overriding lease, and informed Milton Gate of their intention to vacate the premises. During April and May, Milton Gate and BT had discussions, which led to an agreement in principle, in early June 2002, that BT would vacate by March 2003.

As so often happens, one party's change of tack was followed by a corresponding change on the part of the other. The rent due under the BT Underleases on 24th June 2002 was paid by BT, because, even if those Underleases had determined on 24th June 2002, they were continuing under the 1954 Act as BT were in occupation of the lower five floors. However, that did not apply to the upper two floors, because no one was in occupation: MLEL had moved out, and BT had not moved in. Accordingly, BT did not pay any rent under the MLEL Underleases on 24th June 2002 on the basis that they had expired on that date.

On 4th July 2002, Milton Gate's solicitors wrote to BT's solicitors threatening to wind up BT if they did not pay the rent due under the MLEL Underleases on 24th June 2002. This was explained in a letter from Milton Gate to BT's agents written on the following day. The letter stated that:

"[F]ollowing a further review, the Pennell -v- -Payne issue is becoming more difficult for us to pursue and Milton Gate... has decided not to take the issue any further. Accordingly... I can confirm that the [under] leases... including those relating to the fifth and sixth floors, continue in full force and effect notwithstanding the service of a break notice by PWC."

BT replied through their solicitors, pointing out that Milton Gate had taken precisely the opposite position as against BT for over a year, and that it was too late for Milton Gate now to contend that Pennell did not apply. Thereafter, perhaps not surprisingly, BT's solicitors wrote to Milton Gate's solicitors asking why "the Pennell -v- Payne issue is becoming more difficult_ to pursue", and, perhaps equally unsurprisingly, Milton Gate did not reply. BT thereafter paid the rent for 24th June 2002 in respect of the upper two floors "under protest".

Meanwhile, the first set of negotiations between PW and Milton Gate had been proceeding fitfully. PW was pressing a surrender of the Headlease on Milton Gate, who was being generally unreceptive to the notion. On 24th April 2002, Milton Gate's solicitors wrote to PW's solicitors making it clear that Milton Gate expected PW to pay the Penalty, which amounted to just over £5.17m, on 24th June 2002. This was justified by Milton Gate on the basis that the proviso to clause 5(6) could not be satisfied because the effect of Pennell was to determine the Underleases on 24th June 2002. PW's solicitors replied on 23rd May 2002, challenging this contention, and renewing PW's offer to surrender the Headlease to Milton Gate. Milton Gate did not reply to this offer and, to protect its position, PW issued the instant proceedings on 11th June 2002, seeking a declaration that the Penalty would not be payable.

In its Defence and Counterclaim dated 23rd July 2002 (and which has subsequently been substantially amended), Milton Gate alleged against PW that the Underleases had determined on 24th June 2002, that the proviso to clause 5(6) accordingly did not apply, and that PW was therefore liable to pay the Penalty, with interest. On 9th October 2002, BT was added as a Part 20 Defendant to the current proceedings, on the application of Milton Gate.

Meanwhile, as BT's case was that their tenancies of the lower five floors were continuing pursuant to the 1954 Act, they served notices under section 26 of that Act on Milton Gate for the grant of new tenancies of those floors of the premises.

Summary of the issues

BT's case is that the conclusion and reasoning in Pennell. as discussed and approved in Barrett, means that all the seven Underleases determined on 24th June 2002 as a result of the service of the Notice. As a consequence, because BT was in occupation of the lower five floors as at 24th June 2002, the BT Underleases were continued, but only through the operation of the 1954 Act, and as no one was in occupation of the upper two floors, the MLEL Underleases expired on that date for all purposes.

This analysis is challenged by PW and (in what is now its primary case) by Milton Gate. They contend that clause 5(6) was so expressed as to make it clear that, notwithstanding the fact that the Headlease came to an end as a result of the service of the Notice, the Underleases nonetheless continued. In effect, they say that by clause 5(6), the Landlord, originally CLRP, and the head-tenant, PW, agreed not only that PW could serve the Notice to determine the Headlease on 24th June 2002, but also that any subtenancy, provided it was a Permitted Underlease, would continue according to its terms, and would not determine, despite the normal rule. established in Pennell, which I shall call "the general rule".

In these circumstances, the first issue to be considered is whether, and if so in what circumstances, a landlord and a tenant can validly contract out of the normal consequence on a subtenancy of the service of a notice under a provision such as Clause 5(6) (a "break notice") under a head-tenancy, and whether there has been such a valid contract in this case.

If BT lose on this issue, then that is the end of this case (subject to one point about enforceability of the covenants in the Underleases). The Underleases, all of which are now vested in BT, will continue until their respective contractual term dates, and the proviso to clause 5(6) is satisfied, so that PW is not obliged to pay the Penalty for exercising its right under clause 5(6).

If CLRP and PW did not contract out of the general rule (either because they did not purport to do so or because they could not do so), then PW and Milton Gate contend that clause 5(6) should be construed so that a Notice served thereunder has the effect of surrendering the Headlease, rather than determining it by effluxion of time. BT's reply is that this is impermissible, and that, in any event, it does not assist Milton Gate and PW in their case.

If BT succeed on this issue, then Milton Gate and PW argue that the same result for which they contend as a matter of contract, is nonetheless effectively justified on the basis of estoppel. This estoppel argument is put in two ways. First, they rely on estoppel by deed. Secondly, they rely on estoppel by convention. Each estoppel is said to be between the landlord under the Headlease ("the Landlord") and PW as tenant thereunder. BT contend that the facts and documents in this case do not satisfy the requirements of either type of estoppel, and, anyway, that the estoppel never bound them.

If Milton Gate and PW succeed on one or both of the estoppel points, BT contend that Milton Gate, whether or not it is bound by the estoppel as against PW, cannot insist on the estoppel as against BT. If BT succeed in defeating Milton Gate on estoppel, PW argues that Milton Gate would be estopped from recovering the Penalty from PW.

Having briefly explained the disputes between the parties, the issues which require to be determined may be conveniently summarised and considered in the following order:

i) Contracting out of the general rule that a subtenancy is determined by determination of the head-tenancy.

a) Did CLRP and PW purport to contract out?

b) Can a landlord and a tenant contract out of the general rule?

c) Is there a problem due to lack of reversion?

d) Does the 1998 Act affect this conclusion?

ii) Characterising the effect of the Notice as a surrender.

a) Can this be done?

b) Does it help "Milton Gate" and PW?

iii) Estoppel by deed.

iv) Estoppel by convention: the existence of a convention.

a) Was there a convention as a result of the Headlease?

b) Was there a convention as a result of the Underleases and Licences?

c) Were BT bound by any such convention?

d) Was Milton Gate entitled to rely on any such convention?

v) Estoppel by convention: the existence of an estoppel.

a) An estoppel between Milton Gate and BT.

b) An estoppel between Milton Gate and PW.

c) How can the tripartite relationship be reconciled?

Contracting out of the rule in Pennell -v- Payne

Was there a purported contracting out?

It is common ground that, in the absence of the words which I have emphasised when quoting the clause (namely "subject to any Permitted Underleases" and "such Permitted Underlease or Underleases have an expired term of at least 5 years"), the effect of service of a Notice under clause 5(6) would be to determine the Underleases. That must follow as a result of the decisions in Pennell and Barrett. Milton Gate and PW contend that the effect of the words I have emphasised is that there was an agreement between CLRP and PW to the effect that the Underleases would survive notwithstanding the determination of the Headlease. The first question it is convenient to consider is whether the parties did in fact purport to contract out of the consequences of the general rule laid down in Pennell.

On behalf of BT, Mr Nicholas Dowding QC, who appears with Mr Jonathan Evans, contends that the wording of clause 5(6) did not give rise to any agreement between CLRP and PW to the effect that any Permitted Underlease would continue after 24th June 2002, notwithstanding the service of a Notice, and notwithstanding the general rule to the contrary. In particular, he argues that the words "subject to any Permitted Underleases" are the only words upon which such contention could rest, and that those words are insufficiently clear and unequivocal to justify the conclusion for which Milton Gate and PW argue.

On a fair reading of clause 5(6), it appears to me that the parties agreed that, on service of a Notice, the Headlease would determine on 24th June 2002, and, equally, that they agreed that such determination would be "subject to any Permitted Underleases". At least in the absence of the general rule, as decided in Pennell, the natural meaning of those words is that any Permitted Underlease would be unaffected by, and continue notwithstanding, determination of the Headlease. To put it at its lowest, that is at least a natural meaning of the expression "subject to any Permitted Underleases". Indeed, in my view, it is the natural meaning. If the exercise of a right is described as "subject to" another right or interest, then the normal meaning of the words is that the other right or interest should not be detrimentally affected by the exercise of the first right. Further, it is not very easy to envisage what other purpose the words in question were intended to have. Mr Dowding's suggestion that the words were included to protect PW against the possibility of any subtenant remaining in occupation, does not strike me as convincing. It does not give the words a natural meaning. In so far as the subtenant was entitled to remain in occupation by virtue of the 1954 Act, his presence would not represent a breach of PW's obligation to deliver up possession. If he was not entitled to remain in occupation, that would be because his subtenancy had come to an end, in which case the words "subject to any Permitted Underleases" would not protect PW: the former subtenant would be a trespasser.

At first sight, it could be said that, if the parties had intended to produce a different result from that normally required by law as laid down in Pennell, CLRP and PW would have made their intention clearer than they did. There are two reasons for rejecting that argument. The first is that the Headlease was entered into at a time when the decision in Brown -v- Wilson was assumed to represent the law; consequently the parties could well have expressed their agreement fairly shortly, believing that it was consistent with the general law. Secondly, it appears to me that the closing words of clause 5(6) make it abundantly clear that the parties did believe that any subtenancy would survive the expiry of the head-tenancy by a notice under clause 5(6). Accordingly, to give the words "subject to any Permitted Underleases" the meaning for which PW and "Milton Gate" contend, accords entirely with the understanding as expressed at the end of the clause. I am by no means convinced that I would have reached a different conclusion, even in the absence of these two arguments, but they render the likelihood of the parties having agreed that Permitted Underleases would survive the expiry of the Headlease pursuant to a Notice, relatively easy to accept.

Mr Dowding also relies on the fact that, in the first draft of clause 5(6), before the Penalty provision or the proviso were added, the words "subject to any Permitted Underlease" were included. I do not consider that this assists BT. First, one cannot invoke an earlier draft as an aid to the construction of an agreement (whether a Deed or otherwise). Secondly, given that the solicitors like the parties, believed that the subtenancies would or could survive the expiry of the Headlease, this point would take matters no further in any event.

In these circumstances, I conclude that CLRP and PW purported to contract out of the consequences of Pennell, at least in relation to Permitted Underleases, in the event of the Headlease being determined by a Notice pursuant to clause 5(6). The next question is whether that was an effective agreement.

Was the agreement effective as a matter of law?

The question which arises in this connection is, as in Pennell, a pure question of law. It is whether a landlord and head-tenant can effectively agree that, notwithstanding the head-tenancy's determination by a break notice, a subtenant will nonetheless be entitled to retain possession as against the landlord, for the contractual duration of the subtenancy. Mr Kirk Reynolds QC, who appears with Mr Nicholas Taggart for PW, and Mr David Hodge QC, who appears with Miss Nicole Sandells, for Milton Gate, contend that it is open to a landlord and a head-tenant so to contract. They argue that the point was not disposed of against them by the decisions in Pennell and Barrett that it is therefore an open point, and that it should be answered in the affirmative, both in light of the general thrust of the reasoning in those two cases, and because of the general principle that parties should be free to contract as they wish, save in so far as public policy or statute indicate otherwise. In that connection, Mr Reynolds relies on a dictum of Sir George Jessel MR in Printing and Numerical Registry Co. -v-Samson (1875) LR 19 Eq 462 at 465.

This contention is based on the proposition that the general rule as laid down in Pennell is, to quote Mr Reynolds, "not one of deep legal principle" or, to quote Mr Hodge, a "general default position based on clear policy arguments" which does not apply "regardless of the bargain between the parties".

The only authoritative material available to test those propositions is the reasoning in the judgment of Simon Brown LJ in Pennell, and in the speech of Lord Millett in Barrett. Because that material is not only authoritative, but also potentially directly on point and recent, it justifies close scrutiny. On the other hand, because the instant problem does not appear to have featured in the arguments, or in the minds of the tribunals, in those cases, it could be a little dangerous to place too much weight on the precise words used during the course of the judgment or speech, as the case may be.

On the particular issue which falls to be decided here, the judgment of Simon Brown LJ in Pennell does not seem to me to provide much assistance. Mr Dowding, on behalf of BT, points out that the issue was described as "one of general principle" at 195D. Mr Reynolds and Mr Hodge rely on the fact that, in order to decide the issue, it was considered "necessary to examine the policy considerations in play, and, above all, the necessary consequences of a decision either way" - see at 201C. As they say, Simon Brown LJ then went on to consider the practical consequences of either party being right, and concluded that, if the normal rule was that the subtenancies survived the service of an upwards notice by the head-tenant on the landlord, it could cause such injustice to the landlord that the contrary conclusion was "compelling" (see at 202H).

The fact that the issue was described as "one of general principle and application" does not seem to me necessarily to assist BT in the present case. A "general principle" can always have exceptions, and there are no clear grounds for thinking that Simon Brown LJ might not have accepted that a prior agreement between a landlord and head-tenant might not have amounted to such an exception. Equally, the fact that Simon Brown LJ relied on policy as justifying his decision does not mean that, if the policy considerations do not apply in a particular case, a different result would obtain. Many, indeed, I suspect the great majority, of legal principles and rules (whether of equitable, common law, or statutory origin) will have been based on what could be called policy considerations. Yet that does not result in the rule or principle not applying simply because, on the particular facts of a case, the policy considerations do not apply.

The reasoning of Lord Millett in Barrett appears to me to provide significantly more assistance on this issue. However, because he did not have in mind the facts of a case such as this, it is almost inevitable that each side can take comfort from at least parts of what he said.

At 270D to E, Lord Millett said this:

"A lease or tenancy for a fixed term comes to an end by effluxion of time on the date fixed for its determination. A periodic tenancy comes to an end on the expiry of a notice to quit served by the landlord on the tenant or by the tenant on the landlord. ... [I]t also comes to an end by effluxion of time. In each case the tenancy is determined in accordance with its terms. By granting and accepting a periodic tenancy with provision, express or implied, for its determination by notice to quit, the parties have agreed at the outset on the manner its of termination."

By contrast, said Lord Millett at 270F to 271A:

"A surrender is ineffective unless the landlord consents to accept it, and is therefore consensual in the fullest sense of the term... On its surrender the tenancy is brought to an end prematurely at a time and in a manner not provided for by the terms of the tenancy agreement. . In this respect it differs from the case where a tenancy is determined by notice to quit. It is because the landlord or his predecessor in title has not, by granting the tenancy, previously agreed that the tenant should have the right to surrender the tenancy prematurely that the landlord's consent is necessary."

As I have mentioned, Lord Millett had "no doubt" that Pennell was rightly decided (see at 274A), although he considered that a different result obtained where the head-tenancy was determined by surrender. At 271 D to E he stated:

"It is a general and salutary principle of law that a person cannot be adversely affected by an agreement or arrangement to which he is not a party. So far as he is concerned, it is res inter alios acta. It would conflict with this principle if the destruction of a tenancy by surrender carried with it the destruction of the interest of a subtenant under a subtenancy previously granted."

At 272 A to F, Lord Millett set out four "major differences" between a determination of a head-tenancy by surrender, and determination of a head-tenancy by service of a tenant's break notice. The first difference is that a surrender determined a tenancy "prematurely otherwise than at the time and in the manner stipulate& in the tenancy, whereas determination by a break notice was "at a time and in the manner previously agreed between the parties". Secondly, because a surrender had not been agreed in advance, it requires the landlord's consent, whereas "no further consent [was] necessary" in the case of a break notice. Thirdly:

"A subtenant holds a derivative title which cannot be prejudiced by the surrender of the head-tenancy from which it is derived or any other agreement between the parties to the head-tenancy which is later than the creation of a subtenancy. His title is, however, precarious, for it cannot survive the natural termination of the head-tenancy in accordance with its terms agreed before his subtenancy was create& (see at 272C to D).

Fourthly, if the head-tenancy is surrendered, it is fictionally treated as continuing, so as to support the existence of the subtenancy, but, if determined by notice, the head-tenancy "has come to the end of its natural life", and cannot be treated as surviving.

In Barrett, the landlords served a notice to determine a head-tenancy (which they were contractually entitled to do) and the head-tenant agreed in advance not to serve a counter notice, which would have given him protection under the Agricultural Holdings legislation. The Court of Appeal had held that this arrangement should be treated as a surrender, because the head-tenant had bound himself not to serve the counter notice, and accordingly the subtenancy survived as against the landlord. That reasoning was described by Lord Millett as inconsistent - with "the orderly development of the common law" (see at 274G).

There is undoubtedly a formidable pragmatic case for concluding that a landlord and head-tenant can agree precisely that which CLRP and PW purported to agree in clause 5(6), namely that, notwithstanding determination of the head-tenancy by an upward (or indeed a downward) notice, any subtenancy will survive. Indeed, given that such a consequence could not be said to be contrary to public policy or to any statutory principle, the natural instinct of anyone other than a property lawyer, and, indeed, the natural instinct of some property lawyers, would be to conclude that such a result is permissible. This instinct is supported by principle, as is indicated by the observation of Sir George Jessel at 465 in Printing and Numerical upon which Mr Reynolds relies:

"It must not be forgotten that you are not to extend arbitrarily those rules which say that a given contract is void as being against public policy, because if there is one thing which more than another public policy requires it is that men of full age and competent understanding shall have the utmost liberty of contracting, and that their contracts when entered into freely and voluntarily shall be held sacred and shall be enforced by courts of justice. Therefore, you have this paramount public policy to consider - but that you are not lightly to interfere with this freedom of contract."

Nonetheless. as is made clear in that passage, Sir George Jessel was dealing with contracts which were ' said to be void on public policy grounds, and, more importantly for present purposes. he was concerned with cases which might involve an arbitrary extension of pre existing law.

It is fair to say that there are aspects of the speech of Lord Millett in Barrett which can be said to support, or at least to be consistent with, the notion that a landlord and a head-tenant can contract out of the general rule. Thus, the thrust of Lord Millett's reasoning can be analysed as meaning that the landlord's right to treat the head-tenancy as determined by the head-tenant's break notice should have priority over the rights of a subtenant under the subtenancy, because the break clause pre-dates the grant of the subtenancy. If that is right, then one sees how it can be said that there is no reason why the landlord cannot agree in advance, for instance in the head-tenancy itself, that his right to possession pursuant to the break notice will, as it were, be postponed to the right of a subtenant (or, as in this case, a certain class of subtenant) to remain in possession under his subtenancy.

While I was and remain impressed by that argument, I have come to the conclusion that it is contrary to principle, and, indeed, contrary to the reasoning of Lord Millett, and consequently that it would not represent an "orderly development of the common law'

As explained by Lord Millett, it appears to me that the principle is that the determination of a head-tenancy in accordance with its terms will lead to the destruction of any subtenancy, but where the determination of the head-tenancy arises from some consensual arrangement between the landlord and the head-tenant not provided for in the head-tenancy, then the subtenancy will survive. This view seems consistent, in particular, with the observations of Lord Millett at 272A to B, and 272C to E. As I have said, care must be taken lest one places too much weight on words in a judgment where the tribunal did not have in mind the point at issue in the instant case. Nonetheless, in those two passages, Lord Millett did say that the subtenancy "cannot survive the natural termination of the head-tenancy in accordance with its terms agreed before the subtenancy was created", and that, "when [the head-tenancy] is determined by notice to quit, it has come to the end of its natural life. There is no further period remaining during which the tenancy can have continuance". It appears to me that, in those observations, he was stating, or applying, a point of general principle.

I consider that this conclusion is supported by three further points. First, it is important to remember in connection with this sort of issue that a tenancy is not merely a contract: it is, and it creates, an estate in land. When a tenant grants a subtenancy, he is granting a subsidiary estate out of the estate vested in him by the head-tenancy. As a matter of principle, it would seem to follow ineluctably that, if and when the head-tenancy determines, and the estate thereby created ceases to exist, any subsidiary estate carved out of it, including any subtenancy, must also determine. It is, I suppose, an example of the maxim Nemo dat quod non habet. Ultimately, that is the simple proposition upon which the general rule and decisions such as Pennell and Barrett rest. If that is the right analysis, it is difficult to see how a subtenancy can survive a destruction of the head-tenancy simply because the landlord and head-tenant agree that it should.

The fact that it is well established that a subtenancy does survive in the case of a consensual surrender of the head-tenancy admittedly provides some apparent basis for a different conclusion. If, in such a case, the head-tenancy can be notionally treated as continuing for the purpose of preserving the subtenancy, that could be said to provide support for the view that such a notional continuance could be agreed between the landlord and the head-tenant, when agreeing that the head-tenancy is to be determined by notice. However, as I see it, the fact that there is a well established exception of general application to the normal rule, which is of itself of general application, does not by any means automatically mean that, in a specific case, it is open to the parties to agree another category of exception.

The surrender exception to the general rule is of great antiquity: it is described in Coke's Commentary upon Littleton Vol 11 s636 p 338b, cited in Barrett at 271E to F. It was explained by Lord Millett on the basis that, after surrender, "the head-tenancy... is treated as continuing until its natural termination so far as this is necessary to support the derivative interest of the subtenant" (see at 272D).

In my opinion, the exception is, to all intents and purposes, as well established in authority and principle as the general rule. Given that the general rule is based upon an easily identifiable consequence of the fact that a lease or tenancy creates an estate, I do not think it is legitimate to invent a new exception to the rule.

The fact that the common law developed the principle that, in a specific type of case, a head-tenancy which had determined could be treated as notionally continuing, albeit only for a specific and limited purpose, does not seem to me to justify the conclusion that it is therefore open to a landlord and tenant to agree in advance that there should be such a notional continuance in rather different circumstances. The fact that a legal fiction has been developed over the centuries by the common law to deal with a particular set of circumstances and to satisfy a legal principle (identified by Lord Millett at 271D-E in Barrett). cannot, at least without more, justify the conclusion that it is open to parties contractually to agree that the fiction will apply in a different set of circumstances.

I believe that this conclusion is supported by the fact that the common law exception to the general common law rule is not itself cut down merely because it can lead to an unjust result. As Simon Brown LJ said in Pennell at 202C, the decision in that case (whichever way it went) would apply equally to unlawful subtenancies as to lawful subtenancies. That is well demonstrated by Parker -v- Jones [1910] 2 KB 32, where a landlord, who accepted a surrender of a tenancy, was bound by an unlawful sub-tenancy. If the pragmatic approach, urged by "Milton Gate" and PW, rather than the principled approach, adopted by BT, were correct then this decision would presumably have gone the other way.

Secondly, as pointed out by Mr Dowding, the fact that a lease creates an estate does sometimes produce results which do not accord with the plain contractual intention of parties. Thus, in Jenkin R Lewis & Son Ltd -v- Kerman [1971] Ch. 477 at 496B to F, the Court of Appeal emphasised that a landlord and tenant cannot vary a lease by adding land to the demised premises, or by extending the duration of the term, even if that is what they plainly agreed and intended. At 496E to F, Russell LJ said this:

"It is not possible simply to convert the existing estate in the land into a different estate by adding more years to it, and even if the parties use words which indicate that this is what they wished to achieve the law will achieve the result at which they are aiming in the only way in which it can, namely by implying a fresh lease for the longer period and a surrender of the old lease..."

A little later, he made effectively the same point in relation to the purported addition of land to the existing demised premises.

Perhaps even closer to the issue in the present case is the principle that a tenant, who purports to grant a person a subtenancy for a term equal to, or greater than, the term of his tenancy, will in fact thereby assign his tenancy to that person, even though it is quite clear that, both parties intended the arrangement to be a subtenancy. The principle was confirmed by the Court of Appeal in Milmo -v- Carreras [1946] 1KB 3 06, where Lord Greene MR said this at 310 to 311:

"[I]n accordance with a very ancient and established rule, where a lessee, by a document in the form of a sub-lease, divests himself of everything that he has got (which he must necessarily do if he is transferring to his so-called sub-lessee an estate as great as, or purporting to be greater than, his own) he from that moment is a stranger to the land, in the sense that the relationship of landlord and tenant, in respect of tenure, cannot any longer exist between him and the so-called sub-lessee. That relationship must depend on privity of estate.... You cannot have a purely contractual tenure. Tenure exists by reason of privity of estate."

Both these cases, and in particular Milmo, expressly emphasised, indeed they turned on, the tenurial nature of a lease. They also provide very good examples of how that tenurial nature can, on occasion, result in the contractual intention of the landlord and tenant, however unequivocally and forcibly expressed, being thwarted, not by virtue of public policy or statute, but by the limitations placed on their freedom to agree by reason of the inherent nature of a leasehold estate. As I see it, just as a tenant cannot grant a subtenancy for a term equal to or exceeding that of the head-tenancy, so can a tenant not effectively agree with his landlord that any subtenancy will survive the expiry of the head-tenancy according to its terms.

Thirdly, there is the practical consequence if there is an ability to contract out of the general rule. The position is explained by Lord Millett in Barrett at 271B to D:

"Formerly the extinguishment of the tenancy by surrender also extinguished the reversion to any subtenancy, so that the remedy for the rent and the covenants attached to the reversion ceased with the reversion to which they were annexed. The subtenant held the property as tenant of the head-landlord for the residue of the term of the extinguished tenancy but without privity of estate and accordingly without an obligation to pay the rent or perform the tenant's covenants: see Webb -v- Russell (1789) 3 Durn & E 393. This unsatisfactory state of affairs was remedied by statute in two stages. Section 6 of the Landlord and Tenant Act 1730... (now section 150 of the Law of Property Act 1925) effected a partial reversal of the common law rule. Section 9 of the Real Property Act 1845... (now section 139 of the Act of 1925) reversed it more generally. Those provisions apply only where the head-tenancy is surrendered."

Section 139(1) of the Law of Property Act 1925 ("the 1925 Act") provides:

"Where a reversion expectant on a lease... is surrendered or merged, the estate or interest which as against the lessee for the time being confers the next vested right to the land, shall be deemed the reversion for the purpose of preserving the same incidents and obligations as would have affected the original reversion had there been no surrender or merger thereof."

Subject to the arguments to which I turn in the next two sections of this judgment, it appears to me that this represents another reason why a landlord and head-tenant should not be able to contract out of the general rule. As Lord Millett explained, although a subtenancy survived the surrender of the head-tenancy, there was, in common law, no basis upon which the landlord could enforce the covenants in the subtenancy, because surrender destroyed the reversion to the subtenancy and there was no privity of estate between the landlord and the subtenant. Although such a result could, at least in theory, have been avoided by treating a surrender as a notional assignment of the term of the head-tenancy to the landlord, that was not the analysis which the common law adopted: by a surrender, a tenancy became "absolutely drowned" (see Coke on Littleton, op. cit.). The fact that this was a conclusion reached only after investigation and with obvious regret by Lord Kenyon CJ in Webb (see at 403) appears to me, if anything, to emphasise its strength.

If the position in common law was that, on the surrender of the head-tenancy, a subtenancy survived, but the landlord could not enforce the covenants against the subtenant, then it seems to me that the position in common law must be the same if a subtenancy can survive determination of the head-tenancy by the exercise of a break clause. The legislature has dealt with this problem, but only where the head-tenancy is surrendered (or merged), through the medium of what is now section 139 of the 1925 Act ("section 139"). In those circumstances, subject to the arguments raised by Mr Reynolds and Mr Hodge, to which I must now turn, it seems to me to follow that, if a landlord and a head-tenant can contract out of the general rule, the result would be that the landlord is prevented from obtaining possession for the remaining contractual period of the subtenancy, but would not be entitled to recover any rent under (or enforce the lessee's covenants in) the subtenancy. I suppose it is possible that the landlord and the tenant in a case such as this could have overlooked that problem. However, in my view, the more telling point is that, by restricting the ambit of section 139 (and its predecessors) to surrender (and merger), the natural inference is that the legislature considered that it was not necessary to extend it to any other case, because in no other case would a subtenancy survive the determination of the head-tenancy.

The problem due to the lack of a reversion (absent the Human Rights dimension)

As mentioned, Mr Dowding for BT argues that, if a landlord and head-tenant can agree that a subtenancy will continue notwithstanding the determination of a head-tenancy by notice, then it would lead to the problem identified and reluctantly upheld by the court in Webb, a problem which has only been solved by Statute in the case of a surrender. Mr Dowding accordingly contends that this problem would exist in the present case, if the seven Underleases have survived the expiry of the head-tenancy. If that contention is correct, it is significant in the present case for two reasons. First, as I have said, it assists the conclusion that a landlord and head-tenant cannot contract out of the general rule. Secondly, if I am wrong in my view as to the ability of a landlord and head-tenant to contract out of the general rule, and there has been such a contracting out in the present case, the result would be, at least to the thinking of any fair minded commercial person, absurd. Each of the seven Underleases would continue beyond 24th June 2002, and would be binding on Milton Gate, until their respective contractual expiry dates, but Milton Gate would not be able to recover any rent because there would be no privity of estate between Milton Gate and BT. Mr Hodge and Mr Reynolds have a number of points to rebut Mr Dowding's contention, to which I now turn. (Mr Hodge also has a point under the 1998 Act, which was raised rather late, prompted by a very recently reported decision of the European Court of Justice. I propose to deal with it separately, reflecting the way in which the case was argued).

First, it is contended that the common law rule, as exemplified by the decision in Webb, has been effectively reversed in all cases by subsection (1) of section 141 of the 1925 Act ("section 141"). This provides, so far as relevant:

"Rent reserved by a lease, and the benefit of every covenant or provision therein contained... and every condition of re-entry... shall be annexed and incident to and shall go with the reversionary estate in the land... immediately expectant on the term granted by the lease..."

This provision had its origin in the Grantees of Reversions Act 1540, a statute enacted as a result of the Dissolution of the Monasteries, to enable those who acquired the former Monasteries and their lands to enforce the covenants in any pre-existing leases. However, not surprisingly, the provisions of section 141 are very differently worded from those of the 1540 Act, and consequently cases on the 1540 Act are not of assistance when construing the section(see e.g. per Upjohn LJ at 490 in In re King (Deceased) [1963] Ch 459). It appears to me that the argument that section 141(1) enables a landlord to enforce covenants in a subtenancy in any circumstances where the head-tenancy has been determined suffers from three problems. The first is that it would result in section 139, a provision in the same part of the same Act, being wholly otiose. The 1925 Act was intended to consolidate and significantly to change the pre-existing law, and given its authoritative and careful drafting, I think that any construction of a section which would result in another section being wholly otiose is unlikely to be correct. (It should be added that if section 141(1) has the effect contended for, section 65(2) of the 1954 Act would also be wholly otiose).

Secondly, 1 consider that the argument that section 141(1) reversed Webb involves a misunderstanding of the basis of the conclusion in that case. The effect of section 141 (1) is to make the right to claim rent under a lease go "with the reversionary estate in the land... immediately expectant on the term granted by the lease [in question]". The basis of the reasoning in Webb, as explained by Lord Millett in Barrett at 271B, is that "the extinguishment of the tenancy by surrender also extinguished the reversion to any subtenancy". Accordingly, section 141 (1) does not take the argument any further, because "the reversionary estate" referred to in the statute is the same as the "reversion to any subtenancy" referred to by Lord Millett. I do not consider that it is legitimate to construe "the reversionary estate in the land... immediately expectant on the term granted by the [under]lease" as being a reference to the estate in the land which, if the subtenancy did not exist, would give the right to possession of the land, in a case where it is very well established that, but for the existence of section 139, the reversion in question had ceased to exist.

Thirdly, there is the history of section 141(1). Consideration of the legislative history of a particular statutory provision is often unhelpful, even potentially misleading, particularly where the wording of the provision has changed. That is a point made graphically clear in the passage to which I have referred in King. However, the immediate predecessor of section 141(1) is the first and main portion of section 10(1) of the Conveyancing Act 1881, whose wording was effectively identical. It is clear from the tenth (1913) edition of Wolstenholme's Conveyancing and Settled Land Acts (edited by Sir Benjamin Cherry, who was primarily responsible for the drafting of the 1925 Act) from the notes to section 10(1) of the 1881 Act, that the section was not intended to have the effect for which Mr Reynolds and Mr Hodge contend. Indeed, the point was made by Farwell LJ giving the judgment of the Court of Appeal in Turner -v- Walsh [1909] 2 KB 484 at 494:

"The 10th section makes no alteration in the rights of anyone, but merely alters procedure, so as to give the right of action to the person entitled to the proceeds of such action."

A second, and not dissimilar, argument is raised by Mr Reynolds and Mr Hodge based on the provisions of section 141(2), which provides as follows:

"Any such rent, covenant or provision shall be capable of being recovered, received, enforced and taken advantage of, by the person from time to time entitled, subject to the term, to the income of... of the land leased."

The first reason for rejecting the reliance of Milton Gate and PW on section 141(1), namely that it would render section 139 (and section 65(2) of the 1954 Act) otiose, applies equally to their reliance on section 141(2). The third reason, namely the legislative history of section 141(2), also applies equally. That is because section 10(1) of the 1881 Act contained, almost word for word, in a single subsection, what are now two subsections, namely subsections (1) and (2) of section 141. The observations of Sir Benjamin Cherry in Wolstenholme (op.cit.), and of Farwell LJ in Turner, apply with equal force to the argument based upon section 141(2) as they do in relation to that based on section 141(1).

It is fair to say that the second reason for rejecting the argument based upon section 141(1) does not apply with such immediate force in relation to the argument based upon section 141(2): there is no specific reference to "the reversion" in the latter subsection. However, it appears to me that it would be extraordinary if an identically worded provision in a later Act had a much wider meaning than it had in an earlier Act, in circumstances where the later Act is a consolidating (albeit also a significantly amending) statute, which is intended to repeal and substantially to re enact the earlier Act, and the only distinction is that the new provision is in a separate subsection rather than the closing part of a larger immediately preceding subsection.

In any event, it seems to me that the reference, at the beginning of section 141(2) to "any such rent..." is to rent "reserved by a lease" which is "annexed and incident to... the reversionary estate..." in section 141(1). That brings one straight back to the second point which can be raised against the argument based on section 141(1).

An argument is also advanced by "Milton Gate" and PW based on section 141(3), which provides, so far as relevant:

"Where that person becomes entitled by conveyance or otherwise, such rent, covenant or provision may be recovered, received, enforced or taken advantage of by him notwithstanding that he becomes so entitled after the condition of re-entry or forfeiture has become enforceable..."

It is true that the words "by conveyance or otherwise" could scarcely be wider, as was recognised by the Court of Appeal in Electricity Supply Nominees Limited -v- Thorn EMI Retail Limited [1991] 2 EGLR 46, at 48H to J. However, I do not think that section 141(3) takes the argument any further, because it is merely concerned to defeat an argument against a reversioner bringing a claim against a lessee, based on how the reversioner acquired the reversion, or based on the point that the right to forfeiture has arisen. It does not deal with the centrally relevant issue for present purposes, namely whether it can be said that the person in question, in this case "Milton Gate", has the reversion vested in him at all.

In these circumstances, I do not consider that any aspect of section 141 answers the general point that, if a landlord and a head-tenant can contract out of the general rule, the landlord will not be able to enforce the covenants, and in particular will not be able to recover the rent due, under any subtenancy which survives the determination of the head-tenancy, by a break notice.

It seems to me that that view is supported by consideration of section 142, which is concerned with enforceability of the lessor's covenants. So far as relevant, section 142 provides that the lessor's obligations are "annexed and incident to and shall go with [the] reversionary estate". If, as was held in Webb, and still remains the common law, as explained by Lord Millett, the survival of a subtenancy following the determination of the head-tenancy carries with it the problem that the subtenant's covenants cannot be enforced because the reversion is destroyed, it appears to me to follow as a matter of logic that, in such a case, common law must also result in the subtenant being unable to enforce the head tenant's covenants as contained in the subtenancy, essentially for the same reason. Additionally, as Mr Dowding points out, section 142 only annexes the benefit of the lessor's covenants to the reversionary estate "if and as far as the lessor has power to bind" that estate. In the present case "the lessor" under the Underleases was PW and it had no power to bind its immediate landlord, CLRP or Allmanna. Whatever arguments can be raised in relation to section 141, it therefore appears to me that there is nothing in section 142, which could fairly be said to fill the gap on the other side.

On the facts of this case, it was suggested by Mr Reynolds and Mr Hodge that the conveyancing documentation provides an answer to the problem. I do not agree. If anything, it highlights the problem. Reliance is placed on clause 11 of the Licences, which contained a direct covenant between the subtenant (BT or MLEL) and the landlord (i.e. CLRP or Allmanna).

In my view, there are three, or, at any rate, two problems with that argument. First, it is clear that clause 11 expressly excludes any liability in relation to the payment of rent. Secondly, the liability under clause 11 only lasts "so long as the subtenant... is liable to the lessor thereunder". As the reasoning in Webb establishes, if the Underleases survive the determination of the head-tenancy, then "the subtenant", namely BT, is no longer "liable to the lessor thereunder", namely PW, and therefore the provisions of clause 11 would have fallen away at the very moment when Milton Gate's argument seeks to invoke them. Thirdly, given that clause 11 was a covenant with Milton Gate's predecessor, it is not immediately easy to see how the basis upon which Milton Gate could enforce it: there is no privity of contract, and no privity of estate between Milton Gate and PW. Mr Hodge suggests that the answer to that point might be found in section 56 of the Law of Property Act 1925, but, as he accepts, that would be a revolutionary proposition inconsistent with the characteristically trenchant observations of Walton J in Re Distributors and Warehousing Limited [1986] 1 EGLR 90 at 94E to F. In particular, Walton J said that, in order for section 56 to apply:

"It is essential that the third party, upon whom rights are thus sought to be conferred, should be in existence and identifiable."

On reflection, it may be that the answer to this third point is to be found in the reasoning of the House of Lords in P&A Swift Investments -v- Combined English Stores Group plc [19891 AC 632. However, the first two points seem to me to dispose of any reliance which Milton Gate can place on clause 11 of the Licences.

The other argument raised on behalf of Milton Gate in this connection rests on the definition of "the lessors" in each of the Underleases. In my judgment, for reasons which have already been explained, this cannot assist Milton Gate. The expression "the lessors" extends to "the estate owner for the time being entitled to the reversion immediately expectant on the determination of the term [of the subtenancy]", but, as with section 141 (1), any attempt to rely upon this definition runs straight into the reasons for the decision in Webb, as explained by Lord Millett in Barrett at 271B. In any event, if section 141 does not assist Milton Gate as a matter of general law, it is hard to see how Milton Gate could succeed as a matter of private contract, given that there is no privity of contract or privity of estate between "Milton Gate" and BT. and section 56 of the 1925 Act cannot assist. In this connection, I do not think that the decision in P&A Swift can help Milton Gate, because in no sensible way can it be said to be successor of "the lessors" under the Underleases, namely PW.

In these circumstances, I am of the view that both as a general proposition, and on the particular facts of this case, it would follow that, if the Underleases survived the determination of the Headlease on 24th June 2002, then the covenants in the Underleases, and in particular the covenants to pay rent, could not be enforced by Milton Gate against BT. In my judgment, that is a powerful reason, at the very least as a matter of commercial common sense, for concluding that Milton Gate and PW cannot have contracted out of the general rule.

The effect of the 1998 Act and the Convention

I turn now to Mr Hodge's argument based on the 1998 Act, which he raises if (as he and Mr Reynolds accept) section 139, and if (as I have found) section 141, do not enable a head-landlord to recover the rent under the subtenancy in the event of the head-tenancy being determined other than by surrender, when construed in accordance with the traditional approach to construction. In that event, he argues that either or both sections of the 1925 Act should be construed in a different way, and in particular should be construed so as to apply to a case where the head-tenancy has been determined by notice, in light of the provisions of section 3(1) of the 1998 Act ("section 3").

The argument proceeds on the basis that, if the determination of the head-tenancy by the Notice did not put an end to the Underleases, but the lessee's covenants in the Underleases are unenforceable by Milton Gate, that would infringe Milton Gate's rights under Article 1 of the First Protocol to the European Convention on Human Rights ("the Convention"). This Article ("Article 1") provides, in summary, that every person is not to be deprived of his possessions and is to be entitled to the peaceful enjoyment of his possessions, subject to law. In those circumstances, Mr Hodge argues that, in light of section 3, the court should strain to arrive, and can arrive, at a construction of a relevant statutory provision which deals with the problem (i.e. section 139 and/or section 141) so as to extend it to apply to a case such as this. Section 3 requires legislation, so far as possible, to be read and given effect in a way compatible with rights under the Convention.

It seems to me that Mr Hodge's argument raises four issues. The first is whether section 3 can be invoked in a case such as this, where the rights concerned arise under an agreement, namely the Headlease, entered into long before the 1998 Act came into force. The second issue is whether Article 1 is engaged at all. The third issue is whether it is properly possible to construe sections 139 and/or 141 so as to render the covenants in a subtenancy enforceable as between the landlord and the subtenant, in the event of the head-tenancy coming to an end other than by merger or surrender. The final issue which then has to be considered is whether this justifies a different result from that which I have so far arrived at on the question of whether the determination of the Headlease resulted in the determination of the Underleases.

The extent to which section 3 of the 1998 Act can be retroactively applied has been authoritatively considered in Wilson -v- First County Trust (No. 2) [2003] 3 WLR 568.

At paragraph 19 of his speech, Lord Nicholls of Birkenhead approved what was said by Staughton LJ in Secretary of State for Social Security –v- Tunnicliffe [1991] 2 All ER 712 at 724:

"The true principle is that Parliament is presumed not to have intended to alter the law applicable to past events and transactions in a manner which is unfair to those concerned in them, unless a contrary intention appears. It is not simply a question of classifying an enactment as retrospective or not retrospective. Rather it may well be a matter of degree - the greater the unfairness, the more it is to be expected that Parliament will make it clear if that is intended."

In the following paragraph, Lord Nicholls, in agreement with an earlier judgment of Mummery LJ, concluded that "in general... section 3(1) does not apply to causes of action accruing before the section came into force".

Lord Hope of Craighead said at paragraph 98 that the presumption of retrospectivity "is based on concepts of fairness and legal certainty". He continued:

"These concepts require that ' accrued rights and the legal effect of past acts should not be altered by subsequent legislation. But the mere fact that a Statute depends for its application in the future on events that have happened in the past does not offend against the presumption."

In the next paragraph he said:

"To restrict the application of the interpretative obligation [in section 3 of the 1998 Act],'without exception, to 'events' that happened or 'transactions' entered into on or after 2 October 2000 would be to introduce a restriction which is not stated expressly anywhere in the 1998 Act."

At paragraph 161, Lord Scott of Foscote, while accepting that section 3 should not "have a general retrospective effect", went on to refer to an example which is pertinent to the present case:

"Where transactions calculated to continue for some considerable period are entered into, intervening legislation may in some respect or other affect the rights and obligations that accrue under the transaction after the legislation has come into force. Landlord and tenant legislation is a good example.

If a lease is granted for, say, 99 years, there might well be intervening legislation capable of affecting the ability of the landlord to forfeit the lease, to operate a rent review clause... But it would be unusual for the legislation to alter the rights and obligations of the parties resulting from events that had already taken place, such as a forfeiture notice already served... rent review machinery already in train, and so on."

Lord Rodger of Earlsferry, with whom Lord Hobhouse of Woodborough agreed, considered the question of retrospectivity, both in general, and in relation to section 3 of the 1998 Act, in illuminating detail in a passage beginning at paragraph 186. At paragraph 196, he said this:

"The presumption is against legislation impairing rights that are described as 'vested'. The courts have tried, without conspicuous success, to define what is meant by 'vested rights' for this purpose ... [A]s Lord Mustill observed... the basis of any presumption in this area of the law 'is no more than simple fairness, which ought to be the basis of every general rule'."

In paragraph 210, Lord Rodger accepted that "the presumption would be that Parliament did not intend that, when used to give to effect to them, the operative provisions [of the 1998 Act] should interfere with vested rights or pending actions". At paragraph 212, he concluded that "none of the operative provisions of the [19981 Act, including section 3, is retroactive".

In the present case, Mr Dowding contends on behalf of BT that it is not open to Milton Gate to invoke section 3 when construing the 1925 Act, because such an invocation would be in relation to transactions, namely the Headlease and the Underleases, entered into before the 1998 Act came into force, indeed, before the 1998 Act was enacted, or even in gestation.

While I see the force of this argument, I consider that it is not well founded. Although I accept that the notion of "vested rights" is somewhat elusive and unsatisfactory, it appears to me that the earliest that any "vested rights" could be said to have arisen under clause 5(6), was the date of the service of the Notice under that clause. Unless and until clause 5(6) was operated, the rights and obligations of any of the parties as a result of the exercise were merely contingent, and not vested. The reasoning of Lord Scott in relation to leases appears to me to be very much in point, and to support this conclusion.

If one approaches the question by reference to fairness, it does not appear to me to be unfair that the 1998 Act should be capable of being invoked to produce a result which the parties clearly intended at the time when they entered into their contracts, and was intended by the person who exercised the unilateral right which gives rise to the relevant issue. It is true that the Notice was served only four days after the 1998 Act came into force, but two points may be made in answer to that. First, the issue is simply whether the relevant step was taken before or after the 1998 Act came into force, and all lawyers are familiar with the vital importance of a particular date in many different circumstances. Secondly, the 1998 Act had, self-evidently, been on the Statute Book for around 2 years by the time it came into force; indeed, as Lord Rodger pointed out in paragraph 183 in Wilson, "the 1998 Act did not arrive on the scene unheralded".

Having reached this conclusion on the first issue, it is next necessary to consider whether Mr Hodge is correct in his contention that the provisions of Article 1 are engaged. In this connection, Milton Gate's argument is that if, following the determination of the Headlease pursuant to clause 5(6), the Underleases continue as against Milton Gate, but Milton Gate is unable to recover any rent due under the Underleases (or to enforce the other tenant's covenants therein), then this inability would constitute an infringement of Milton Gate's Article 1 rights.

The case upon which Milton Gate primarily relies in this connection is the decision of the European Court of Human Rights (Fourth Section) in Stretch -v- The United Kingdom (24 h June 2003, Times Law Reports of 3rd July 2003). In that case, a District Council had granted an applicant a lease of land for a term of 22 years. The lease required the applicant to erect buildings at his own expense on the land, and included an option in his favour to renew for a further 21 years. The applicant constructed the buildings, and in due course exercised the option. The Court of Appeal held that the grant of the option had been beyond the Council's statutory powers. The applicant applied for relief from the European Court of Human Rights ("the European Court") on the basis that he had been deprived of the benefit of the option in the lease, which infringed Article 1. The UK Government contended that, as the Council never had the capacity to grant the option, it had been a nullity from the start, and therefore there was no question of the tenant being deprived of any "possession".

The European Court rejected the UK Government's case. In paragraph 32, it explained:

"[A]ccording to the established case-law of the Convention organs, 'possessions' can be 'existing possessions' or assets, including claims, in respect of which the applicant can argue that he has at least a 'legitimate expectation' of obtaining effective enjoyment of a property right."

The Court then said in paragraph 34:

"While it is true that under English law the option was rendered invalid due to the operation of the doctrine of ultra vires, the Court observes that the applicants had entered into the agreement with [the Council] on the basis that he would have the possibility of extending the term of the lease. Neither party had been aware that there was any legal obstacle to this term forming part of the applicant's consideration for agreeing to the contract. The applicant proceeded to build... He clearly expected to be able to renew the option. ..."

In these circumstances, in paragraph 35, the European Court concluded that:

"[I]n the circumstances of this case, ... the applicant must be regarded as having at least a legitimate expectation of exercising the option to renew and this may be regarded, for the purposes of Article 1..., as attached to the property rights granted to him by [the Council] under the lease."

It was accordingly decided in paragraph 36 that, by refusing to renew the lease, the Council "may be regarded as frustrating the applicant's legitimate expectations under the lease and depriving him in part of the consideration which he gave in entering into the agreement". The European Court said that Article 1 was engaged either on the basis that there was "interference with the peaceful enjoyment of the applicant's possessions" or on the basis that there was "a deprivation of possessions", in each case within Article 1.

In paragraphs 37 to 41, the Court went on to consider whether the interference or deprivation could be justified on the basis that the law nonetheless represented a "fair balance" between "the demands of the general interests of the community and the requirements of the individual's fundamental rights". While accepting that the doctrine of ultra vires "provides an important safe guard against abuse of power by local or statutory authorities acting beyond the competence given to them", the Court effectively concluded in paragraph 39 that there was no question in that case of the Council acting against the public interest "in the way in which it disposed [or purported to dispose] of the property" (i.e. by the grant of the lease including the option) or "by giving effect to a renewal option". In paragraph 40, the European Court also expressed the view that it was no answer to say that the applicant or his advisers should have been aware of the law, given that the mistake was shared by all parties, including the Council.

So far as the present case is concerned, it seems to me plain that, at least until Milton Gate came on the scene, the Landlord under the Headlease, namely CLRP and then Allmanna, and the tenant thereunder, PW, clearly anticipated not merely that Permitted Underleases would continue, notwithstanding the expiry of the Headlease pursuant to clause 5(6), but that the Landlord under the Headlease would be able to enforce the tenant's covenants in the Underleases. Apart from anything else, it would have been wholly absurd for the Landlord to be entitled to the Penalty if the proviso to clause 5(6) was not satisfied (for instance, if the premises were empty), but was not to be entitled to the Penalty if the proviso was satisfied (and at least 75% of the premises was subject to some tenancies which had more than 5 years to run). If there had been any question of the Landlord not being able to recover the rents under such continuing Underleases, it would have been worse off, not better off, as a result of the continuation of any Underleases.

It is true that, from the moment it acquired the Landlord's interest,. Milton Gate had a different view of matters. However, given that it was effectively tied into whatever arrangement bound its predecessors, I do not think that makes any difference. I draw some support from the facts and decision in Stretch, where the identity of the landlord Council changed between the date of grant of the lease and the date of exercise of the option. Quite apart from this, although Milton Gate's expectation was that the Underleases would not survive determination of the Headlease pursuant to clause 5(6), that does not even begin to support the notion that it believed, or would have believed, that, if its understanding was wrong, the Underleases would survive the determination of the Headlease, without Milton Gate being able to enforce the covenants therein.

It is true that, as pointed out by Mr Dowding, Stretch involved a misunderstanding as to one party's capacity to enter into the contract, whereas in the present case, the misunderstanding was as to the consequences of a subtenancy surviving the determination of the head-tenancy by an upward notice. However, in each case, the misunderstanding was the oversight of a legal technicality, shared by all parties involved. In terms of culpability, if it is relevant, it appears to me that, if anything, less blame could be attached to the parties in the present case than to those in Stretch. In that case, the powers of the Council were set out in sections 164 and 172 of the Local Government Act 1933, a fairly obvious place to look once one appreciated that the Council was a Statutory Body whose powers could well be circumscribed. The unenforceability of covenants in subtenancies following the determination of head-tenancies is a rather esoteric point, which may well not occur even to reasonably experienced conveyancers, bearing in mind that the only circumstance in which it would normally otherwise be encountered has been disposed of by section 139.

In light of the observations of the European Court at paragraph 32 in Stretch as to what constitutes "possessions" in Article 1, it seems to me that the ambit of the word is wide enough to cover the present case. Furthermore, if the Underleases would survive the determination of the Headlease, without the tenant's covenants being enforceable, Milton Gate would be kept out of four floors of the premises for over 5 years, and of the remaining three floors for some 13 years, without being able to recover any rent whatever. That is scarcely "peaceful enjoyment of [its] possessions".

I have considered whether the point really falls outside the ambit of section 3, because it is ultimately the common law, rather than Statute, which gives rise to the Article 1 infringement in the present case. On balance, I consider that that involves too narrow a reading of section 3. Where the legislature has, through a Statutory provision, sought to reverse a common law rule which would (albeit only as it subsequently transpires) infringe a Convention right, it seems to me that section 3 is properly engaged by that provision. The purpose of the section, as I see it, is to require the Courts, so far as they properly can, to construe a Statute so as to eliminate (or failing that, to minimise) the extent to which English law involves, or results in, infringement of Convention rights.

In these circumstances, I am of the view that Article 1 would be engaged in the present case if the Underleases survived the determination of the Headlease on 24th June 2002, without Milton Gate being able to enforce PW's covenants therein, and in particular the obligation to pay rent.

The next question is whether it is possible to construe section 139 or 141 in such a way as to avoid the conclusion that, should the Underleases survive the expiry of the Headlease, there would nonetheless be no infringement of Milton Gate's Article 1 rights. In this connection, the obligation under section 3 to construe the 1925 Act so as to produce a result which is compatible with Convention rights, is only to be effected "so far as it is possible to do so". However, as Lord Steyn said in an important passage in R -v- A (No. 2) [2002] 1 AC 45 at 67F to 68E, section 3 imposes an "interpretative obligation [which is] strong" and permits, indeed may require, "an interpretation which linguistically may appear strained" (see also per Brooke LJ in Goode -v- Martin [2002] 1 WLR 1828 at 1840D to 1841A).

In my judgment, it is possible to construe sections 141 and 142, in light of the requirements of section 3, so as to enable a landlord under a head-tenancy, which has been determined by a break notice, to enforce the lessee's covenants as contained in the subtenancy. The arguments, which I have set out above, for rejecting such a conclusion, in circumstances where section 3 is not engaged, appear to me to be outweighed by the requirements of section 3.

As to section 141 (1), although the landlord's interest in such a case would not technically fall within the words "the reversionary estate... immediately expectant on the term granted by the [subtenancy]", it seems to me that it is capable of falling within those words as a matter of ordinary language. Thus, the landlord's interest can fairly be said to be "reversionary", in the sense that, after the determination of the head-tenancy, he would be kept out of possession unless and until the subtenancy expires. For the same reason, it would not be a misuse of language, to described his estate as "immediately expectant" on the term granted by the subtenancy, in the sense that there is no interest between the landlord's interest and that of the subtenant. In other words, the argument based on the historical and technical context of section

1), is outweighed by the fact that the words can be given a meaning which would enable the section to achieve a result which is compatible with the Convention, which is a result which would not otherwise be achievable. The fact that it could then be argued that section 139 is otiose, which would otherwise be difficult to accept, does not present nearly so much difficulty if it is part of the cost of complying with section 3. Equally, it seems to me that the current pressures exerted by section 3 should outweigh the relatively historic factor of the legislative ancestry of the section.

Even if that is not right, I would reach the same conclusion in relation to section 141(2). It is possible to read "such rent [or] covenant" as being a reference merely to the "rent reserved by a lease, and ... every covenant... therein contained" in section 141 (1), which would present Milton Gate's case with no difficulty. It is also possible to read the words "entitled, subject to the term, to the income... of the land..." as being a reference to the landlord's interest after the head-tenancy has been surrendered, on the basis that the words "subject to the term" mean "in the absence of the subtenancy". On that analysis, section 141(3) has a similar result.

It is right to add this. On the basis of essentially the same reasoning, it appears to me that the subtenant can enforce against the landlord the lessor's covenants in the subtenancy where the head-tenancy has been determined by a notice, on the proper construction of section 142, when construed in accordance with the requirements of section 3.

In these circumstances, I am persuaded that, in light of the provisions of section 3 and the decision in Stretch, the 1925 Act could, in this case, be construed, if possible, in such a way as to enable the covenants in a subtenancy to be enforceable as between head-landlord and subtenant, in the event of the head-tenancy being determined by a notice, and that this result can be achieved through the medium of a modem and purposive construction of section 141. That, then, brings me to the fourth issue, namely, whether this causes me to revise my conclusion that the Underleases did not survive the determination of the Headlease as a result of the Notice served pursuant to clause 5(6).

In my view, the above discussion and conclusion do not justify a different result, and, notwithstanding the impact of the Convention and the 1998 Act, I am of the view that the Underleases did not survive determination of the Headlease as a result of the Notice served pursuant to clause 5(6).

In the first place, it appears to me that the fact that the covenants therein would have been unenforceable if the Underleases had survived the determination of the Headlease, was only a supporting factor, albeit quite an important supporting factor, for the conclusion I have reached. That conclusion was primarily based on the nature of legal estates and well established common law principle, and, indeed, the desire to maintain "the orderly development of the common law".

Secondly, it seems to me that my conclusion as to the effect of section 3 by no means destroys any value which the argument based on Webb -v- Russell affords to BT's case. It is clear from the provisions of clause 5(6) that CLRP and PW intended, indeed agreed, that the Underleases would survive the determination of the Headlease pursuant to clause 5(6). However, as at the date of the grant of the Headlease, and indeed as at the date of the Underleases, there would have been no question of the 1925 Act being construed by reference to the approach required by section 3. Accordingly, it remains the case that the terms of Clause 5(6) (and the Underleases) were agreed at a time when, in light of the legislation then in force, the legislature appears to have assumed that a subtenancy would not survive the determination of the head-tenancy pursuant to a break notice.

Treating clause 5(6) as a surrender provision

On the basis that CLRP and PW could not, or did not, contract out of the general principle that a break notice served by a head-tenant determined any subtenancy, PW and Milton Gate contend that the Headlease should be treated as determined by surrender, rather than determined by a notice to break, on 24th June 2002, with the result that the Underleases continued, notwithstanding the determination of the Headlease. In my view, both limbs of that argument should be rejected.

Can clause 5(6) be treated as a surrender provision?

So far as the first limb of the argument is concerned, I accept that the law, particularly since equity prevails over the common law, is concerned with substance rather than with form. The law on this topic was discussed and applied in characteristically authoritative fashion by Wilberforce J in In re Stirrup's Contract [1961] 1 WLR 449 at 452 to 453. At 453, he quoted Lord Mansfield CJ in Edwards -v- Bail (1777) 2 Cowp. 597 at 600 as saying:

"The rules laid down in respect of the construction of deeds are founded in law, reason, and common sense: That they shall operate according to the intention of the parties, if by law they may: And if they cannot operate in one form, they shall operate in that, which by law will effectuate the intention."

Wilberforce J then immediately went on to apply that approach to the case in front of him:

"[A]s a matter of general approach, once the court is satisfied that the intention of a document is to pass the legal estate ..., it should not allow that intention to be defeated by the fact that instead of the word 'grant' or 'conveyance', the word 'assent' is used."

The words of Lord Mansfield are salutary, and I would not wish to derogate in any way from them; on the contrary. However, it would scarcely be consistent with what Lord Millett called "the orderly development of the common law" in Barrett at 274G, if those words were interpreted so widely as to enable the court, as a matter of construction to produce the result which it believes that the parties to the contract wanted, by rewriting the contract, characterising the nature of the contract as being different from what it is, provided only that that would not infringe public policy or statute.

To construe a document, which on its face, plainly intends to effect a transfer of a property, as conveying the property, notwithstanding the fact that the parties have not used the appropriate language of "conveyance" or "grant", but instead have used the word "assent", is a long way away from the characterisation of the effect of the service of a notice pursuant to a break clause in a lease, as effecting a surrender of the lease. Indeed, it appears to me that, the proper characterisation of the relationship between the parties or the effect of an act permitted by contract, is ultimately a question of law, although the intention of the parties (particularly if expressed in the document concerned) may be a significant factor in some cases. As Mr Dowding says, that proposition is familiarly demonstrated in the landlord and tenant field by the decision of the House of Lords in Street -v- Mountford [1985] AC 809. In that case, because the agreement in question involved the grantor giving exclusive possession to the grantee in return for a periodic payment, the arrangement created a tenancy, notwithstanding the fact that it was plain from its terms that the parties intended it to create a licence.

In the present case, clause 5(6) permitted PW to determine the Headlease on 24th June 2002 simply by serving a notice of requisite length, and nothing more was required to put an end to the term. To my mind, if the parties had characterised that arrangement as a surrender, they would have been attempting that which Lord Templeman said in Street they could not effectively do. As he graphically observed, the parties can no more turn an agreement, which in law creates a tenancy, into a licence by calling it a licence, than they can turn a five pronged digging instrument, which as a matter of concept and language is a fork (albeit with a supernumerary tine), into a spade by calling it a spade. Clause 5(6) in the instant case is a provision which entitles the tenant thereunder to determine it simply by service of a notice: that is determination by notice, and not determination by surrender. A surrender, as a matter of law, involves some subsequent consensual act, such as execution of a deed of surrender or handing over and acceptance of the key, before the transaction is completed. The fact that there can be a unilateral aspect to surrender, in the sense that the landlord - or indeed the tenant - can contractually commit himself in advance to execute a deed of surrender, does not call that point into question, in my view.

What if clause 5(6) was a surrender provision?

Even if the determination of the Headlease on 24th June 2002 could be characterised or treated as a determination by surrender, rather than a determination by notice, I tend to the view that it would not have resulted in the Underleases continuing, if they would not have continued if the determination was pursuant to a break notice. It appears to me that the contrary view proceeds on the assumption that a subtenancy will survive the surrender of the head-tenancy even where the surrender of the head-tenancy is effected pursuant to a unilateral right to require a surrender contained therein. In my judgment, that assumption is incorrect, and, indeed, is inconsistent with the reasoning of Lord Millett in Barrett.

The distinction which Lord Millett drew was between two categories of case namely a unilateral right to determine the head-tenancy, contained in the head-tenancy itself, and a bilateral agreement to determine the head-tenancy entered into after it had been granted - see the passages cited above from his speech. In the present case, characterising clause 5(6) as giving rise to a surrender would not alter the fact that it is a provision in the head-tenancy, permitting the head-tenant unilaterally to determine it, albeit by surrender. It seems to me that such a re-characterisation of clause 5(6) would not alter the fact that determination of the Headlease pursuant to its terms would fall into Lord Millett's first category, rather than his second category.

However, it is right to acknowledge that, if clause 5(6) could be characterised as giving effect to a surrender of the Headlease, it would dispose of one of the reasons I have identified (subject to the impact of the 1998 Act) as to why the Notice should not have put an end to the Underleases. The reason in question is, of course, that, if the Underleases survived the determination of the Headlease, then "Milton Gate", although bound by the Underleases, could not enforce any of the covenants therein. If, however, clause 5(6) can be interpreted as effecting a surrender, then this particular problem is avoided, because section 139 would apply. Nonetheless, the fact that that problem would be avoided, if clause 5(6) can be interpreted as a surrender provision, does not seem to me to justify disapplying the general principle which, in my view, is to be extracted from the reasoning in Barrett, as discussed above.

In these circumstances, I am of the view that, although CLRP and PW purported to agree to disapply the general rule (that a subtenancy could not survive the unilateral determination of the head-tenancy in accordance with its terms), it is not possible to contract out of this general rule, and, accordingly, the Underleases all determined on 24th June 2002, when the Notice took effect. This conclusion is, however, subject to two qualifications. The first is that the determination of the Underleases in accordance with the general rule does not alter the fact that the BT Underleases could continue pursuant to the provisions of the 1954 Act, because BT was in occupation of the lower five floors on 24th June 2002. Secondly, this conclusion is subject to the effect of the argument based on estoppel, to which I now turn.

Estoppel by deed

The principle of estoppel by deed is explained in summary form in paragraph 1018 of Halsbury's Laws of England (4th edition) Vol. 16(1), in the following terms:

"Estoppel by deed is based on the principle that, when a person has entered into a solemn engagement by deed as to certain facts, he will not be permitted to deny any matter which he has so asserted. ... The averment relied upon to work an estoppel must be 'certain to every intent' without any ambiguity, but may be contained in the recital or any part of the deed."

Estoppel by deed differs from estoppel by convention, not only by virtue of the relatively restricted circumstances in which it can arise (as described in the passage I have just quoted), but also in that it requires no subsequent conduct or any other act of reliance by the party invoking the estoppel ("the claimant").

In my view, the only provision in the present case upon which an estoppel by deed could even be arguably raised is clause 5(6) itself, which is also to be treated as included in the recital to each of the Licences by virtue of section 58 of the 1925 Act, in light of the fact that each Licence is described as being "supplemental to" the Headlease. I do not consider that any estoppel by deed could arise from the fact that, in each Licence, the Underlease thereby permitted is confirmed to be a Permitted Underlease. First, as I have mentioned, clause 3(25) of the Headlease requires every underletting both to be licensed by the landlord and to be by the way of a Permitted Underlease. Therefore, the confirmation in the Licence could be said to take matters no further. Secondly, even if that is wrong, the furthest any reliance upon such confirmation in each Licence gets, is to suggest by implication that the subtenancy will survive the determination of the Headlease pursuant to clause 5(6). However, an estoppel by deed cannot arise by implication: see paragraph 1020 of Halsbury (op. cit.).

For the same two reasons, I do not consider that the provision in each Underlease, to the effect that the service charge sinking fund provisions will only apply as from 18th June 2002 can give rise to an estoppel by deed. First, as I have mentioned, it is justifiable on the basis that the Underleases could well be expected to continue under the provisions of the 1954 Act. Secondly, in any event, the argument that the sinking fund provision indicates that the parties expected the Underleases to survive the expiry of the Headlease pursuant to clause 5(6), is based on an implication.

I turn, then, to consider whether an estoppel by deed could have arisen as a result of the provisions of clause 5(6) itself. I do not consider that it could. In Greer -v- Kettle [1938] AC 156, Lord Maugham said at 170, after referring to the fact that the court had decided that an estoppel by deed can arise from a recital in a deed:

"Subsequent cases laid down that the recital must relate to specific facts, must be certain, clear and unambiguous, and would not avail persons who were not parties or privies to the deed."

In this connection, see also Onward Building Society -v- Smithson [1893] 1 Ch 1 at 14, and re Distributors at 93.

In the present case, far from being a clear and unambiguous express statement as to present fact, clause 5(6) appears to me to encapsulate an agreement as to what may arise in the future as a matter of law: I use the word "may", because the event is dependent upon a Notice being served under clause 5(6), and upon there being extant Permitted Underleases at the date such Notice takes effect. I do not think that that can found an estoppel by deed. I believe that this conclusion is also consistent with a passage at the end of the judgment of Park J in CP Holdings Ltd -v- Dugdale (unreported) to the following effect:

"X and Y enter into an agreement: 'Whereas we believe that the effect of the new agreement will be MNO, we now agree as follows.' X later wants to argue that the effect of the new agreement on its true construction is not MNO, but is PQR instead. He is not estopped from doing so. He may have an uphill struggle in his arguments on construction, but he is not estopped from putting them forward. In a case like that it is rectification or nothing."

It is argued by Mr Hodge that this view is inconsistent with the decision of the Court of Appeal in Poulton -v- Moore [1915] 1 KB 400 where, by a majority, the Court of Appeal concluded that a recital in a deed was sufficiently precise to estop a party from asserting that a right of way was not released. It is clear that that was regarded as a statement as to present fact: see per Buckley LJ at 412, where he referred to the relevant provision being "a recital that the state of facts exists at the moment" (in a passage with which Phillimore LJ agreed at 415). However, the fact that Pickford LJ, at 416, was "not satisfied that the statement in this recital is so precise as to operate as an estoppel", indicates just how strictly the court approaches an estoppel by deed, given that the relevant words in the recital in that case were that the predecessor of the person estopped "is entitled" to a right of way.

It may, at least at first sight, seem out of keeping with the modem, relatively flexible view of estoppel that such a strict approach should be adopted to any type of estoppel. However, as I have mentioned, unlike most other estoppels, an estoppel by deed will, if its strict conditions are satisfied, normally operate as an estoppel without more - see e.g. per Lord Mansfield CJ in Edwards -v- Bailey (1777) 2 Cowp. 597 at 600 to 601. Estoppel by deed is described in Halsbury paragraph 1018 (op. cit.) as "a rule of evidence", and there is, therefore, normally no question of considering the issue of unconscionability, which looms so large in relation to other estoppels, including estoppel by convention.

In those circumstances, it is not hard to understand why the court should tend to limit the ambit of estoppel by deed. After all, if a provision in a deed, such as clause 5(6) in this case, is insufficient to found an estoppel by deed, but is sufficient to indicate a common understanding or common assumption, then, if there is subsequent conduct based on that understanding, such that it would be unconscionable for one party to resile from the common understanding, then estoppel by convention would frequently come into play.

Estoppel by convention: the existence of a convention

In Spencer Bower and Turner on Estoppel by Representation (3rd Edition) at page 157, estoppel by convention is described in these terms:

"This form of estoppel is founded, not on a representation of fact made by a representor and believed by a representee, but on an agreed statement of the facts the truth of which has been assumed, by the convention of the parties, as the basis of a transaction into which they are about to enter. When the parties have acted in their transaction upon the agreed assumption that a given state of facts is to be accepted between them as true, then as regards that transaction each will be estopped against the other from questioning the truth of the statement of facts so assumed."

In Amalgamated Investment & Property Co Ltd -v- Texas Commerce International Bank Ltd [1982] QB 84, Lord Denning MR put it thus at 122C-D:

"When the parties to a transaction proceed on the basis of an underlying assumption - either of fact or of law - whether due to misrepresentation or mistake makes no difference - on which they have conducted the dealings between them - neither of them will be allowed to go back on that assumption when it would be unfair or unjust to allow him to do so. If one of them does seek to go back on it, the courts will give the other such remedy as the equity of the case demands."

The circumstances in which an estoppel by convention can arise have been considered in a number of cases. Given its relatively flexible nature, it is perhaps inevitable that one can detect differing approaches in various judgments. In some cases, such as Amalgamated Investment, the court has sought to lay down and apply general principles; in other cases, such as Keen -v- Holland [1984] 1 WLR 251, the court has identified limits to the general principles; in yet other cases, such as The Vistafjord [1988] 2 Lloyds LR 343, the court has extended the principles; there are also cases, such as Johnson -v- Gore Wood & Co. [2002] 2 AC 1, in which authoritative, if somewhat generalised, doubts about the extent of the principle have been expressed; finally, there are cases, such as John -v- George (1995) 71 P&CR 375, where the principle has been expressed virtually as being one of unconscionability, almost no more and no less.

It is convenient to start by considering whether a convention, by which I mean a convention sufficient in principle to found an estoppel, has been established between PW and its landlord under the Headlease (i.e. the Landlord) up to the time Milton Gate acquired the freehold reversion. I shall then consider whether BT were also bound by this convention on taking the Underleases. Next I shall deal with Milton Gate's position on acquiring the reversion to the Headlease.

Milton Gate and PW contend that a convention arose between them by virtue of.

i) The fact that CLRP and PW entered into the Headlease on the common understanding that the general rule would not apply, at least in relation to Permitted Underleases, in the event of the Headlease being determined under clause 5(6);

ii) The fact that the Licences were entered into and the Underleases granted after the grant of the Headlease and that they were based on the common misunderstanding of PW and the Landlord that the general rule would not apply.

Was there a convention between the Landlord and PW as a result of the Headlease?

I accept that it is clear from the terms of clause 5(6) itself (supported by the correspondence between, and evidence as to the then-state of mind of the parties and their advisers) that CLRP and PW both believed that the general rule would not apply, and that clause 5(6) was negotiated and agreed on this basis. However, I do not consider that that is capable, on the basis of the law as it stands, of founding a convention which could give rise to an estoppel.

I reach this conclusion because of the reasoning of the Court of Appeal in a judgment delivered by Oliver LJ in Keen. The issue in that case arose from the curiosity that, while a tenancy of an agricultural holding of a periodic nature, or of a fixed term for one year or more than two years, is afforded Statutory protection, a tenancy for a fixed term of between one year and two years is not afforded such protection. In that case, it was clear from the negotiations between the parties that the landlord was only prepared to grant a term between one and two years, and that the parties therefore intended that the term be for such a period. However, they made a mistake in believing that the term granted ran from the date from which it was expressed to run, rather than from the date of the tenancy. When the tenant claimed the protection of the agricultural holdings legislation, the landlord contended that the tenant was estopped from relying on such an argument.

The first ground upon which the Court of Appeal dismissed the estoppel argument was that it was impossible for parties to estop themselves out of a Statute such as the Agricultural Holdings Act 1948. However, Oliver LJ continued at 261F to 262A:

"That is sufficient to dispose of the argument but there seems to us to be other insuperable obstacles to a successful plea of estoppel. This is not strictly a case of the parties, having established by their construction of their agreement or by the apprehension of its legal effect, a conventional basis upon which they have regulated their subsequent dealings... The dealing alleged to give rise to the estoppel is the entry into the agreement itself in the belief that it would produce a particular legal result. In fact, for reasons which have nothing to do with the defendant, the plaintiffs got it wrong: and what [counsel] appears to us to be contending for is a much wider conventional estoppel than has yet been established by any authority, namely, that where parties are shown to have had a common view about the legal effect of a contract into which they have entered and it is established that one of them would not to the other's knowledge have entered into it if he had appreciated its true legal effect, they are, without more, estopped from asserting that the effect is otherwise than they originally supposed.

So broad a proposition cannot be deduced from the actual decision in the Amalgamated Investment case and although it may be supported on the basis of the very wide proposition of Lord Denning MR..., it cannot, in our judgment, be right."

Unless that passage does not represent the law, it seems to me that it must dispose of the first way in which the estoppel by convention argument is run by Milton Gate and PW. Mr Hodge's suggestion that the words "without more" indicate that a convention could be established in the circumstances described by Oliver LJ, provided that the essential additional element of unconscionability could also be established, does not appear to me to accord naturally with the meaning of the words in that context. In my opinion, Oliver LJ was indicating that some course of dealing after the contract in question had been entered into was necessary. That conclusion is supported by the fact that the Judge below had expressly dealt with the question of unconscionability, and it seems to me that, if by "without more" the Court of Appeal had had in mind unconscionability, they would have dealt with the Judge's finding on that issue.

It is true that passages in two subsequent Court of Appeal judgments, namely in The Vistafjord and in John -v- George, support the view that estoppel by convention has a wide and flexible doctrine, and there is nothing in the way in which some of the observations are expressed which appears to exclude in terms a convention arising before the contract in question was entered into. However, to construe any of those observations as being inconsistent with, let alone overruling, what was said by the Court of Appeal in Keen, appears to me to be illegitimate. In neither of the two later cases was there any question of the estoppel being based on a convention which had arisen from the negotiations for the contract in question, or as a result of the terms of the contract in question. Accordingly, the point with which I am now concerned was simply not in issue.

In any event, the principle articulated in the judgment of Oliver LJ in Keen, as I understand it, has been accepted and adopted in at least three first instance decisions. They are The Nile Rhapsody [1992] 2 Lloyds LR 399, per Hirst J at 407 to 408, Wilson -v- Truelove [2003] All ER(D) 302 per Mr Simon Berry QC at paragraphs 22 and 23, and CP Holdings per Park J in the passage quoted above.

Mr Hodge suggests that, in that passage, Park J was concerned with the meaning of what the parties had agreed, rather than the effect of what they had agreed. I do not consider that such a very fine, almost teleological, distinction justifies a different result. First, if one goes back to the facts of Keen, it was a case concerned with effect rather than meaning. Secondly, it is now well established that rectification, referred to by Park J in CP Holdings, is available where the mistake sought to be rectified relates to effect and not merely where it relates to meaning.

I also consider that it would be wrong to extend too readily the circumstances in which estoppel by convention can apply, in light of the concerns expressed by Lord Goff of Chieveley at 39D to 40H (and, arguably, by Lord Millett at 61A to B) in Johnson [2002] 2 AC 1. It is right to add that the other members of the House of Lords in that case were prepared to accept "the correctness in principle of Lord Denning MR's statement in Amalgamated Investment" as referred to, and quoted, by Lord Bingham of Cornhill at 33C to F.

Was there a convention as a result of the Underleases and Licences?

I turn, then, to the second basis upon which the convention is said to arise. In this connection, PW contends that, during 1994 and 1995, it granted each of the seven Underleases, and entered into each of the Licences in the belief that each of the Underleases would survive the determination of the Headlease, and that the Landlord under the Headlease shared this belief. On behalf of PW, Mr Reynolds contends, with the support of Mr Hodge for Milton Gate, that this was sufficient to create a convention, which could give rise to an estoppel which would prevent Milton Gate and PW, as between each other, from asserting that the Underleases cannot survive the determination of the Headlease pursuant to clause 5(6).

For BT, Mr Dowding first argues that, because the mistake in question was one of general law, it cannot found a convention which gives rise to an estoppel. It seems to me clear that an estoppel by convention can be founded on a mistake of law. That is most clearly established in terms by the judgment of Peter Gibson J in Hamel-Smith -v- Pycroft & Jetsave Limited (unreported, 5th February 1987) in a passage cited with warm approval by Bingham LJ in The Vistafjord at [1988] 2 Lloyds LR 351 to 352. Peter Gibson J suggested that the passage I have quoted from Spencer Bower is "better understood as [an] illustration... of estoppel by convention rather than a definition of that category of estoppel". He gave three reasons (which I do not think were intended to be exhaustive) as to why the statement was unsatisfactory as a general principle, the first of which was:

"The agreed assumption need not be of fact, but may be of law. That is exemplified by the Amalgamated case itself, the assumption there being as to the effect [in] law of a guarantee."

I do not see the reasoning of the Court of Appeal in the passage in Keen quoted above as being inconsistent with this view. The fact that the mistake in that case was one of law does not appear to me to have been a problem for the person alleging the estoppel. That point is perhaps best illustrated by the reference early on in the passage I have quoted from the judgment of Oliver LJ to "their apprehension of its legal effect".

Mr Dowding suggests that there may be a difference between a case, such as the present, where the misunderstanding related to a general legal principle, and could not be characterised in any sensible way as a mistake of fact, and a case such as Amalgamated or the Vistafjord, where the mistake, although it could be characterised as one of law, could also be characterised as a mistake of fact, in the sense that the parties were under a common misapprehension as to the effect of their agreement. There is a distinction between a mistake as to the general law or a legal principle, such as whether a subtenancy can survive the determination of a head-tenancy , and the mistake as to the effect of a particular agreement, such as a mistake as to the type of under-tenancy to which a provision in a tenancy applies. However, it does not appear to me that there is any good reason for concluding that the former type of misapprehension cannot in any circumstances give rise to an estoppel by convention, if the latter type of misapprehension can do so.

Of course, in some cases of the former type, it may be impossible to give effect to the estoppel, either because it would infringe public policy or statute to do so, or because it would be simply conceptually impossible to do so, whereas such a problem is much less likely to arise in the latter type of case. However, I can see no reason in principle or practice why a common misapprehension as to the general law cannot, in an appropriate case, found an estoppel by convention. That view is reinforced by the fact that some misapprehensions can be characterised, according to one's viewpoint, either as mistakes as to the general law or as mistakes as to interpretation of private rights. The mistake in Keen provides a tolerably good example; the mistake there could be characterised as one of general law, namely a belief that the term of a lease runs from the date from which it is expressed to run, rather from the date of grant. It could also be described as a mistake as to private rights, namely the effect of the particular agreement into which the parties had entered. So, too, the mistake in this case.

PW's case, in so far as it is based on conduct subsequent to the execution of the Headlease, is as follows. First, when entering into the Headlease, and when entering into the subsequent Underleases and Licences, PW believed that the Underleases would survive the determination of the Headlease pursuant to a break notice, and that the proviso to clause 5(6) could apply, so that the Penalty would not be payable. Secondly, the Landlord for the time being shared this belief. Thirdly, on a fair reading of all the documents, namely the Headlease, the Underleases, and the Licences, that common understanding was effectively communicated between the parties. Fourthly, to a reasonable person in the position of the Landlord under the Headlease, it was, or at least should have been, apparent that PW entered into the Underleases and the Licences, in the belief and expectation that the Underleases would continue notwithstanding the determination of the Headlease under clause 5(6). Fifthly, PW would have acted differently if it had not had the relevant belief and expectation.

So far as the first step is concerned, it is clear on the evidence that, until "Milton Gate" suggested otherwise to PW in November 2000, PW plainly believed that the Underleases would survive the determination of the Headlease pursuant to a Notice pursuant to clause 5(6). As to the second step in the argument, it is clear that CLRP, and its successor, Allmanna, shared this belief, and, between them, they were the Landlord at the date of the grant of the Headlease, and at the date of the grant of each of the Underleases and the Licences.

I turn to the third step. It appears to me that this common understanding is clear from the Headlease, entered into between PW and the Landlord, the Licences between PW and the Landlord, and the Underleases, whose terms were specifically approved by the Landlord in the Licences. The terms of clause 5(6) show, as I have stated, that CLRP and PW believed that, at least if it was a Permitted Underlease, a subtenancy would survive the determination of the Headlease by a Notice, and that, if any such sub-tenancies had more than 5 years to run on 24th June 2002, and the total area occupied thereunder was more than 75% of the area of the premises, the Penalty would not be payable.

On a fair reading, it seems to me that the terms of the Licences, particularly when read together with the Underleases they permitted, demonstrate that this understanding continued to be held by PW and the Landlord, and was impliedly communicated between them. First, there seems little point in the provision in each Licence that the underlease thereby permitted would be a Permitted Underlease, unless such an acknowledgement was to ensure that such Underlease would survive determination of the Headlease pursuant to clause 5(6) thereof, and therefore falls within the proviso to that clause. Mr Dowding contends that PW may have wanted the comfort of knowing that the Landlord accepted that the Underlease was lawful, given that PW could only sublet pursuant to a Permitted Underlease, but I think that is pretty fanciful. Given that the Landlord was permitting the particular subtenancy by virtue of entering into the Licence, it would have added nothing for the landlord to agree that the subtenancy was a Permitted Underlease, unless it was for the purpose of confirming that the Underlease was capable of falling within the proviso to clause 5(6).

Each Licence was described as being "supplemental to" the Headlease, which, by virtue of section 58 of the Law of Property Act 1925 means that all the terms of the Headlease, including clause 5(6), are treated as set out in the recital to each Licence. Accordingly, the provisions of clause 5(6) of the Headlease, with all that they entailed, were treated as set out in a notional recital to each of the Licences.

Further, there is the fact that each of the Underleases was granted for a term which would extend well beyond that of the Headlease, if determined pursuant to clause 5(6), unless the subtenant exercised its right to determine the Underlease shortly before 24th June 2002. The fact that the Underleases could not be determined before their respective contractual term dates by PW strongly suggests that PW must have believed that the Underleases would survive determination of the Headlease under clause 5(6). Otherwise, PW would have been liable in damages to BT and MLEL, if it had served Notice under clause 5(6), because that would have led to a premature determination of the Underleases, to the potential, and quite possibly very substantial, disadvantage of BT and MLEL. That such liability would arise on the basis of derogation from grant appears clearly from the speech of Lord Millett in Barrett at [2000] 2 AC 274C.

I am of the view that the fourth step is also satisfied, essentially for the reasons discussed. The Landlord and PW were parties to each of the Licences. Confirmation from the Landlord in the Licences that the Underleases were Permitted Underleases could only have been required by PW because it might want to exercise its right to break under clause 5(6). The grant of Underleases (which were attached in draft form to the Licences) for terms going well beyond 24th June 2002 without any right in PW to break them, would only make commercial sense if PW thought that they would survive such break.

In this connection, I do not consider that this is a case such as The August Leonhardt [1985] 2 Lloyds LR 28, where the Court of Appeal rejected a claim based on estoppel by convention, because there was no "mutually manifest conduct by the parties ... based on a common but mistaken assumption"; each party "acted - or failed to act - independently from the other on the basis of a mutual mistake which remained uncommunicated between them" (per Kerr LJ giving the judgment of the Court of Appeal at 34 to 35). In the passage in the judgment in Hamel-Smith, which Bingham LJ endorsed in The Vistafjord at [1988] 2 Lloyds LR 351 to 352, Peter Gibson J said:

"What is important for an estoppel by convention is that there should have been a common assumption which has been acted upon. Further, such action need not follow immediately after the common assumption is made, nor be in the course of the same transaction between the parties."

Bingham LJ went on at 353 to refer to unilateral conduct being sufficient to give rise to a convention, if it was conduct "of which the other party was fully cognisant".

The fifth step is a little more difficult, in the sense that there is no statement in the evidence filed on behalf of PW to the effect that, if PW had not believed that the Underleases would survive the exercise of its right to break the Headlease, it would have acted differently in 1994 and 1995. The partner concerned with the grant of the Underleases of PW, Mr Walter Clark (now retired), and the person involved on behalf of CLRP at the time, have provided witness statements. Mr Clark does not say that PW would not have granted Underleases for terms which significantly extended beyond 24th June 2002, without some sort of protection for PW should it serve a Notice under clause 5(6), if it had appreciated that the Underleases would be determined by the service of such a Notice. Nor is there any suggestion that there would have been discussions between PW and CLRP to protect PW's position in relation to the underlessee, let alone. the Penalty, in the event of the true position having been appreciated.

Nonetheless, I am satisfied that, as a matter of commercial common sense, it is highly improbable that PW would have simply granted Underleases extending well beyond 24th June 2002 if it had appreciated the general rule, at least unless it had decided not to exercise clause 5(6). The effect of the general rule would have been not only that the Penalty of over £5m would have been payable to the Landlord, but unquantifiable and potentially very substantial damages would have been payable to BT (and to MLEL) by way of damages for derogation from grant. It seems very likely, in light of Mr Menzies's evidence, that PW was expecting to determine the Headlease under clause 5(6) at the time the Underleases were granted. Even though the evidence indicates that it was a tenant's market at the time, I believe it is not much short of fanciful to think that PW would not have tried to take steps to insulate itself against any liability for damages, if it had appreciated the true effect of exercising its right under clause 5(6) on any Underlease.

I accept that PW would have underlet the various floors of the premises in 1994 and 1995, irrespective of their understanding of the effect of clause 5(6) of the Headlease: the premises were surplus to PW's requirements, and were costing PW a substantial amount of money in terms of rent, and would continue to do so until, at the earliest, 24th June 2002. The obvious course for PW was either to sublet or to assign. However, I consider it substantially more -likely than not that PW would not have granted Underleases for terms expiring beyond 24th June 2002, unless it had some sort of protection from any claim by BT or MLEL for damages caused by the premature determination of each Underlease as a result of the Headlease being determined pursuant to clause 5(6). PW might have been able to obtain protection from such a claim by granting subtenancies expiring on 24th June 2002, by expressly precluding such a claim in each Underlease, or even (but much more speculatively) by obtaining an agreement from the Landlord to grant new tenancies to BT or MLEL in the event of PW serving a Notice. I accept that it is possible that the underlessees would not have accepted shorter Underleases, and that PW might not have been able to obtain any protection against liability to the underlessees in the event of exercising its right under clause 5(6). However, on the balance of probabilities, and at the risk of being accused of making a commercial assessment without any expert evidence on the point, I think that it is not merely very likely that PW would have been able to act and would have acted differently if it had appreciated the true position, but that it is more likely than not that PW would have acted differently. More specifically, I believe am prepared to assume that PW would have negotiated a provision in the Underleases exonerating PW for a liability for damages if it exercised its right to break the Headlease, especially bearing in mind that the Underleases had the protection of the 1954 Act. However, in that event, I accept that PW would have had to accord the underlessees a quid pro quo, probably by way of a somewhat reduced rent.

In these circumstances, I consider that there was a sufficient course of conduct after the grant of the Headlease, a sufficient common understanding, a sufficient communication of that understanding, sufficient acts of reliance on the part of PW, known to the Landlord, and sufficient detriment. Accordingly, I conclude that PW has established that, by 1994/1995, there was a sufficient convention to maintain an estoppel by convention against Milton Gate.

Were BT (and MLEL) bound by the convention?

On the basis that a convention arose between PW and the Landlord under the Headlease as a result of the Licences and the Underleases, Mr Hodge contends, on behalf of Milton Gate, that BT and MLEL were bound by the convention as a result of entering into the Underleases and the Licences. This argument is advanced on two bases. The first is that the convention that came into existence as between the Landlord and PW, as landlord and tenant under the Headlease, was of such a nature as to bind persons holding derivative interests from either of those parties, and therefore it extended to BT and MLEL. Secondly, it is contended that BT and MLEL became bound by the convention in light of the Licences (to which they were parties) being expressly supplemental to the Headlease - and in particular clause 5(6)- and the fact that each of the Underleases was for a term which, subject to the subtenant's right to break, would extend well beyond the determination of the Headlease if a Notice under clause 5(6) was served.

The issue of whether a subtenant is bound by (and entitled to rely on) a convention which has been established as between his immediate landlord and the head-landlord, particularly when the very act which gives rise to the convention is said to be the grant of the subtenancy itself (and associated licence), appears to be free of direct authority.

In that connection, I refer to paragraphs 1024 and 990 in Vol. 16(1) of Halsbury's Laws (op.cit.). Paragraph 1024, which is concerned with estoppel by deed, states that such an estoppel extends to the parties to the deed and their "privies". For that expression one is referred back to paragraph 990, which is concerned with the identity of "privies" to parties estopped by judgment, and the expression appears to extend to assigns. However, those passages are concerned with estoppel by deed or by judgment, and do not deal, at least in clear terms, with a subtenant of one of the parties.

In Brikom Investments Ltd. -v- Carr [1979] QB 467, Lord Denning MR expressed the view that an assignee of a tenant (or indeed of a landlord) could take the benefit of, and would be bound by, a promissory estoppel which had arisen between his predecessor and the landlord At 484G, he cited Coke on Littleton (op. cit.) vol. H p 352, a. b., to this effect:

"[E]very estoppel ought to be reciprocal... privies in estate, as the feoffee, lessee & ... shall be bound and take advantage of estoppels..."

However, Roskill LJ (at 489H to 490A) expressed doubts about this view where the estoppel was of a promissory nature and Cumming-Bruce LJ agreed with him (see at 490A). Nonetheless, all ' three members of the Court of Appeal agreed in the result, as the majority based their decision on waiver.

In Cuthbertson -v- Irvin (1860) 6 H&N 135, it was accepted in the judgment of Wightman J at 139 that, where there is a lease by estoppel, and the lessee assigns, the assignee would be estopped from denying the landlord's title, just as much as his assignor, the grantee of the lease, would have been estopped. The reasoning in Poulton also assumed, if not decided, that a successor in title of a person who is estopped by deed is himself estopped by deed.

In Hopgood -v- Brown [1955] 1 WLR 223, the Court of Appeal had no hesitation in holding that a successor in title would be bound by an estoppel between his predecessor in title and a neighbour as to the location of a boundary between their respective properties: see per Evershed MR at 224 to 225, and at 229 and 231 per Jerkins LJ and Morris LJ respectively.

In Valentine -v- Ash [2003] EWCA Civ 915 (unreported) a right of access was held to have arisen by estoppel (probably proprietary estoppel). At paragraph 65, Chadwick LJ said that the estoppel bound "successors in title".

None of these cases were concerned with the position of a subtenant, as opposed to an assignee. In London and Northwestern Railway Co. -v- West (1867) LR 2 CB 553, that point did arise. West was a tenant by estoppel from Budd, who was in turn a tenant by estoppel of the Company. It was held that West could not dispute the Company's title, as he derived his rights from Budd, who was himself estopped from denying the Company's title. The effect of the judgment of Willes J was that, as West could not deny Budd's title, and Budd could not deny the Company's title, it must follow that West could not deny the Company's title.

Estoppel is a flexible doctrine, which depends in almost all cases on unconscionability, and there are different species of estoppel, each of which has its own characteristics. Thus an estoppel of the sort considered in West is very different from that considered in Brikom. It is easy to see why a subtenant would be more easily bound by the former type of estoppel - directly concerned with title - than with the latter type - based on an oral promise. In these circumstances, it may be dangerous to conclude that as a matter of principle. any estoppel (or any convention) between landlord and tenant will be binding on, and enforceable by, a subtenant. However, I think it is possible to draw some conclusions as to the normal position.

The general thrust of the authorities to which I have referred tends to support the proposition that, once an estoppel (perhaps other than estoppel by representation) is established as between a landlord and a tenant, then it would normally bind their respective successors in title. As a matter of logic the same conclusion must apply to a convention which in due course may give rise to such an estoppel. If that is correct, then it appears to support the conclusion that a subtenant would also be normally bound, at least provided his subtenancy does not predate the coming into being of the convention. In my judgment, such an estoppel, and therefore such a convention, which unambiguously relates to the relationship of landlord and tenant, and can only fairly work if it extends to the subtenant, will generally do so, particularly if the subtenant had prior notice of it. In such a case, unconscionability would normally, I would have thought, work in favour of the estoppel binding the subtenant rather than not.

It would seem surprising if a subtenant of the tenant was in a stronger position than an assignee of the tenant: the former will have a shorter, more precarious, interest, and one more dependent on the continued observance of the obligations of the tenant, than the latter. It would be odd if, in the case of an estoppel between a landlord and a tenant under a 999-year lease, an assignee of that 999-year tern was bound by the estoppel, but a subtenant for 998 years, or, indeed, for 2 years, was not so bound. Further, if the rights of the landlord under the convention predate the rights of the subtenant under the subtenancy, the observations of Lord Millett as to the "general and salutary principle of law" at 2711)-E in Barrett suggest that the landlord's rights under the convention should prevail over those of the subtenant under the subtenancy. At least in some cases, it would be more difficult where the subtenant had no reason to know of the facts giving rise to the convention until after he took the subtenancy.

In this case, there is no problem in concluding that BT were bound by the convention even though the first four Underleases they were granted could be said to have created the convention. First, the convention was created by the first four Underleases. granted to BT, they subsequently took the fifth, and (by assignment) the sixth and seventh, Underleases. Secondly, in any event, I consider that BT were "locked into" the convention when they took the first four Underleases. The convention thereby established between landlord and tenant was one which was not merely intimately connected with both the landlord-tenant and the tenant-subtenant relationships. It was equally intimately connected with the landlord-subtenant relationship.

The conclusion that BT became bound by the convention by taking the first four Underleases might be said to appear somewhat inconsistent with the reasoning at the end of the judgment of the Court of Appeal in Keen. That is because it involves BT being bound by a convention as a result of entering into the very transaction which is said to give rise to it. However, I consider that there is a very considerable difference between a convention arising purely as a result of two parties entering into an agreement on the basis of a common understanding, and a third party becoming bound by a convention which arose from an agreement which he entered into with other parties (one of whom can be said to be his privy), who were already contractually bound, and between whom there was already a common understanding.

Another ground for justifying the inclusion of BT and MLEL in the convention between Milton Gate and PW, is that the convention could not work properly or fairly between landlord and head-tenant, unless the subtenant was also bound by it. If Milton Gate was bound by the convention as against PW, it would not be able to recover the Penalty on the basis that the Underleases were treated as continuing; it would scarcely be fair on Milton Gate, at least in the absence of further facts, if it could not treat the Underleases as continuing as against BT. Equally, from PW's point of view, the convention would only be of limited value if, having granted the Underleases with a view to escaping the Penalty, it nonetheless found itself facing a substantial claim for damages by the underlessees on the ground that the Underleases had been determined prematurely, in derogation from grant.

There is no question of BT being unfairly prejudiced if they are treated as having become bound by the convention as a result of entering into the BT Underleases, and the associated Licences. They entered into each Underlease with knowledge of all the relevant information, namely the terms of clause 5(6), the fact that the Licence stated that the Underlease was a Permitted Underlease, and the fact that the duration of such Underlease was, subject to BT's right to break, to extend well beyond the term of the Headlease if determined in accordance with clause 5(6). It is also clear that BT entered into the BT Underleases, and indeed the associated Licences, in the belief that the Underleases would continue even if the Headlease was determined pursuant to clause 5(6). Quite apart from this, at the time of the grant of the Underleases, and indeed even after service of the Notice, BT wanted their Underleases to continue until their respective expiry dates. No doubt, BT would have wished to enforce any convention, and they could only do so if they were bound by it.

Did Milton Gate become party to the convention?

In these circumstances, it is necessary to consider whether the convention could be relied on by PW as against Milton Gate and by Milton Gate against BT, at least before the events that have happened since the service of the Notice. In light of the authorities and principles I have just been considering, one would expect Milton Gate to have become bound by, and entitled to enforce, the convention, as successor in title to one of the parties to the convention.

However, Mr Dowding contends that, even if BT was bound by the convention, as against PW and CLRP/Allmanna, Milton Gate never had the right to rely on the convention against BT. This contention is based on the proposition that, from the moment it acquired the reversion to the Headlease, and indeed before that, Milton Gate never shared the common assumption that Permitted Underleases would survive the determination of the Headlease pursuant to clause 5(6), and that the sharing of a common assumption is an essential ingredient of estoppel by convention.

However, the existence of the convention had arisen and, due to the grant of the Underleases, its effect was, as it were, complete, before Milton Gate acquired the reversion to the Headlease. Further, as a successor in title, it seems to me that Milton Gate cannot be in a stronger position on this issue than CLRP or Allmanna, if they had remained the Landlord under the Headlease, but had been disabused of their misconception as to the effect of clause 5(6), in June 2000 (when, in fact, Milton Gate acquired the reversion). The convention arose as a result of the grant of the Underleases and Licences in 1994 and 1995, and the fact that the Landlord under the Headlease appreciated the true position in 2000 would plainly not be enough, of itself, to prevent PW enforcing the estoppel against the Landlord. Apart from anything else, it would have been necessary for the Landlord to communicate the change in its understanding (again applying the reasoning in the August Leonhardt), but even that would not have been enough, on its own, to dispose of the convention.

If, as I believe to be the case, the mere fact that the Landlord no longer suffered from the common misconception after 19th June 2000, was of itself insufficient to prevent PW enforcing the convention, then I think it must follow that Milton Gate was equally entitled to enforce that convention. Estoppel is based on equitable principles, and those principles include at least a presumption of mutuality. It seems to me that it cannot be right that a party is bound by a common entitlement to the benefit of a convention so long as he is ignorant of the law, but that he is no longer entitled to that benefit, but continues to be bound by the convention, simply upon appreciating the true legal position.

There is nothing unfair about this conclusion so far as BT is concerned. On the contrary. As I have explained, so long as the convention applies, it is not merely that BT is bound by it: BT would also be entitled to enforce it. For BT to lose the right to enforce the convention because of a change of head-landlord would seem unjust and contrary to the general principle that estoppels bind privies of the original parties.

Estoppel by convention: the existence of an estoppel

The position as between Milton Gate and BT

On 6th October 2000, when the Notice was served, the position was that, as a matter of strict law, it would have had the effect of determining the Underleases on 24th June 2002. However, by virtue of the existence of the convention to which I have referred, Milton Gate and BT were potentially estopped, as against each other, from so contending, and from asserting that the Underleases would not continue until their respective contractual expiry dates. If such an estoppel took effect, I think it would follow that, subject to the estoppel being in some way discharged thereafter, Milton Gate would be entitled to enforce the lessee's covenants, and would be bound by the lessor's covenants, in the Underleases. In that connection, quite apart from the effect of the 1998 Act, I see no warrant for extending the principle embodied in Webb -V- Russell to a case where a subtenancy becomes directly binding on a head landlord through estoppel. Equity is flexible and strong enough to ensure that any estoppel results in a sensible commercial outcome, which is not thwarted by archaic and technical rules of property law, unless those rules are based on public policy, or are so fundamental as to be incapable of being overridden.

The issue which has to be determined is whether the events subsequent to the service of the Notice enable BT to escape the consequences of the convention as against Milton Gate, thereby enabling BT to contend that the Underleases expired on 24th June 2002. Perhaps, it is more correct to formulate the issue as being whether Milton Gate is entitled to rely on the convention so as to give rise to an estoppel against BT in light of those events.

In this context, I consider that the centrally relevant question for the purpose of establishing an estoppel by convention in the present case is whether it would be unconscionable in all the circumstances for BT to go back on the common assumption, and to contend that the Underleases were in fact determined by PW's exercise of its right to break the Headlease under clause 5(6). Peter Gibson J made that point very clearly in Hamel-Smith in the passage quoted by Bingham LJ in The Vistafjord at 352, where he said this:

"A third reason why the two sentences in Spencer Bower should not be taken as definitions is that they are not qualified, as I think they should be, by considerations of justice and equity. Even if parties had acted on a common mistake or assumption it does not follow that the estoppel will follow as of course. ...[T]he court is not so rigid and inflexible as to insist on the parties being held to an assumed and incorrect state of fact or law when there is no injustice in allowing a party to resile therefrom..."

In John -v- George, Morritt LJ clearly considered unconscionability a vital factor: see at 384-385; Evans LJ agreed generally (at 391) and specifically on the issue of unconscionability (at the top of 393). At 396, Simon Brown LJ said:

"[F]or estoppel by convention to apply there must be, first, an assumption (shared and communicated between the parties) underlying the transaction and, secondly, unfairness or injustice in allowing the party seeking to benefit to go back on that assumption."

In Johnson. at [2002] 2 AC 41B to C, Lord Goff said

"In the end, I am inclined to think that the many circumstances capable of giving rise to an estoppel cannot be accommodated within a single formula, and that it is unconscionability which provides the link between them."

BT's case is that, as a result of the negotiations and discussions between BT and Milton Gate, there was a mutual resiling from the convention by April, or, at the latest, May, 2002. BT say that, in those circumstances, an estoppel by convention cannot be invoked, and in particular the convention could not be unilaterally revived against BT by Milton Gate, as it attempted on 0 and 5th July 2002. In my judgment, subject to one point, that contention is well founded.

So far as the facts are concerned, I have already described them. Shortly after the service of the Notice, namely from about April 2001, in its communications with BT, Milton Gate consistently and unequivocally contended that the Underleases would determine on 24th June 2002; in other words, it was unambiguously seeking to resile from the convention. BT were unhappy about this, and were effectively seeking to maintain the convention, but they negotiated with Milton Gate for the overriding lease of the premises, which, if granted, would have rendered the whole issue academic, at least as between BT and Milton Gate. However, in April 2002, BT pulled out of those negotiations, and, during April and May 2002, they made it clear to Milton Gate that BT were accepting, indeed (due to a change of plan on their part) relatively nthusiastically accepting, that the Underleases would indeed come to an end on 24th June 2002. There were then negotiations between BT and Milton Gate during May and June 2002 with a view to agreeing a basis upon which BT would remain in the premises until some time in 2003, when they would vacate.

In my view, subject to the effect of the convention in this case being tripartite, there was a clear and common understanding between BT and Milton Gate that the convention would no longer apply between them. Milton Gate had sought to resile from it, and, albeit after a substantial delay, in April 2002 BT accepted that resiling. Thereafter, late in June or July 2002, the parties were negotiating on the basis of the true legal position, namely that (subject to the effect of the 1954 Act) the Underleases would determine on 24th June 2002. Accordingly, subject to this being a tripartite convention, by the time that date arrived, the convention no longer existed as between Milton Gate and BT.

I believe that this conclusion derives support from the decision of the Court of Appeal in Gloyne -v- Richardson [2001] 2 BCLC 669, where at 683 to 684 Laws LJ said this:

"40. The doctrine of estoppel by convention was explained by this court in .... The Vistafjord... Bingham LJ said this at 352:

'It is sufficient... to say that it [sc. the doctrine of estoppel by convention] applies where (1) the parties have established by their construction of their agreement or their apprehension of its legal effect a conventional basis, (2) on that basis they have regulated their subsequent dealings, to which I would add (3) it will be unjust or unconscionable if one of the parties resiled from that, convention.'

41. I accept without demur conditions (1) and (2) in this formulation are established on the facts of this case. My difficulty is with (3). At the very least, once [Mr Richardson] had resiled from the terms of the letter agreement by asserting a claim based on option rights... it cannot be said that it was unconscionable for [Mr Gloyne] thereafter to depart from the 'convention' constitutedly the letter agreement, and to stand on the original term I s of [the agreement of 25th July 1988]. In these circumstances, in my judgment there is no estoppel by convention..."

Blackburne J and Aldous LJ, at paragraphs 55 and 65 respectively, agreed.

There is, however, a problem with applying the reasoning in Gloyne without modification to the present case. In Gloyne, as in most cases where estoppel by convention is alleged, there were two parties to the convention, and accordingly it was open to one party to resile from the convention once the other party had done so. There could normally be no reason why such a bilateral resiling could lead to an unconscionable result. However, in the present case, Mr Hodge and Mr Reynolds contend that it would be unconscionable to conclude that, as between Milton Gate and BT, there had been an effective resiling from the convention because of the consequences as between PW and Milton Gate. Otherwise, either Milton Gate would have the disadvantage of the strict legal position as against BT and the disadvantage of the convention as against PW, or PW would lose the benefit of the convention through no fault of its own. The first possibility would arise if the proviso was treated as satisfied against Milton Gate in favour of PW on the basis of estoppel by convention, although it is not satisfied as a matter of strict law, and BT could rely as against Milton Gate on the strict legal position. The second possibility would arise if, as a result of the mutual resiling between Milton Gate and BT, in which PW played no part, Milton Gate lost the ability to insist on the continuation of the Underleases and therefore became entitled to claim the Penalty from PW.

Mr Hodge says that it is clear from the way in which Laws LJ expressed himself in Gloyne that a mutual resiling from a convention does not dispose of any estoppel simply on the basis of an agreement between the parties. It is more that such a mutual resiling will thereafter no longer normally render it unconscionable for either party thereafter to rely upon the strict legal position. That appears to me to be correct.

Where the convention involves only two parties, and they communicate a desire or intention to resile from it, then, at least in the absence of unusual circumstances, there would be no justification for permitting either party to revert to the convention against the wish of the other party. However, where there is mutual communication of a desire or intention to resile between two of three parties bound by a convention, that may be insufficient to prevent one of those parties from unilaterally reverting to the convention as against the other party, because the third party was in no way involved in resiling, and is entitled to insist on the convention.

In these circumstances, it is right to turn to the question of estoppel by convention as between Milton Gate and PW and, having come to a view in relation to that, to reach a final conclusion with regard to the argument based on estoppel by convention as between all three parties.

The position as between Milton Gate and PW

I turn then to the question of whether PW can avoid having to pay the Penalty to Milton Gate (ignoring for the moment that BT was party to the convention). If the Underleases are due to continue until their respective contractual expiry dates as a result of the convention, then there can be no doubt but that Milton Gate is not entitled to the Penalty. If the Underleases determined on 24th June 2002, there is equally no doubt that, in light of the applicability of the general rule, PW would be liable to pay the Penalty to Milton Gate, at least as a matter of strict law.

There was a convention as between PW and Milton Gate, and, unlike the position as between Milton Gate and BT, there was no question of any mutual resiling from the convention. Throughout 2001, and into July 2002, Milton Gate was effectively trying to resile from any convention by contending that the Underleases would determine in June 2002, but PW was insisting that the convention subsisted and gave rise to an estoppel. In those circumstances, it appears to me that the question which needs to be considered, reverting back to the formulation of Bingham LJ (quoting from Peter Gibson J) in The Vistafjord, as quoted by Laws LJ in Gloyne, is whether it would be unconscionable for Milton Gate to insist on its strict legal right to recover the Penalty from PW.

In almost all cases, such unconscionability must be based on the prejudice which would be caused to the claimant if the strict legal position applied. As I see it, the claimant must also establish that that prejudice arises from its reliance upon the convention. In other words, the Court generally must be satisfied that (a) the claimant will suffer real prejudice, and (b) the prejudice arises from its reliance upon the convention. It should be emphasised that, even if the claimant satisfies these criteria, there may still be no estoppel, because there may be other, more powerful, factors pointing the other way.

In the present case, it appears that the prejudice which PW might allege as a result of having granted the Underleases is, first, liability for the Penalty to Milton Gate, and, secondly, liability to BT for damages based on derogation from grant owing to the premature determination of its Underleases. However, on analysis, it seems to me that, in this connection, neither of these two liabilities can properly found a contention that it would be unconscionable if Milton Gate were permitted to resile from the convention.

So far as the liability for the Penalty is concerned, it arises from the terms of clause 5(6) itself, and the service of the Notice in October 2002. It does not arise from the grant of the Underleases. Further, there is no evidence to show that, in 1994 or 1995, PW could or would have entered into arrangements (e.g. involving CLRP and/or Allmanna) which would have enabled the proviso to clause 5(6) to be satisfied. I accept that some sort of put and call arrangement could have been arranged between the Landlord and the underlessees, but that is far too speculative, hypothetical, and unusual an arrangement on which to base any estoppel. For PW to allege estoppel by convention against Milton Gate on the basis of a very unusual arrangement with third parties which Milton Gate's predecessors might have entered into, but were under no obligation to enter into and were not even asked to enter into, and of which PW might well have been asked to pay, possibly substantially, seems quite inappropriate, particularly where that arrangement would have been designed to deprive Milton Gate's predecessors of a potential substantial benefit, namely the Penalty. Indeed, no argument to that effect has, as I understand it, been raised. In these circumstances, in my judgment, PW's liability to pay the Penalty cannot be relied on to support an estoppel based on the convention arising out of the grant of the Underleases.

PW's reliance on its potential liability to BT for derogation from grant, arising from premature determination of the Underleases, looks much stronger. Such a potential liability to BT was a very real and substantial risk as at October 2000 when the Notice was served, as BT were anxious to remain in occupation of the premises. However, on closer examination, I am not convinced that it assists PW. As I have mentioned, I do not consider that the mere fact that PW granted sub tenancies can be said to give rise to any estoppel. By 1994, the premises were plainly surplus to PW's requirements, and, whatever its belief as to the effect of clause 5(6), it clearly would have granted sub-tenancies (albeit I suppose that it might have contemplated an assignment of the Headlease, although there is nothing to suggest that was or would have been in its mind). However, the fact that the Underleases were for terms going substantially beyond 2002 with no protection for PW if it exercised the break clause in the Headlease, which give rise to the convention.

On the other hand, by 24th June 2002, the date when the Headlease came to an end by virtue of the Notice, and the date when Milton Gate contends that its entitlement to the Penalty from PW arose, there was no possibility of any such claim being raised by BT, and, indeed, as at today, there is no such possibility. The reason is that, albeit contrary to their position up to that time, in April 2002 BT no longer wished to retain the Underleases, and were only too pleased to go along with Milton Gate's contention that those Underleases would expire on 24th June 2002, albeit subject to continuation (in the case of the BT Underleases) under the 1954 Act. Even if BT now wished to change their mind, I do not consider that they could sue PW for derogation from grant. The effective reason the Underleases would have determined on 24th June 2002 would be BT voluntarily joining Milton Gate in resiling from the convention. (This could be alternatively characterised as a failure to mitigate on the part of BT).

Accordingly, in my view, unless some other aspect of unconscionability or some other legal principle can be identified by PW, its defence to the claim for the Penalty should fail, subject to its being entitled to insist on the convention applying as between BT and Milton Gate. I turn to consider whether there is any other ground of unconscionability PW could raise against Milton Gate.

The only actions of PW which could be linked to the convention, are the entering into the Licences and the Underleases. and the service of the Notice. Other than the duration of the Underleases, it is not suggested that any of the terms of the Licences or the Underleases which were influenced by PW's belief as to the effect of the exercise of the right to break the Headlease under clause 5(6) on any Permitted Underlease, could be relied on to support an estoppel. The sinking fund provisions in the Underleases take matters no further in this connection. The existence and terms of the break clauses in the Underleases were plainly influenced by the existence and terms of clause 5(6). However, in the first place, those break clauses were never operated, and, secondly, if they had been operated it would have been to the benefit of PW, because of the Penalty provisions.

There is no evidence that the rent charged under the Underleases was any less than it would have been if PW had not been under a misconception as to the effect of the exercise of the provisions of clause 5(6). Indeed, commercial common sense, and the implication in the evidence of Mr Clark (the partner in PW responsible for negotiating the Underleases), suggest the contrary. If anything, the rents under the Underleases were higher than they would have otherwise been, because BT and MLEL believed that they had the benefit of guaranteed longer terms than was in fact the case.

I turn, then, to the issue of whether PW can rely on its service of the Notice in October 2000 to establish unconscionability. PW's case appears to be that it served the Notice on the basis that it would not have to pay the Penalty of just over £5m. (That would not be so if one or two of the Underleases had been determined by BT or MLEL, but, in that case, PW would have been paid a penalty, albeit of a smaller amount). That raises a difficult question, not least in light of the relative dearth of relevant evidence from witnesses. the lack of any direct or expert evidence of the commercial realities, and the briefness of the argument on this issue.

The operation of clause 5(6) with effect from 24th June 2002 was the only opportunity available to PW to determine the Headlease, for which it had no operational use, and which would otherwise run until 2015, at a very substantial rent of at least around £7m per annum. As at the date of the service of the Notice, the aggregate annual rents under the Underleases appear to have been around £2m less than the rent under the Headlease. (This is based on Mr Menzies's evidence that the rent under the BT Underleases was around £22.50 per square foot against £43.50 under the Headlease). However, by March 2001, when the reviewed rents under the BT Underleases were agreed, the gap had become much smaller, around £240,000 per annum. Accordingly, PW was facing a substantial shortfall at the time it served the Notice, but, given that it had until June 2001 to serve the Notice, PW could have waited until the reviewed rent was agreed with BT, and then would have appreciated that the shortfall was much less, albeit still significant, before deciding whether to serve the Notice.

However, PW might well have anticipated that the shortfall could widen over the remainder of the Headlease, although it could also have narrowed, or even have turned into a surplus. Furthermore, PW would have been aware that over half of the premises, four of the seven floors, might well become vacant in 2007, when four of the five BT Underleases were due to expire. Accordingly, at least with effect from about 2007, PW would have appreciated that it might have been facing an increased shortfall, but, even then, there was no guarantee that there would have been any shortfall, and it is even possible that there could have been a surplus. Additionally, because the Headlease contained a tenant's full repairing obligation, and the Underleases contained service charges, PW would have been aware that, if it had not determined the Headlease, it would continue to suffer the cost and bother of management.

The only witness statement which throws light on PW's thought processes and intentions leading up to, and at the time of, the service of the Notice, is that of Mr Menzies, its Estates Manager. He made it clear that PW had always intended to serve the Notice since Summer 2000 at the latest, but that takes matters little further, because that was plainly at a time when PW believed that the Underleases would survive the determination of the Headlease.

Mr Menzies said that Milton Gate's stance in earlier November 2000 caused him "considerable surprise" and that PW had "concern at the possibility of having to pay a large cost which had not been foreseen or budgeted for" and that the Penalty "is not currently budgeted for as [PW] strongly believe[s that it is] not liable to pay [the Penalty]". However, he nowhere stated, or even hinted, that PW would, or even might, not have served the Notice if it had appreciated the true effect on the Underleases of determining the Headlease by the Notice. Nor was there any suggestion of such a factor in the mind of PW in Mr Reynolds's skeleton argument.

PW's primary argument is, I think, that, judging the matter at the date of the service of the Notice, it must be likely that the Notice would not have been served if PW had been aware that the Underleases would, or even would probably, determine on 20 June 2002. On that basis, the Notice would give rise not merely to the liability for the Penalty of over £5m, but also to a potential substantial liability for damages to BT, who at that time were anxious to stay in the premises.

PW would therefore argue that it is pretty clear that it would not have served the Notice if it had not been proceeding, in effect, under the convention. In those circumstances, the consequences of the service of the Notice must, runs the argument, at least as between Milton Gate and PW, be governed by the convention.

If it is wrong to judge the matter by reference to the date of service of the Notice, PW must, I think, accept that any such argument is significantly weaker. As I have mentioned, any risk of liability to BT for substantial damages for derogation from grant had evaporated by the date the Notice took effect. As at 24th June 2002, and as at present, PW's only liability as a result of serving the Notice is for the payment of the Penalty of just over £5m. If, at the date it had served the Notice (or at any date before 24th June 2001, the last moment for the service of the break Notice), PW had appreciated that the service of the Notice would give rise to a liability for £5m, it presumably would still contend that it would not have served the Notice. Although it was suffering a net rental shortfall, it was less than £250,000 per annum. Although there was a danger of further shortfalls accruing (due to differential valuations in relation to the rent review under the Headlease and the Underleases) and, after 2007, much greater uncertainty (because four of the seven Underleases were due for determination), that would have been some way away, and in any event these various uncertainties might not have produced much of a shortfall, and could even have resulted in a profit. Even taking into account the disadvantage of having to manage the premises until 2015, it might be said to be unlikely that PW would have been prepared to pay more than £5m simply to get out of a rather small present disadvantage, and a future potentially much larger disadvantage which might never materialise. In these circumstances, PW would argue, even judging the matter on 24th June 2002 or today, it would be unconscionable for Milton Gate to be permitted to resile from the convention.

On the face of it, there is obvious force in saying that the issue of unconscionability must be judged at the date of the service of the Notice, because, on the present argument, it is that irrevocable act upon which PW relies to found the unconscionability which is an essential ingredient of an estoppel by convention. However, it appears to me that, when considering the question of unconscionability in connection with an estoppel by convention, the court must ultimately carry out its assessment by reference to facts and matters known to it at the date of the hearing. Estoppel is a doctrine designed to do justice, and, at least normally, it seems scarcely consistent with doing justice to ignore facts, which have occurred since the date upon which an action was taken in reliance upon the estoppel, and which may well impinge significantly, or even determinatively, on the issue of unconscionability. It appears to me that it would be an undesirable fetter on the flexibility of the equity, which could lead to injustice rather than justice, if the court, at least in a case such as this, had to ignore the effect of events which occurred after the action taken allegedly in reliance on the convention.

What if the only reason it could be said to be unconscionable for PW to pay the Penalty was that it had exposed itself to a very substantial potential liability to BT and MLEL by serving the Notice without appreciating that it would thereby put itself in derogation from grant under the Underleases? If one judged the question of unconscionability at the date of the service of the Notice, this might well be a powerful argument for holding that PW escaped liability for the Penalty on the basis of estoppel by convention. It appeared at that time that BT would be likely to seek substantial damages if their Underleases were prematurely determined. However, if one moves forward to the date upon which the Penalty would have become payable, 24th June 2002, or the present time, there would be no unconscionability in requiring PW to pay the Penalty. The sole unfairness of Milton Gate insisting on its strict legal rights, namely the possible liability of PW for derogation from grant, would have disappeared. The result of PW's estoppel argument would be a windfall for PW, and a consequent unjustifiable deprivation of Milton Gate's strict legal rights, which seems positively unconscionable.

On this basis, it appears to me that the correct date on which to assess unconscionability is either the date upon which the Penalty would fall due, or the date of the hearing. Of course, different, and indeed possibly difficult, considerations might apply if, shortly after the Notice was served, PW had embarked on a course of action, such as disposing of all its assets, or effecting a substantial investment, in the honest and reasonable belief that the Penalty was not payable. However, there is no suggestion of that having occurred in the present case, and therefore such considerations do not arise. In these circumstances, I think that the correct date for assessing unconscionability is 24th June 2002 or the present day. It could be that the financial effect of having served the Notice is different on those two dates, but there is simply no evidence to suggest that.

I was prepared to accept that the Underleases would not have been granted for terms extending beyond 24th June 2002, without some provision protecting PW from liability to BT and MLEL in the event of the Underleases being prematurely determined, as a result of the service of a Notice under clause 5(6), on the basis of the commercial probabilities, without any specific evidence to that effect. However, I am not prepared to accept that the service of the Notice would not have occurred, if PW had appreciated that it would thereby be liable for the Penalty. Given that clause 5(6) gave only one opportunity to break the Headlease, which would otherwise run to 2015, it appears to me quite possible that PW would have exercised its rights under clause 5(6), even if it had appreciated that it would lead to its having to pay the Penalty. It would have been only too easy for PW's Estate Manager, Mr Menzies or someone else on behalf of PW, to say that PW would not, or even may well not, have served the Notice, if it had appreciated that it would thereby expose itself to the Penalty, and to justify this by expert evidence.

There was only relatively brief discussion as to the burden of proof on this question. It appears to me that, in this connection, it is important to distinguish between reliance and detriment. In Greasley -v- Cooke [1980] 1 WLR 1306, a case of proprietary estoppel, Lord Denning MR said at 1311D to F:

"These statements to Miss Cooke were calculated to influence her - so as to put her mind at rest - so that she should not worry about being turned out. No one can say what she would have done if Kenneth and Hedley had not made those statements. ... There is a presumption that she [stayed on in the house] relying on the assurances given to her by Kenneth and Hedley. The burden is not on her, but on them, to prove that she did not rely on their assurances."

Waller and Dunn LB took the same view as to the onus of proof. see at 1312G, and 1313H to 1314A

The reasoning in Greasley is consistent with what Lord Denning had said earlier in Brikom at [1979] QB 48217- 483A and with the analysis of Balcombe LJ in Wayling -v- Jones [1995] 2 FLR 1029, at 1031G to 1032B. The reasoning was considered and explained by Mr Jonathan Parker QC, sitting as a Deputy Judge, in Coombes -v-Smith [1986] 1 WLR 808 at 821D to E where he said this:

"The decision as it seems to me, was concerned with the presumption of reliance, rather that with the existence of detriment in isolation. The statement of Lord Denning MR... that 'There is no need for her to prove that she acted to her detriment or to her prejudice' must not be taken out of context. Read in context, I take it to mean merely that where, following assurances made by the other party, the claimant has adopted a course of conduct which is prejudicial or otherwise detrimental to her, there is a rebuttable presumption that she adopted that course of conduct in reliance on the assurances."

That analysis is consistent with the reasoning of the Court of Appeal in Stevens and Cutting Ltd. -v- Anderson [1990] 1 EGLR 95. Stuart-Smith LJ accepted that Greasley was authority for the proposition that:

"If... there was a clear representation or promise... and... the appellant had suffered detriment by adopting a course of conduct consistent with the assumption that the representation was true or the promise binding, I would be disposed to conclude that in the absence of evidence to the contrary the appellant had adopted that course in reliance on the representation or promise" (see at 97M).

However, he went on to indicate that he was not prepared to assume, in the absence of any specific evidence to that effect, that any detriment had been suffered (see at 98A to C). The same view was taken by Farquharson LJ, who, at 99F, said that "[t]he difficulty" for the claimant was "that it called no evidence of detriment".

I do not know whether, as at October 2000 (when the Notice was served), June 2001 (the last month for serving the Notice), June 2002 (when the Notice took effect, and the Penalty became, or would have become, payable), or today, PW was or is better off or worse off if it is free of the Headlease but liable for the Penalty. On any view, it has rid itself of a tenancy of property it does not want, which would cost it about £7m or more a year until 2015. It is true that there was only a relatively small (£245,00 per annum) current rental shortfall, but there was obviously a real risk that it could well increase, possibly substantially, from 2006 or 2007. There was also the disadvantage to PW of having to manage the premises until 2015. The financial implications of that risk and that disadvantage are a matter for an expert valuation surveyor and depend very much on the state of the market and market perception. I simply do not know whether, judged from an objective market-value basis, it would be worth paying the Penalty to be shot of the Lease. Nor do I know the answer to that question, if judged from the subjective viewpoint of PW.

It seems pretty likely that the Headlease had and would have a significant negative value, in light of the features I have identified. That view is supported by the very fact that PW served the Notice. That it may not have expected to pay the Penalty does not impinge on the point that, if the Headlease had any significant value, it appears unlikely that PW would have determined it. PW would presumably have assigned it. I, have no basis upon which I can even hazard an informed guess at the likely value to PW of having determined the Headlease.

PW has the benefit of having exercised its rights under clause 5(6) against Milton Gate. PW wishes to avoid the concomitant burden under the clause on the ground of an estoppel by convention. This ground involves establishing unconscionability, which in a case such as this means relevant detriment to PW. As a matter of principle, it seems to me that the burden should be on PW to establish that detriment. That view is supported by the cases to which I have just referred. It is also consistent with the fact that it appears likely, indeed, I think very likely, that the determination of the Headlease was and is of substantial value to PW.

In all these circumstances, I do not consider that PW has made out its case on estoppel by convention, subject to considering the issue on a tripartite basis. However, that is not the end of the issues as between PW and Milton Gate, because PW has a further argument based on the doctrine of approbation and reprobation.

PW's case in this connection is that it is not open to Milton Gate to contend as against PW that the Underleases have determined (and therefore Milton Gate is entitled to recover the Penalty) in circumstances where Milton Gate are contending, and have contended since 4th July 2002, as against BT, that the Underleases are continuing. The technical expression upon which PW relies is that a person cannot approbate and reprobate; the expression used frequently by lawyers is that one cannot blow hot and cold; less technical versions are that one cannot have one's cake and eat it, and that one cannot have the penny and the bun.

In my view, that argument cannot succeed in the present case. It seems to me that there is normally nothing to prevent someone running two inconsistent arguments (whether both arguments are against the same person or each argument is against different persons), and at some point opting for one argument rather than the other. It would be pretty surprising if it were otherwise. If A runs two inconsistent arguments against B, it cannot be right that simply because of that fact, B can either choose which of the two arguments to accept, or defeat both of those arguments. Equally, if A runs two inconsistent arguments, one against C and the other against D, there would seem no warrant for concluding that C and D could get together to decide which of the two arguments should succeed; equally, it would seem ridiculous if C could defeat A because A had run an inconsistent argument against D, and D, could defeat A because A had run an inconsistent argument against C.

Of course, different considerations may apply where A has achieved success on one of the arguments. In that event, it may very well be right that A will have disentitled himself from seeking to run the other argument. However, in such a case, there would rarely be a need to rely on approbation and reprobation, because A would probably be prevented from relying on his alternative argument on the basis of election or estoppel. Of course, in some cases it may not be inequitable or otherwise inappropriate to permit A to run the other argument, albeit subject to some qualification.

In the present case, I can see no basis, either in fairness or in legal principle, for holding that Milton Gate should be disentitled from recovering the Penalty from PW, simply because it has contended as against BT since about 4th July 2002 that the Underleases have been continuing. If Milton Gate establishes, or BT accept, that the Underleases are to be treated as continuing, then that would prevent Milton Gate contending for the opposite conclusion as against PW. However, there has been no determination or acceptance of Milton Gate's argument as against BT that the Underleases are continuing. In those circumstances, there is no question of Milton Gate having its cake and eating it. While Milton Gate is trying to have its cake (by claiming against BT that the Underleases are continuing) there is a risk that it will not be able to eat the cake (in the form of recovering the Penalty from PW), but I do not see why it should be prevented from pursuing inconsistent remedies. Even if one regards Milton Gate's negotiating stance between November 2000 and July 2002 as pretty tough, there was nothing dishonest about it, and there was nothing even unusual about its taking inconsistent positions in an open way in the present proceedings.

Mr Reynolds cites four authorities to support his argument on approbation and reprobation. The first is in Lissenden -v- CAV Bosch Limited [1940] AC 412 at 418 to 419, where Viscount Maugham explained that the doctrine was Scottish in origin and that it related to inconsistent claims under wills and instruments inter vivos. Essentially, it appears to be a species of election. In that case, the fact that a successful claimant received instalments of an award did not prevent him subsequently appealing that award on the basis that it was too small. The unsuccessful claim based on approbation and reprobation was stronger than PW's, because the claimant had received some benefit under the award which he was seeking to impugn.

Express Newspapers plc -v- News (UK) Limited [1990] 1 WLR 1320, involved two copyright disputes between two newspapers. In the first dispute, the Daily Express alleged that Today had infringed its copyright in one article, and in the second, Today alleged that the Daily Express had infringed its copyright in another article. The Daily Express obtained summary judgment in its action. In Today's application for summary judgment in its action, the Daily Express ran an argument, which, if correct, would have meant that Today had a defence in the Daily Express's action, and that the Daily Express should accordingly not have obtained summary judgment.

Sir Nicolas Browne-Wilkinson V-C accepted that Today was entitled to judgment in its action, saying at 1329:

"There is a principle of law of general application that it is not possible to approbate and reprobate. That means that you are not allowed to blow hot and cold in the attitude that you adopt. A man cannot adopt two inconsistent attitudes towards another: he must elect between them and, having elected to adopt one stance, cannot thereafter be permitted to go back and adopt an inconsistent stance."

I consider that the Vice-Chancellor was saying that, having obtained a benefit against Today on the basis that a particular argument was wrong, it was not open to the Daily Express then to deny Today a similar benefit on the basis that the argument was right. By contrast, in the present case, Milton Gate have not obtained judgment against BT to the effect that the Underleases are continuing: it would only be if they had obtained such judgment that the reasoning in Express Newspapers would apply. Even then, there is the difference that in this case the inconsistent cases were advanced against different persons.

In Oliver Ashworth (Holdings) Limited -v- Ballard (Kent) Limited [2000] Ch. 12 at 31, Robert Walker LJ considered that in some circumstances the law could well deprive a person of the right to raise inconsistent arguments in proceedings, in certain circumstances, even when no election had occurred, on the basis of approbation and reprobation. However, as he appears to have acknowledged in his observations on the point, in most cases the issue would be determined by reference to the doctrine of election.

Robert Walker LJ's observations in that case were tentative and obiter. In any event, his reasoning does not apply here. On the basis that a notice to quit was defective, a landlord in that case had issued a claim for rent, but when it subsequently appeared that the notice was probably valid, the landlord sought to amend his pleadings to claim double rent on the basis that the tenant had been a trespasser. The point made by Robert Walker LJ at 31 was that it was "at least arguable that by demanding and suing for rent the landlord was unequivocally treating the tenant as not being a trespasser", and that, accordingly, any subsequent amendment of the landlord's pleading to claim that the tenant was not a tenant should not operate retrospectively. In my judgment, it does not follow from this that, in the present case, Milton Gate was disentitled from alleging, on 23rd July 2002, in its Defence against PW, that the Underleases had expired, merely because, as against BT, Milton Gate had been maintaining the opposite from 4th July 2002.

Mr Reynolds also refers to the decision of the Court of Appeal in Public Trustee -v-Pearlberg [1940] 2 KB 1. The effect of that decision was that, so long as a party had live proceedings seeking specific performance of a contract. he could not seek to rescind the contract. There is nothing in the reasoning in Pearlberg to prevent a party putting forward alternative cases, particularly when having to face two different parties with conflicting positions. Indeed, if anything, Pearlberg is consistent with Milton Gate's case, as it is implicit in the reasoning that, once the specific performance action had been discontinued, rescission could then be sought.

Quite apart from this, the sequence of events does not help PW's case. From 2000, Milton Gate argued against PW and BT that the Underleases had expired: that was its position until early July 2002. Further Milton Gate pleaded its case against PW (that the Underleases had expired) before it pleaded its case against BT (that they had not expired). Therefore, it would appear that, if there is anything in this argument, it would assist BT, and not PW.

I should mention one further point in relation to this aspect. It might be said that Milton Gate did in fact receive a benefit from alleging against BT that two of the Underleases were continuing, because it was only on this basis that Milton Gate was entitled to the rent in respect of the upper two floors. However, it is common ground between Milton Gate and BT that the basis upon which BT paid the rent in respect of the upper floors to Milton Gate from 24th June 2002 would entitle BT to repayment of the rent if, as between Milton Gate and BT, the Underleases were determined on 20 June 2002.

Reconciling the estoppel as between Milton Gate, PW and BT

It appears to me that the effect of the discussion in the two immediately preceding sections of this judgment is as follows. As between Milton Gate and BT, there was a convention, but it had been mutually resiled from, and, in those circumstances, were it not for the involvement of PW in the convention, there would be no estoppel by convention, and BT would be entitled to insist on their strict legal rights, and to contend that the Underleases expired on 24th June 2002. As between Milton Gate and PW, although there was a convention which could have given rise to an estoppel, PW is unable to establish the essential ingredient of unconscionability to make good its estoppel, because it has not established that it would suffer relevant detriment if the strict legal position obtains. In those circumstances, at least at first sight, it would seem that there is no good reason why the strict legal position should not be enforced as between all the parties.

In my judgment, it seems to me that the issue is not, at least necessarily, quite as simple as that. First, it is appropriate to consider the one aspect of the tripartite relationship not so far discussed in this context, namely that between PW and BT. Secondly, even after considering that aspect, it is appropriate to consider the overall justice of the situation if the strict legal position obtains. After all, as has been emphasised in a number of cases, estoppel is a flexible doctrine to be applied to arrive at a just result, albeit on a principled basis. I turn to consider those two questions.

PW's contention, that BT is estopped by convention from contending, as against PW, that the Underleases expired on 24th June 2002, appears to me to suffer from the same problem as PW's similar contention as against Milton Gate. - While able to establish a convention to that effect, PW's problem is establishing any unconscionability in BT resiling from that contention. My reasons for that conclusion are the same as my reasons for concluding that PW cannot raise an estoppel by convention as against Milton Gate in this connection.

I turn, then, to consideration of the tripartite relationship on a broader basis. Given that (a) BT and Milton Gate effectively agreed a mutual resiling from the convention in and from April 2002, (b) there is no question of BT claiming damages against PW for derogation from grant due to the premature determination of the Underleases, and (c) PW cannot establish that, by serving the Notice, and thereby putting an end to the Headlease but becoming liable to the Penalty, it is worse off than if it had not served the Notice, there is, as I have said, at least on the face of it, no unconscionability, as between any of the parties, if the strict legal position applies. In those circumstances, I think it would require an exceptional case before the court should be prepared to conclude that the strict legal position does not apply as between each of the parties.

It appears to me that the only possible oddity in this conclusion is that PW has become liable for the Penalty essentially as a result of a mutual resiling from a tripartite convention by the two other parties to the convention, in circumstances over which PW had no control, and indeed of which it was unaware. This fact might, of itself, be said to give rise to a separate, rather more subtle, argument of unconscionability in favour of PW.

However, I do not think that that is a good argument. First, it appears to me that, where there is a tripartite convention which could give rise to an estoppel, each of the three parties to the convention runs the risk that an arrangement between two of them may result in it not being unconscionable for those parties being able to resile from the convention as against the third party. Of course, each case will turn on its own particular facts, but such a possibility is inherent in any tripartite convention. Secondly, on the facts of this case, the convention is and was primarily between Milton Gate and PW, and BT were only brought into the convention as a privy of PW, and in order to make the convention work in a fair and equitable way, as explained above.

Indeed, this leads to a wider point. It seems to me that there is a powerful argument for saying that, even if there had been no mutual resiling from the convention by Milton Gate and BT in and after April 2002, it would probably have been right to conclude that, as BT did not want the Underleases to continue, there should be no estoppel by convention as between the parties in any event, and each of them should be entitled to insist on the strict legal position. Given that I have rejected the argument that PW could rely on the service of the Notice in October 2000 to raise an argument based on unconscionability, the only ground upon which it can raise a claim for unconscionability is, in effect, its potential liability to BT for derogation from grant due to the premature determination of the Underleases. By the same token, the only basis upon which Milton Gate could run an argument based on estoppel by convention would be that it would be wrong if the convention enabled PW to prevent it from recovering the Penalty, unless it also enabled Milton Gate to insist upon the Underleases continuing until their respective contractual expiry dates. In other words, Milton Gate's estoppel argument against BT is based on the contention that Milton Gate should not have the burden of the estoppel as against PW without having the benefit of the estoppel as against BT. Neither Milton Gate nor PW appears to me to have any other basis for raising the estoppel argument.

Given that the convention is primarily between Milton Gate and PW, and that BT's subjection to the convention arises from the need to make the convention 'work fairly, it seems to me that if BT are able, even unilaterally, to arrange matters so as to avoid the damage on which PW could otherwise justifiably rely to give rise to the estoppel, there would be no injustice in the estoppel argument falling away. In other words, by making it clear that they would not claim, or by effectively disentitling themselves from claiming, damages for premature determination of the Underleases, it seems to me that BT could thereby put an end to the ability of PW, and therefore of Milton Gate, to contend that the convention gives rise to an estoppel. The fact that BT cannot claim such damages from PW deprives PW of any argument that it would be unconscionable for Milton Gate to enforce the estoppel (because the only other ground upon which PW could rely, namely service of the Notice itself, fails on the facts). As a result, Milton Gate would be entitled to. recover the Penalty from PW, and, in those circumstances, far from it being equitable, it would be positively unconscionable if Milton Gate could insist on the Underleases continuing beyond 24th June 2002, as against BT.

On the particular facts of this case, there is, nonetheless, some ground for sympathy with PW's argument, in light of Milton Gate's conduct. Milton Gate acquired the reversion to the Headlease knowing all the relevant facts, in particular the effect of clause 5(6) of the Headlease, the terms of the Licences, and the fact that the Underleases were for terms extending well beyond the break date under clause 5(6) of the Headlease. It also knew that the effect of the service of a Notice under clause 5(6) of the Headlease would lead to the determination of the Underleases. If (as seems unlikely) it had not appreciated that PW and BT had a very different view as to the effect of the determination of the Headlease by a Notice on the Underleases, I consider that the position of PW and BT was made quite clear to Milton Gate by May 2001.

It is true that the case based on estoppel by convention was not spelt out in terms to Milton Gate until after June 2002, but the facts giving rise to the estoppel were plain to see, and would have been obvious to reasonably sophisticated commercial property lawyers, such as Milton Gate's solicitors, and, indeed, to sophisticated commercial property advisers, such as Milton Gate apparently employed. It was apparent from clause 5(6) that the parties to the Headlease anticipated that Underleases would survive the service of the Notice. Further, the fact that the Underleases, all extended well beyond the break date under clause 5(6), showed that they were entered into on the basis that they would continue beyond the break date. Bearing in mind that Milton Gate and its solicitors appreciated the legal position, they must also have realised that the parties to the Headlease, the Licences and the Underleases were under a misapprehension as to the law. Pennell had been decided in December 1994, and Barrett had been decided in January 2000. Milton Gate were being advised by their solicitors by March 2000, and had appreciated the fact that a subtenancy could not survive the determination of a head-tenancy by notice served in accordance with its terms. Further, the first time the dispute was discussed in detail between Milton Gate and PW, on 15th November 2000, it was made clear in terms to Milton Gate that PW's case was that the application of the general rule in the present case would be an "unjust conclusion": BT were clearly subscribing to this view in their discussions with Milton Gate, until their change of heart in April 2002.

It was in those circumstances that Milton Gate chose in early 2001 to state in unequivocal terms to BT that the Underleases would determine on 24th June 2002, and it maintained this position to BT in unequivocal terms until after 24th June 2002. Milton Gate was not, at any time, under any misapprehension as to the law or the facts, and it appreciated the position of BT and PW. It decided to take advantage of the situation, believing that it could drive a hard bargain with BT, who, it thought (correctly), had been unexpectedly and inconveniently thrown back on their potentially precarious and uncertain rights under the 1954 Act, rather than being able to look forward to a continuing substantial contractual term.

Milton Gate attempted to get the best of both worlds. It sought to achieve a situation which (in fact if not in law) would satisfy the proviso to clause 5(6) by negotiating the overriding lease with BT. Indeed, in light of its higher rent and longer term than the combined rents and average terms of the Underleases, the overriding lease would have represented a substantial improvement for Milton Gate. At the same time, Milton Gate was refusing a surrender of the Headlease (which was consistently on offer from PW from November 2002 until May, or even June, 2002), so that it could recover the Penalty from PW on the basis that the proviso to clause 5(6) had not been satisfied.

Even though it can therefore be said that Milton Gate might have brought such an outcome on itself, the fact remains that it would be unfair on Milton Gate if it neither had its cake nor could cat it, in the sense that it could neither enforce the continuation of the Underleases against BT, nor recover the Penalty from PW. Although it negotiated toughly, it was not in any way dishonest, and, although it got close, it did not receive a tangible benefit from its tough negotiations. To let PW off the Penalty, if it is not liable to BT for damages, would seem something of a windfall. It seems likely that PW has obtained a benefit of substantial financial value, although it is not possible to quantify that value in light of the absence of any relevant evidence, by determining the Headlease. Additionally, although it is a matter of speculation, it is a fair inference from the evidence that a higher rent was paid by BT (and presumably by MLEL) under the Underleases at the time of their grant, because of their expected duration, than would have been paid if it had been appreciated that the Underleases would be determined by the service of the Notice under clause 5(6). PW has therefore probably obtained a hidden benefit from the common misunderstanding. Additionally, given that Barrett had been decided some 9 months before the Notice was served, PW had the opportunity of appreciating the risks of serving the Notice.

In all these circumstances, I am of the view that it would not be unconscionable if the strict legal position applied as between all three parties in the present case. If I am wrong, and there is in fact an estoppel between Milton Gate and PW, then it would be necessary to consider whether, in those circumstances, Milton Gate's activities after acquiring the freehold in June 2000 have disentitled it from insisting on the convention as against BT. On the one hand, there would be the obvious injustice to Milton Gate of having the burden of the estoppel by convention against PW, but not having the benefit of that estoppel as against BT. On the other hand, there is the fact that there was a clear and mutual resiling from the convention as between Milton Gate and BT, and, as the facts outlined above indicate, Milton Gate can be said only to have themselves to blame for this outcome. Having appreciated that it was in a highly advantageous position because of the misunderstanding on the part of BT and PW as to the strict legal position, Milton Gate entered into negotiations with BT for the grant of the overriding lease and did not accept PW's proposal of a surrender of the Headlease.

Milton Gate pushed its advantage as hard as it could, perfectly properly. Indeed, its stance almost came off. the overriding lease for a longer term, at. an increased rent, was on the point of being executed. However, at the last minute, BT had a change of mind, and went along with what Milton Gate had been unequivocally saying and relying on for more than a year. Even after its attempt failed to get the best of both worlds, Milton Gate could have tied BT into the Underleases by accepting a surrender of the Headlease, which was plainly on offer from PW.

Although I accept that BT have not identified any act which they took in reliance of there having been a mutual resiling from the convention, I do not consider that it would be inequitable or unconscionable for BT to be permitted to rely on the mutual resiling from the convention, even if PW was entitled to raise an estoppel on that convention against Milton Gate. While a sword is normally invoked in a rather different context in connection with estoppel, it seems to me that, having chosen to live by the sword of insisting on the strict legal position as against BT, and seeking to have it both ways (i.e. the benefit of the reversion to the overriding lease, and payment of the Penalty from PW), Milton Gate could scarcely have complained if it were unable to raise the shield of unconscionability to avoid dying by the sword of suffering both ways. However, that point does not, in my opinion, arise.

Conclusion

In these circumstances, my conclusions are as follows:

i) By clause 5(6) of the Headlease, CLRP and PW purported to contract out of the general rule that a subtenancy comes to an end, on the determination of the head-tenancy in accordance with its terms;

ii) However, as a matter of law, it is not possible to contract out of the general rule, and therefore the Underleases expired on 24th June 2002 (subject to the other arguments);

iii) If the position were otherwise, then, in a case such as this, where the determination of the head-tenancy is by a break notice, the covenants in the subtenancy would have been unenforceable between the landlord and the subtenant, but that position has now been reversed as a result of section 3 of the 1998 Act;

iv) Accordingly, if my conclusion in (ii) is wrong, the covenants in the Underleases would be enforceable as between Milton Gate and BT;

v) A break clause such as clause 5(6) cannot be recharacterised as a surrender provision, and, even if it can, it would not alter conclusion (ii);

vi) There can be no estoppel by deed in the present case;

vii) There can be no estoppel by convention in the present case as a result of CLRP and PW having entered into the Headlease itself,

viii) However, a convention, sufficient in principle to give rise to an estoppel, did arise between the Landlord and PW as a result of the grants of the Underleases and the associated Licences;

ix) BT and MLEL, as the tenants under the Underleases, and as parties to the Licences, became bound by the convention;

x) Although aware of the true position at the time it acquired the reversion to the Headlease, Milton Gate was bound by, and entitled to the benefit of, the convention;

xi) As between Milton Gate and BT, there was a mutual resiling from the convention;

xii) As between Milton Gate and PW, the convention could not give rise to an estoppel if PW was not, in the event, exposed to a claim for derogation from grant by BT;

xiii) Both because of the immediately preceding two conclusions., and, in any event, because BT did not want the Underleases to continue beyond 20 June 2002 and can claim no damages in respect of their premature determination, there is no estoppel by convention operable as between any of the parties;

xiv) In these circumstances, as between all three parties, the Underleases expired on 24th June 2002, and PW became liable for the payment of the Penalty to Milton Gate as at 20 June 2002.