Bath Rugby Ltd v Greenwood & Ors (pre-1926 restrictive covenants affecting land)

This is my judgment on the trial of a claim brought under CPR Part 8, for declaratory relief pursuant to section 84(2) of the Law of Property Act 1925. The declaration sought is that a restrictive covenant contained in a conveyance made in 1922 of certain land in the centre of Bath, formerly part of the Bathwick Estate, is unenforceable and not binding on the claimant. The claimant owns a long lease of the land in question, and is preparing to develop it. The claim form was issued on 6 June 2019 at the Rolls Building, Royal Courts of Justice, in London. The claim form was accompanied by details of claim and a witness statement made by Caroline Preist, the claimant’s solicitor, exhibiting certain key documents.

By order of Master Shuman dated 9 October 2019, the claim was transferred to the High Court, Business and Property Courts in Bristol. Her order also required the claimant to disclose certain documents and information, the fruits of research carried out by the claimant’s solicitors and which had been referred to in a circular letter to local residents dated 1 December 2018. This was done in a further witness statement made by Ms Preist dated 23 October 2019, together with exhibited documents. These documents demonstrated that the properties of some of the defendants, at least, had not been retained in the Bathwick Estate at the time of the 1922 Conveyance. Those defendants thereafter ceased to take an active part in these proceedings.

After the transfer to Bristol, I made an order on 7 January 2020, joining the seventh and eighth defendants, whose property (it was common ground) had been retained by the Bathwick Estate in 1922. I also gave directions for service upon them, and for evidence to be filed by them and in response by the claimant. Further witness statements were then made. One was made on 17 January 2020, by Godfrey White, the eighth defendant, and a director of the seventh. Mr White co-owns the long lease to a flat at 77 Great Pulteney St, Bath, and also a share in the company which owns the freehold of no 77. He is also a director of this company, which has since become the seventh defendant. Another witness statement was made on 30 January 2020 by Ms Preist, in response to that of Mr White. On 28 August 2020 the eighth defendant made a further, short witness statement. The first to sixth defendants have played no effective part in the proceedings since the seventh and eight defendants were joined. The seventh and eight defendants have accordingly borne the burden of defending this claim. But I make clear that the claimant has brought the claim, and it is the claimant that must prove it.

The land concerned in the 1922 conveyance consists of open land in the centre of Bath on the eastern side of the River Avon, in that part of the city known as Bathwick. It has long been used for the playing of sports, especially rugby football, and is popularly known simply as “the Rec” (as I will term it in this judgment). The westernmost part of this land, adjacent to the river, is occupied by a stadium, and leased to the claimant, which uses it for the playing of rugby matches. Over the last

This case is accordingly about the continued enforceability in law of that covenant in the 1922 conveyance. This is a very narrow, legal issue. I should say it is also a very technical one, perhaps one of the most technical in what is already a technical area of the law, and made more rarified by the enormous changes made to English property law in 1925 (which, at least so far as this case is concerned, were not retrospective). I am asked to decide only whether the covenant continues to be enforceable by anyone.

Because of the interest generated by this dispute, I should add that it is important to understand what this case is not about. I am not asked to say whether the redevelopment of the stadium would be a breach of the 1922 covenant if it were enforceable, much less what would be the remedy if any such breach were established. Nor am I asked to decide whether the covenant, if otherwise enforceable, should be modified or discharged. Nor am I concerned in this case with planning, environmental or other public law questions. Those are all potentially at least matters of law, and they may arise in proceedings elsewhere, but they are not before me today. Lastly, I am most certainly not concerned with whether the development project is a ‘good idea’, morally sound, of social benefit or aesthetically pleasing, or whether it would be better to maintain the status quo. I am a lawyer, whose function in this case is to decide a certain question of private right arising between private persons, and not a politician or moral philosopher. My role is limited accordingly.

I have already referred to the written witness statements filed and served in this case, exhibiting numerous relevant documents. Because this is a claim under CPR Part 8, where there are not expected to be significant disputes of fact, oral evidence is not always needed. In the present case, no witnesses were called to be cross-examined on their written evidence, and it is not necessary for me to say anything further about the witnesses themselves. On the basis of the material before the court, I find the following facts.

I begin with the Bathwick Estate, of which the Rec historically formed part. What I say is derived largely from the research carried out for this case and put in evidence before me in behalf of the claimant, but the defendants’ documents have contributed too. The story begins with Sir William Pulteney (1624-91). He had three children: William, who became a colonel in the army and died in 1715, John, who became an MP, and Anne, who married Charles Fitzroy, the illegitimate son of Charles II by Barbara Villiers, the first Duchess of Cleveland. None of these three siblings had any part to play in the history of what became the Bathwick Estate. But the issue of all three of them did.

Colonel William’s son, Sir William’s grandson, also called William (born 1684) became an MP in 1705. In 1727 he bought the Manor of Bathwick (amongst others) from the Earl of Essex. At that time Bathwick was a country village, separated from Bath by the (then unbridged) River Avon. William Pulteney MP was a man of both wealth and political influence. He held a number of public offices during his life, and, in 1742, he was created Earl of Bath. (He was even offered the opportunity by King George II to form an administration in 1746, but this came to nothing.) He saw the possibility of developing the Manor of Bathwick into an affluent suburb of Bath, but although he started the process he did not live to see it completed. He died in 1764, his own children having died in his lifetime without issue. His landed estates (including the Manor) passed to his brother Harry, a retired army general.

Harry himself died unmarried and childless in 1767, and the estates passed to trustees for Frances Johnston, the daughter of William’s and Harry’s first cousin Daniel, the son of John Pulteney MP, the second son of Sir William. Frances was married in 1760 to Scottish lawyer Sir William Johnston, who changed his name to Pulteney on the inheritance. (It is said that he became known locally as “Mr Pulteney”.) Frances died in 1782, and the real property descended to her daughter Henrietta Laura, who (because of her father’s political influence) was created Baroness of Bath in 1792 and Countess of Bath in 1803. She married Sir William Murray in 1794. She and her father were instrumental in developing the Bathwick Estate, largely by laying out streets and granting building leases of 99 years on plots adjoining those streets. They were also involved in the promotion and construction of Pulteney Bridge, across the River Avon, so as to connect the centre of Bath with the Estate, and to make the building plots more attractive.

When the Countess died without issue in 1808, her real estate passed to her third cousin, William Henry Vane, the third Earl of Darlington. He was descended not only from Anne Pulteney, the daughter and third child of Sir William, but also – through the female line – from Charles Fitzroy, second Duke of Cleveland, whom Anne had married, but whose title had become extinct on the death of their son without male heirs. In due course, and in recognition of this family connection, the Earl was created Marquess of Cleveland in 1827 and Duke of Cleveland in 1833. He died in 1842, leaving eight children. These included three sons, none of whom left issue. These were the second Duke (who died in 1864), the third Duke (who died later the same year) and the fourth Duke (who died in 1891, when all the major titles became extinct). The estates descended to each of them in turn.

The first Duke’s five daughters included Lady Louisa Vane, who married Major Francis Forester. Their son, Henry William Forester, in turn had a son, Francis William, born in 1860. He became a captain in the army. No copy of the will of the fourth Duke was available to the court (although once probated it is a public document). However, there is a summary of the relevant parts of that will in the recitals to a resettlement dated 19 July 1920, and a copy of that was in the bundle at trial. From this it is seen that the will created a strict settlement of the Bathwick Estate under the Settled Land Acts 1882 to 1890, of which the duke’s great-nephew Captain FW Forester was tenant for life, with remainder over in tail male. On 18 July 1920 Captain Forester’s only son Henry William (tenant in tail in remainder) attained the age of 21 years. On the next day father and son executed both a disentailing deed and a resettlement of the strict settlement created by the late Duke of Cleveland’s will.

Until the Settled Land Act 1925 was passed, the tenant for life of a settlement governed by the settled land legislation did not become the legal owner of the land concerned. The legal estate was vested in the trustees of the settlement. However, the pre-1926 tenant for life did have powers of sale and leasing, and therefore was in a position to direct the sale or leasing of land comprised in the Bathwick Estate: see the Settled Land Act 1882, sections 3, 6. In 1896 the Rec was leased to trustees of the Bath [Rugby] Football Club. A further lease was granted in 1908, for 21 years. Captain Forester was in fact President of the Club from 1898 to 1926, even though he did not live in Bath, but in Leicestershire.

Thus the Bathwick Estate has passed through all three of the lines of descent from the original Sir William Pulteney, though neither he nor his own children ever owned any of it. In 1919, by direction of Captain Forester, as tenant for life, the whole of the Bathwick Estate was put up for sale by auction as a single lot (which would have included the Rec), but it failed to sell. The auction particulars (which were in evidence) set out the details of the individual properties in some detail. In 1921, Captain Forester directed a further sale by auction, this time of only a part of the estate, and in a series of lots (but not including the Rec). Again, the auction particulars set out details of the properties concerned. Some of the lots sold at the auction, and some did not. Some lots were withdrawn and sold separately. Others were sold privately after the auction. Many remained unsold. Then, in 1922, Captain Forester sold the Rec for £6050 to a company formed for the purpose, The Bath and County Recreation Ground Company Limited, subject to the lease of 1908.

The conveyance of 6 April 1922 is relatively short, and, in order to be properly construed, needs to be set out in full, with the sole exception of the schedule, which is irrelevant to the question of construction, simply giving details of various deeds of title of which the purchaser was entitled to require production. It reads as follows:

“This Indenture made the Sixth day of April, One thousand nine hundred and twenty two Between Frances William Forester of Saxelbye Park, Melton

The Vendor in exercise of the power for this purpose conferred by the Settled Land Acts 1882 to 1890 and of every other power enabling him and as beneficial owner hereby conveys unto the purchasers All that piece or parcel of ground situate in the City of Bath and containing an area of Sixteen acres two roods and eleven perches or thereabouts and known as The Bath and County Recreation Ground Together with the building erected thereon near the North Parade Road formerly used as a Skating Rink and now in the occupation of Artcraft Ltd and The Pavilion near to the Pulteney Mews now the occupation of the Purchasers as Lessees thereof under an Indenture of Lease dated the Twenty fifth day of March One thousand nine hundred and eight and made between the vendor of of the one part and Charles Henry Simpson and others of the other part and also the two buildings formerly used as two cottages adjoining and on the south side of Pulteney Mews now in the occupation of the Purchasers and The Bath and County Croquet Club respectively Except and reserving unto the vendor and his successors in title and his and their heirs and assigns the free and uninterrupted passage and running water and soil from the other buildings and land of the vendor and his tenants adjoining or near to the said hereditaments hereinbefore described through the sewers drains and water courses which are now or may hereafter be in or under the said premises To hold unto and to the use of the Purchasers their successors and assigns in fee simple discharged from all the limitations trusts powers and provisions of the said Settlement and from all estates interests and charges subsisting or to arise thereunder Subject to and with the benefit of an Indenture of Lease dated the Twenty fifth day of March One thousand nine hundred and eight and made between the Vendor of the one part and Charles Henry Simpson, James Edward Henshaw, Egbert Lewis, Alfred George Derwent Moger, William Morgan, and William Frederick Cooling of the other part Whereby the said hereditaments hereinbefore described were demised for a term of Twenty one years from the Twenty fifth day of March One thousand nine hundred and eight at the yearly rent of One hundred pounds

The Purchasers for themselves their successors and assigns hereby covenant with the Vendor his successors in title and assigns and to the intent and so that this covenant shall run with and be binding on such portions of the hereditaments and premises hereby conveyed as are respectively affected thereby into whosesoever hands the same may come but so that the Purchasers shall not be personally liable in damages for any breach thereof after they shall have parted with the same hereditaments and premises that no workshops warehouses factories or other buildings for the purpose of any trade or business which may be or grow to be a nuisance and annoyance or disturbance or otherwise prejudicially affect the adjoining premises or the neighbourhood shall at any time hereinafter be erected upon the said hereditaments and premises except the part thereof now in the occupation of Artcraft Limited and that nothing shall be hereafter erected placed built or done upon the said hereditaments and premises including such part thereof as last aforesaid which may be or grow to be a nuisance and annoyance or disturbance or otherwise prejudicially affect the adjoining premises or the neighbourhood Provided always that no factory chimney shall be erected on the portion of the said hereditaments now in the occupation of Artcraft Limited

Provided always that so far as regards the reversion or remainder expectant on the life estate of the Vendor in the premises hereby conveyed and the title thereto and further assurance thereof after his death the statutory covenant by him implied in these presents shall not extend to the acts or defaults of any person other than and besides himself and persons deriving title under him

The Vendor hereby acknowledges the right of the Purchasers to production of the documents mentioned in the Schedule hereto and delivery of copies thereof and hereby undertakes for the safe custody thereof”

In 1924, the Bathwick Estate Company was incorporated for the purpose of taking over the parts of the Bathwick Estate still remaining in the family settlement. Those parts, which included the property at 77 Great Pulteney Street now owned by the eighth defendant, were valued by one Joseph Stower, and they were the subject of an agreement for sale dated 4 September and made between Captain Forester and the Bathwick Estate Company. The purchase price of £187,492 was satisfied by the allotment of shares in the company to the trustees of the settlement. In the years following, conveyances of various properties on the Estate were made to third parties by Captain Forester and the company acting jointly. Captain Forester died in 1942.

On I February 1956, The Bath and County Recreation Ground Company Limited conveyed the Rec to the Corporation of the City of Bath (“the Corporation”) in consideration of the sum of £11,155, to hold on certain trusts, but subject to the covenants in the 1922 conveyance, if subsisting. In April 1974, the new local authority, Bath City Council, succeeded to the assets and liability of the Corporation, by virtue of the Local Government Act 1972. In April 1996, the new Bath and North East Somerset District Council succeeded similarly, by virtue of the Local Government Act 1996.

In Bath & North East Somerset Council v AG [2002] EWHC 1623 (Ch), Hart J held that the trusts on which the new Council held the Rec were valid recreational charity trusts. The judge said:

“48.

… In the result I have been finally, and narrowly, persuaded that the public character of the Corporation and the fact that it was intended to be the trustee in perpetuity enables one to conclude that the dominant intention of the trusts, to which all the express provisions should be regarded as ancillary, was to provide a recreational facility for the public, and that, construed as such, the trusts are valid charitable trusts.”

In February 1974 the Bathwick Estate Company resolved to go into members’ voluntary liquidation. The company thereafter sold and transferred some of its assets to third parties (including 77 Great Pulteney Street, which was sold to the Corporation). The company was dissolved in 1975, when assets were transferred out to various family trusts connected with the members. In 1995, Bath City Council granted a lease for 75 years from 10 October 1994 of the western part of the Rec to the trustees of Bath Football Club. This lease replaced an earlier lease. It was registered on 8 February 1996, and the register stated that it was subject to the covenants in the 1922 conveyance. On 22 January 2009, the fee simple estate was also registered (in the name of Bath & North East Somerset District Council), and similarly stated to be subject to the covenants in the 1922 conveyance. The residue of the lease was assigned to the claimant, who was registered as proprietor on 17 February 2014. The fee simple estate may have passed through the hands of the Official Custodian for Charities (this is not clear), but was subsequently transferred to Bath Recreation Limited, which was registered as proprietor on 25 January 2018.

No 77 Great Pulteney Street was sold in 1983 by Bath City Council to a Mrs Durston under the “right to buy” legislation in the Housing Act 1980. It is not in evidence how that property devolved thereafter, but as I have said it is now registered at the Land Registry in the name of the seventh defendant, subject to leasehold interests in respect of each of the three flats in the building, of which the eighth defendant owns one.

Although I have referred to a number of transfers of land out of the Bathwick Estate since 1922, it is a striking fact that none of these (including those relating to the property of the seventh and eighth defendants) has ever referred to carrying or transferring the benefit of the restrictive covenant in the 1922 conveyance. Indeed, there are very few instances of any covenants being taken at all on transfers out of the Estate. It is also a feature of this case that the 1922 conveyance does not itself contain any plan, although a contemporaneous plan signed by Captain Forester was made available to the purchaser, and did show the land actually conveyed, but not the extent of the land which at that time he retained. It appears that it was not the practice of his then solicitors to include plans in transfers out of the estate.

The evidence of Ms Preist, of the claimant’s solicitors, sets out the research that she and others carried out to establish what properties were in fact in the settlement of the Bathwick Estate at the time of the conveyance in 1922. The detailed nature and extent of that research make crystal-clear the advantages brought to our conveyancing system by the introduction of registration of title to land. In her witness statement of 23 October 2019, at paragraphs [16]-[31], she explains first of all the documents that were transferred to her firm by the claimant’s former solicitors. These included an auction catalogue (together with the plan) for the sale of the whole Bathwick Estate in 1919 ([17]), a similar auction catalogue (again with plan) for the sale of part of the

At paragraph [29] of her witness statement, Ms Preist says:

“Whilst it is not possible to say conclusively what property Captain Forester retained and owned as part of the Bathwick Estate as at 6 April 1922 (the date of the Conveyance) … nevertheless a reasonably accurate assessment can be made (on a property by property basis) by comparing the two auction catalogues predating the 1922 sale with what was transferred to the Bathwick Estate Company in 1924 (and also taking into account the information derived from the newspaper reports of the 1921 auction and sales).”

In his witness statement of 17 January 2019, the eighth defendant says, at paragraph [23],

“Further research at Bath Records Office produced a Schedule of the properties owned by Captain Forester as remaining unsold in 1924 and to be formed into the Bathwick Estate Company. This Schedule is exhibited hereto at GDW3 and shows that 77 Great Pulteney Street and circa 150 other properties remained the property of Captain Forester at the time the Covenant was imposed. It seems to me that this is a perfectly clear and readily available means of identifying properties retained by Captain Forester as part of the Bathwick Estate, which are capable of being benefited by the Covenant, at the time of the 1922 Conveyance.”

Ms Preist responded to that in her witness statement of 30 January 2020. She made the point that the identification of the historical Bathwick Estate could only be identified by extensive research and extrinsic evidence, and not at all from the conveyance itself. That research took a number of visits to the records office and a number of telephone calls. The process of examining files at the record office can take many hours. Whereas in 2013-14 the catalogues to the records were in paper form, there are now online catalogues for about half the collections held by the records office. Individual documents, however, are not available online. The records of the Bathwick Estate were privately held until they were sent to the records office between 1969 and 1983. Finally, she points out that there are small disparities between the

As I have already said, this claim is brought under the Law of Property Act 1925, section 84(2), for a declaration that the restrictive covenant contained in the 1922 conveyance is unenforceable and not binding on the claimant. Section 84(2) (as amended in 1969) relevantly provides that:

“(2)

The court shall have power on the application of any person interested—

(a)

To declare whether or not in any particular case any freehold land is or would in any given event be affected by a restriction imposed by any instrument; or

(b)

To declare what, upon the true construction of any instrument purporting to impose a restriction, is the nature and extent of the restriction thereby imposed and whether the same is or would in any given event be enforceable and if so by whom. …”

There can be no doubt that the claimant, as the owner of a long lease of part of the Rec, is a “person interested” and able to bring this claim. Bath Recreation Ltd, as proprietor of the registered fee simple estate, is stated in the details of claim accompanying the claim form (verified by a statement of truth by the claimant’s

The declaration which the claimant seeks is that no one is entitled to enforce the covenant. That means that the relief sought is intended not merely to bind the defendants, but in effect also to bind anyone in the world. It is therefore necessary for the court to be satisfied, in practice for the claimant to show, that all those who might be affected had the opportunity of knowing of the claim and therefore of participating in it if they wished. In Re Sunnyfield [1932] 1 Ch 79, a similar application was made under section 84(2) of the Law of Property Act 1925. Maugham J said (at 83):

“When such an order as this is asked for, the court ought to make every effort to see that all persons who may wish to oppose the making of the order have the opportunity of being heard, stating their objections in argument before the court, and inviting the court to refuse to exercise its powers. In the present case, it seems that every effort has been made to give notice to all persons having a probable interest in the property, and accordingly I ought not to refuse to proceed to the hearing of the matter.”

In Norwich City College v McQuillin [2009] EWHC 1496 (Ch), Kitchen J dealt with a similar question, and said:

“17.

The procedure under section 84(2) may usefully be invoked where a claimant believes a covenant is, on its proper construction, no longer enforceable and wishes to obtain a clean title, clear of any restrictions. Its value is that it is binding in rem on all persons entitled to the benefit of the restriction, whether they are parties to the proceedings or not. Accordingly, it is well established that when an order of this kind is asked for, the court ought to make every effort to see that all persons who may wish to oppose the making of the order have the opportunity of being heard, stating their objections in argument before the court, and inviting the court to refuse to exercise its powers: Re Sunnyfield [1932] 1 Ch 79.”

In her witness statements of 23 October 2019 and 30 January 2020, Ms Preist sets out the steps which were taken in order to bring the proceedings to the notice of persons who might wish to object or to oppose the application. These were extensive. They included writing to 1112 local addresses, first of all in March 2013, and again in December 2018, together with a stamped addressed envelope for response. A further 99 addresses were later identified and written to. Further letters were sent out to residents who contacted the claimant’s solicitors. Only 81 responses were received in total. 53 of these stated no wish to rely on the covenant, 19 stated that they did wish to rely on it, 11 provided landlords’ addresses (who were then written to, but without response) and two were defaced.

Further, slightly more detailed, letters were then written to the 19 positive responders. These provided that correspondents who did not respond would be assumed to wish to continue to rely on the covenant and they would be joined into the court proceedings. By May 2019 there were nine properties in respect of which the owners were either indicating a wish to oppose the application or had not responded at all. After a further letter was sent to those nine, only the defendants named in the claim form remained.

There was also a process of public consultation. An exhibition was set up in July 2018 at the Rec with details of the plans of the intended development. This contained details of the 1922 covenant and an email address which could be contacted. It appears that no email comments were received. Further consultations were carried out at an open exhibition at the Guildhall in Bath on 5-8 December 2018, preceded by a flyer for the exhibition being sent to the original mailing list. The same details of the covenant and an email address were provided, but again it appears that no responses were received. In addition to this the claimant and its representatives have held meetings and consultations at which their plans and the covenant have been discussed with local interest groups.

There were criticisms made at the hearing to the effect that some neighbours of the Rec had not received personal notification of these proceedings (see the eighth defendant’s witness statement of 17 January 2019, [16]). Given that many of the properties concerned are in multiple occupation, because they have been broken up into flats, that does not surprise me. But the requirement is not to achieve a 100% success rate in contacting persons who might be affected. It is to make “every effort” to give notice of the proceedings to such persons. It is an obligation of means rather than effects. As I have said, mailshotting was not the only form of notice used. In any event this case has sufficient interest and importance for the locality that I do not believe that by the time of the trial there could have been any local resident who had not heard of it. The evidence ably marshalled by the claimant’s solicitors has satisfied

The benefit of a restrictive covenant runs at common law only where, first, the covenant “touches and concerns” the land, and second, the covenantee had and the assignee now has a legal estate in the land benefited. If the covenantee has only an equitable interest in the land (as Captain Forester had) it will run in equity but not at common law. Those conditions are agreed to be satisfied in the present case. It is further agreed between the parties that, in a case where the question of enforcement arises between successors in title of the original covenantor and the original covenantee (as here), there are three ways in which the benefit of a restrictive covenant may pass to a person who wishes to enforce it (“the enforcer”). These are:

The seventh and eighth defendants say that they are entitled to enforce the covenant by virtue of route (1) (annexation). They do not make any claim to enforcement under routes (2) (assignment) or (3) (building scheme). The claimant denies that anyone is entitled to enforce the covenant by any of the three routes. In addition to dealing in detail with route (1), the claimant also dealt rather more summarily with the others, so that the court should be satisfied that, if the defendants’ claim under route (1) failed, no claim would succeed under the other routes either. I deal first with the question of annexation. Because this is a covenant entered into before 1926, the provisions of section 78 of the Law of Property Act 1925 do not apply (see section 78(2)), and there can therefore be no automatic annexation by virtue of that section.

The requirements for annexation of the benefit of a restrictive covenant to certain land were set out by Farwell J in the leading case of Rogers v Hosegood [1900] 2 Ch 338, and his decision was affirmed by the Court of Appeal. In that case Cubitt & Co, the well-known building firm, owned land to the south of Hyde Park in London, which they laid out in building plots. In 1869 they sold one particular plot at the corner of what is now Palace Gate and Kensington Road to the eighth Duke of Bedford. In the conveyance the Duke,

“with intent that the covenants thereinafter on his behalf contained might so far as possible bind the premises thereby conveyed and every part thereof, into whosesoever hands the same might come, and might enure to the benefit of the said [vendors], their heirs and assigns and others claiming under them to all or any of their lands adjoining or near to the said premises, for himself, his heirs and assigns, covenanted with the [vendors], their heirs and assigns, that no more than one messuage or dwelling-house, with such suitable outhouses and stabling (if any) as it might be thought fit to erect in connection therewith, should at any one time be erected or be standing on”

the purchased land. In 1872 the builders sold another of the plots (not immediately adjacent) to the celebrated painter Sir John Millais, and the plot so purchased was conveyed to him, together with “all the rights, easements, or appurtenances belonging or reputed to belong thereto.” However, the covenants entered into by the Duke of Bedford in relation to his plot were not mentioned, nor was there any assignment of their benefit to him. Indeed, Sir John did not even know of their existence.

In 1872 the eighth Duke died and was succeeded as owner of the plot by the ninth Duke. In 1876 the builder-vendors by deed released the Duke, so far as they lawfully could, from the burden of the covenant entered into by his predecessor in 1869, and the Duke covenanted to indemnify them in respect of this. Thereafter the buildervendors sold two plots to Rogers, who was in fact one of the partners in the firm. In 1896 Sir John Millais died, and his estate passed to the trustees of his will. The ninth Duke sold the burdened land to the defendant Hosegood, with notice of the restrictive covenants. The defendant then proposed to erect a large block of flats on the site. Rogers and the trustees of Millais’s will together sought to restrain the defendant by injunction from breaching the terms of the restrictive covenant. The trustees succeeded both at first instance, and also on appeal. (The effect of the release of 1876 of Cubitt & Co meant that Rogers, as a member of the partnership, could not succeed. But this did not affect Sir John Millais’s trustees.)

The more detailed treatment of the relevant law is contained in the judgment of Farwell J at first instance. He said (at pages 394-97):

“In my opinion, the benefit of the covenants runs at law with the land now vested in the Millais trustees. … The accurate expression appears to me to be that the covenants are annexed to the land, and pass with it in much the same way as title deeds … Covenants which run with the land must have the following characteristics: (1) They must be made with a covenantee who has an interest in the land to which they refer. (2) They must concern or touch the land. It is not contended that the covenants in question in this case have not the first characteristic, but it is said that they fail in the second. I am of opinion that they possess both. Adopting the definition of Bayley J in Congleton Corporation v. Pattison, the covenant must either affect the land as regards mode of occupation, or it must be such as per se, and not merely from collateral circumstances, affects the value of the land. It is to my mind obvious that the value of Sir J. Millais's land is directly increased by the covenants in question. … But a covenant may have the two characteristics above mentioned and yet not run with the land; it is in each case a question of intention to be determined by the Court on the construction of the particular document, and with due regard to the nature of the covenant and the surrounding, circumstances. No covenant can run with the land which has not the two characteristics above mentioned, but every covenant which has those two characteristics does not necessarily run with the land. That it is a question of intention in each case, to be determined on construction, is apparent from the judgment of Hall V-C in Renals v Cowlishaw, a judgment of the highest authority, not only from the weight attaching to the opinion of the learned ViceChancellor, but also from the approval that it has received in the Court of Appeal and the House of Lords: see Spicer v Martin and Nottingham Patent Brick and Tile Co v Butler. The Courts have drawn the inference that the parties intended, or in other words the Courts have held on the true construction of the documents that they have contracted, that the covenants shall or shall not run with the land from various circumstances. … Treating it then as a question of construction, I find the express contract to be that these covenants were entered into with intent that they should bind the premises conveyed by the deeds of 1869 and every part thereof into whosesoever hands the same might come, and should enure, to the benefit of Messrs. Cubitt & Co, their heirs and assigns, and others claiming under them.”

In the first sentence of that extract, the judge used the expression “runs at law with the land”. In fact, the benefit of the covenant could not run with the land at law, because the builder-vendors only had an equitable estate, the legal estate being outstanding in a mortgagee. As is (fortunately) still well known amongst property lawyers today, a legal mortgage before 1926 would have involved conveying the legal estate in the land to the mortgagee, rather than (as post-1925) retaining the legal estate and granting only a legal charge to the mortgagee. When the matter went to the Court of Appeal, that court had to deal with this new point, as well as with the substance of the judge’s reasoning.

The Court of Appeal dismissed the appeal from Farwell J. Collins LJ, delivering the judgment of the court (Lord Alverstone MR, Rigby LJ and himself, three common law judges), said this, at pages 403-04:

“The real and only difficulty arises on the question—whether the benefit of the covenants has passed to the assigns of Sir John Millais as owners of the plot purchased by him on March 25, 1873, there being no evidence that he knew of these covenants when he bought. Here, again, the difficulty is narrowed, because by express declaration on the face of the conveyances of 1869 the benefit of the two covenants in question was intended for all or any of the vendor's lands near to or adjoining the plot sold, and therefore for (among others) the plot of land acquired by Sir John Millais, and that they ‘touched and concerned’ that land within the meaning of those words so as to run with the land at law we do not doubt. Therefore, but for a technical difficulty which was not raised before Farwell J, we should agree with him that the benefit of the covenants in question was annexed to and passed to Sir John Millais by the conveyance of the land which he bought in 1873.”

The substance of the judgment is then taken up with dealing with the technical point not taken in the court below. As I have said, this related to the fact that the land was mortgaged by the vendor-builders, and the Duke’s covenant of 1869 was made with them only, and not with the mortgagee as well. In the event, after considering the matter in some detail, the Court of Appeal held that the mortgage made no difference to the situation, and dismissed the appeal, essentially on the basis that the benefit of the covenant passed in equity, if not at law, because the rules of annexation in equity were the same. I do not need to spend time on that aspect in this judgment.

So the critical question is whether there is manifested in the conveyancing documents, construed in the light of the surrounding circumstances, an intention to benefit certain land. In Rogers v Hosegood, the covenant itself contained the words

“with intent that the covenants thereinafter on his behalf contained … might enure to the benefit of the said [vendors], their heirs and assigns and others claiming under them to all or any of their lands adjoining or near to the said premises,”

This was held to be sufficient. Indeed, it would now be regarded as a “classic formula”: see Megarry & Wade, The Law of Real Property, 9^th ed, [31-061]. The vendors retained other land at the time of the 1869 conveyance, “adjoining or near to” the land sold, and then sold some of that retained land to Sir John Millais and some of it to Rogers. That retained land had the benefit of the covenant annexed to it, although Rogers (but not the trustees of Sir John Millais’s will) lost that benefit by the release of 1876.

However, it is clear that both the intention and the identification of the land must be gathered from the conveyancing documents themselves, construed in the light of surrounding circumstances, rather than simply from surrounding circumstances alone:

On the other hand, the intention to be found in the documents can be implied rather than express. This was not controversial at the hearing. In Shropshire County Council v Edwards (1982) 46 P & CR 270, HHJ Rubin, sitting as a High Court judge, in a case on a covenant made in 1908, considered the authorities, and held (at pages 277-78) that:

“The conclusion I draw from these authorities is that it is not necessary, though highly desirable, that express words should be used to annex the benefit of the covenant to the land with which it is to run. If, on the construction of the instrument creating the restrictive covenant, both the land which is intended to be benefited and an intention to benefit that land, as distinct from benefiting the covenantee personally, can be clearly established, then the benefit of the covenant will be annexed to that land and run with it, notwithstanding the absence of express words of annexation.”

And more recently, in Crest Nicholson (South) Ltd v McAlister [2004] 1 WLR 2409, a case concerning a post-1925 covenant, Chadwick LJ (with whom Auld and Arden LJJ agreed) said:

“23.

… In covenants made before 1926 it was necessary to show, by construing the instrument in the light of surrounding circumstances, that annexation to the covenantee's retained land (or some part of it) was intended. Express words of annexation were not required.”

I should say that Mr Moffett submitted that Marten v Flight Refuelling Ltd [1962] 1 Ch 115, a decision of Wilberforce J, was similarly a case of annexation where the intention to annex the restrictive covenant was implied rather than expressed. But I do

There are many decisions of the courts on whether a particular form of words in a particular conveyance was or was not sufficient to show the intention to annex the benefit of a covenant to particular land. In the Sainsbury case to which I have already referred the covenant was made in 1894, and thus attracted the statutory benefit of section 58(1) of the Conveyancing Act 1881. This provided that covenants relating to lands of inheritance were "deemed to be made with the covenantee, his heirs and assigns, and shall have effect as if heirs and assigns were expressed." The judge (Morritt J) therefore had to construe the covenant on that basis. On the way he noted (at page 597H) that

“If it had been intended to annex the benefit of the purchasers' covenants to the retained land it is remarkable that there is no reference to the retained land or to Alfred Walker junior's successors in title in that covenant” (emphasis supplied).

The judge referred to cases such as Renals v Cowlishaw, 9 ChD 125, Reid v Bickerstaff [1909] 2 Ch 305, Ives v Brown [1919] 2 Ch 314, and Miles v Easter (Re Union of London and Smith’s Bank Ltd’s Conveyance) [1933] Ch 611, and concluded that the (deemed) reference in the covenant to “heirs and assigns” was insufficient to show an intention to annex the benefit to any land. It was a phrase consistent both with retaining the benefit personally and with annexing it to the land.

Morritt J’s interlocutory comments about the absence of the phrase “successors in title” suggest that the judge would, or at least might, have taken a different view in that case if those words had been used, at least as long as the relevant land could be identified. I say this, not only because the use of the phrase “successors in title” is generally restricted to the context of land, or at least to that of property rights, but also because those words had recently been judicially endowed with great significance in this context.

As is well known, section 78 of the Law of Property Act 1925 essentially operated in the same way as section 58(1) of the 1881 Act, in that it deemed certain covenants to be made with certain people. However, whereas section 58(1) of the 1881 Act deemed

In Federated Homes v Mill Lodge Properties [1980] 1 WLR 594, the Court of Appeal held that the effect of section 78(1), adding these words to such covenants, was to annex the benefit of the covenant to the land in question. Of course, it is right to say that the court was assisted in its view by the words in the second sentence of section 78(1), dealing with “owners and occupiers for the time being of the land of the covenantee intended to be benefited”. However, in his leading judgment (with whom Megaw and Browne LJJ agreed), Brightman LJ treated the “owners and occupiers” as simply included in the phrase “successors in title” (see at page 604G). In my judgment, therefore, the use of the phrase “successors in title” is far more potent in showing an intention to benefit particular land of the covenantee than the phrase

“heirs and assigns”. And it will be recalled that the phrase used in the present case is indeed

“covenant with the Vendor his successors in title and assigns…”

Mr Dray submitted, and I accept, that the use of the word “successors” by itself has been held not to demonstrate the intention to annex the benefit to retained land. In Seymour Road (Southampton) Ltd v Williams [2010] EWHC 111 (Ch), the claimant sought a declaration that covenants contained in a conveyance of 1896 were no longer enforceable. The application was in fact not opposed. The restrictive covenant entered into by the claimant’s predecessor in title was (so far as material) as follows:

“the purchaser with intent to bind all persons in whom the piece of land expressed be hereby conveyed shall, for the time being, be vested, … HEREBY covenants with the Vendors, their successors and assigns in the manner following…”

Peter Smith J held that there was no identification in the conveyance of any land to which it could be said that the benefit of the covenant was annexed, nor any identification of any land in respect of “successors”. His conclusion was that the covenants were intended to be exercisable by the covenantee only as long as it retained land. It was accepted that the original covenantee had long since ceased to hold any land, and the judge therefore held that the covenants were now unenforceable.

It is trite law that cases are decided on their own particular facts. The circumstances in which land is conveyed and covenants are taken are infinitely variable, because sellers and buyers of land are free to bargain and agree (subject only to public policy) what they like in relation to the land which they wish to sell and buy. The question of law in that case, as in this, was whether annexation of a pre-1926 covenant had occurred. The judge, having looked at the terms of the conveyance and the surrounding circumstances, was not satisfied that the use of the word “successors” demonstrated the intention to annex the benefit of the covenant to any land. In any event, no land was identified. But in the present case the terms of the conveyance are very different, including the fact that the phrase used is “successors in title”, and the surrounding circumstances are completely different. I must decide on the all-important question of intention by reference to the facts of this case, and not those of any other.

There is a further point with which I must expressly deal. Mr Dray submitted that the words “successors in title” were not used in this covenant with reference to the lands of the covenantee which might thereafter be alienated to third parties. Instead, they were used with reference to his position as tenant for life of the Bathwick Estate. He said that they showed the intention to benefit Captain Forester and those who came after him in right of the strict (re)settlement of 1920. Thus, insofar as the use of the words “successors in title” showed an intention to benefit any land, they went no further than tying the benefit to land remaining in the settlement from time to time thereafter. And all the land in the settlement had disappeared by 1924, when the remaining parcels were transferred to the Bathwick Estate Company in return for shares which were held by the trustees of the settlement for the benefit of the beneficiaries. At that stage the covenant ceased to be enforceable in relation to the land, because the shares replaced the land as the subject of the settlement.

I reject this submission. The phrase “successors in title” in its ordinary property law sense refers to those who become owners successively of an estate or interest in a particular parcel of land. In that ordinary sense, it does not refer to successive tenants for life in a strict settlement. I accept that the then relevant legislation, the Settled Land Act 1882, did refer in sections 10(2), 28(1), 29, 31(2),(3) and 63(2) to the “successors in title” of the tenant for life. But in each case this was for the particular purposes of the section in question, typically describing the position, powers and obligations of successive tenants for life. To that extent I accept that there is a special “settled land” meaning to the phrase.

But the 1922 conveyance was not a document internal to the settlement and governing or amending any part of the structure of the settlement. It was not a constitutive document of the settlement, of the kind that a beneficiary might ask to see as of right in the Re Londonderry’s Settlement [1965] Ch 918 sense. It was a document external to the settlement, facing outwards to and having effect upon third parties, an ordinary conveyance of land out of the settlement to strangers to it, and a document of title to the land concerned for the future. Captain Forester was described throughout as “the Vendor”, and by clause 1 conveyed “in exercise of the power … conferred by the [settled land legislation] and as beneficial owner…” (emphasis supplied). That sale overreached the settlement, and transferred the interests of the beneficiaries to the purchase money, paid to the trustees of the settlement (see section 20(2) of the 1882 Act).

If, before entering into the transaction, the purchaser of the land under that conveyance had asked to see the settlement and its other constitutive documents, on the basis that it wished to ascertain who would be able in future to enforce the covenant as “successors in title” to the tenant for life, I cannot doubt that it would have been told that it was none of its business who were the tenants for life and other beneficiaries of the settlement in future, and that it should look at the conveyance itself. The conveyance refers to successors in title to “the Vendor”, who is described

During the course of the argument I asked Mr Dray for the claimant what would be the point of the vendor’s entering into this form of covenant, binding on the land being conveyed, if not to benefit some retained land of the vendor, in order to keep up or even increase its value in future. Mr Dray’s response was that this simply was a covenant for the personal benefit of the vendor. It enabled the vendor to control any future development of the land conveyed, and to exact a further premium in case the land should increase in value as a result of such development. It was, in some ways, an early form of ‘overage’ provision. I accept that a vendor could do this. In such a case the covenant would be in effect purely for his or her personal benefit, and not at all for that of the subsequent owners of the vendor’s retained land.

But in my judgment that is not this case. This covenant forbids certain kinds of development, namely the erection of

“workshops warehouses factories or other buildings for the purpose of any trade or business which may be or grow to be a nuisance and annoyance or disturbance or otherwise prejudicially affect the adjoining premises or the neighbourhood”.

It also provides that nothing is to be built or done on the land conveyed

“which may be or grow to be a nuisance and annoyance or disturbance or otherwise prejudicially affect the adjoining premises or the neighbourhood”.

Such a form of words as this does not express a desire to take a premium for development that makes money for the developer. This is because it concentrates, not on increases in value of the land, but on things done on it that might be a nuisance in the neighbourhood, yet which might cause no increase in value of the land. In my judgment, these words express a desire to protect neighbouring land, rather than to profit the vendor. Moreover, the form of this covenant does not forbid the most obvious form of lucrative development of all, namely the construction of more houses or other residential development. If the covenant were indeed personal to the vendor, as a means of profiting from future development, it would not sensibly omit this possibility.

In my judgment, taken as a whole, the wording of this covenant, in the context in which it was made, including the use of the phrase “successors in title”, does indicate an intention to annex the benefit of the covenant to land of the covenantee. For the avoidance of doubt, I confirm that the absence of specific reference in subsequent transactions to carrying or passing on the benefit of the covenant does not persuade me to a different conclusion. What matters here (that is, in relation to annexation) is what the parties intended in 1922. The evidence of what they did subsequently, although relevant to the question of intention in 1922, is not strong enough to overcome the effect of the terms of the covenant itself.

But it not enough to demonstrate an intention to benefit land. It is also necessary to identify the land intended to be benefited. So, in Re Union of London and Smith’s Bank Ltd’s Conveyance [1933] Ch 611, a case of pre-1926 covenants, Bennett J at first instance held that they were not enforceable, saying (at 620):

“It is clear that if it be intended to annex to a parcel of land the benefit of a covenant restricting the user of another parcel of land, the deed entered into to give effect to this intention must define the parcel of land to which the benefit of the covenant is to be annexed : see Renals v Cowlishaw.”

The Court of Appeal dismissed an appeal against the decision of Bennett J. However, on this point, Romer LJ, giving the decision of the court (Lord Hanworth MR, Lawrence LJ and himself), said, at page 628, that

“a purchaser from the original covenantee of land retained by him when he executed the conveyance containing the covenant will be entitled to the benefit of the covenant if the conveyance shows that the covenant was intended to enure for the benefit of that particular land. It follows that, if what is being acquired by the purchaser was only part of the land shown by the conveyance as being intended to be benefited, it must also be shown that the benefit was intended to enure to each portion of that land.”

It is to be noted that Romer LJ does not acquiesce in the strict terms of Bennett J’s judgment. He does not say, as the judge below did, that “the deed … must define the parcel of land”. Instead he says “if the conveyance shows that the covenant was intended to enure for the benefit of that particular land”.

I should say that Mr Dray’s skeleton referred to a passage on page 631 of the judgment, but as I read the judgment that passage, beginning “In the next place…”, is actually concerned with a different situation, where the benefit of the covenant has been assigned to others. This is clear from the passage on page 629 introducing the discission leading up to page 631, which says:

“In neither of these cases, therefore, did it become necessary for the Court to inquire into the circumstances in which an express assignee of the benefit of a covenant that does not run with the land is entitled to enforce it. In the present case, however, it is necessary to do so …”

But here also Romer LJ takes a less strict view than Bennett J, saying on page 631,

“the Court will readily infer the intention to benefit the other land of the vendor where the existence and situation of such land are indicated in the conveyance or have been otherwise shown with reasonable certainty” (emphasis supplied).

This admits of the possibility that the land to be benefited is identified by the words of the conveyance, but construed in the light of surrounding circumstances. As it happens, and as was pointed out by Wilberforce J in Marten v Flight Refuelling Ltd

In the later case of Newton Abbot Co-operative Society Ltd v Williamson & Treadgold

On the question whether the benefit of the covenant had been annexed to any land (in particular to Devonia), Upjohn J said this at page 289:

“In this difficult branch of the law one thing in my judgment is clear, namely, that in order to annex the benefit of a restrictive covenant to land, so that it runs with the land without express assignment on a subsequent assignment of the land, the land for the benefit of which it is taken must be clearly identified in the conveyance creating the covenant.”

He referred to the decisions of the Court of Appeal in Renals v Cowlishaw and Re Union of London and Smith’s Bank Ltd’s Conveyance as justifying his view. He held that there was “nothing whatever which identifies the land for the benefit of which the covenant is alleged to be taken”, and that the mere mention in the conveyance of Devonia as the vendor’s address was not enough.

On the other hand, Upjohn J also considered the requirements for the assignment of the benefit of a restrictive covenant to be effective. On this point, he expressly declined to agree with the statement of Bennett J at first instance in Re Union of London and Smith’s Bank Ltd’s Conveyance, [1933] Ch 611, at page 625, that

“the assignee must be able to satisfy the court that the deed containing the covenant defines or contains something to define the property for the benefit of which the covenant was entered into …”

Instead, Upjohn J said (at pages 295-96) that the authority on which Bennett J had relied did not in fact support him, and that what Bennett J had said was inconsistent with a number of Court of Appeal decisions, as well as the observations of Romer LJ in the Court of Appeal on appeal from his own decision (as referred to above). Upjohn J held that the vendor

“took the covenant restrictive of the user of the defendants' premises for the benefit of her own business of ironmonger and of her property Devonia where at all material times she was carrying on that business”.

Accordingly the plaintiff succeeded on the assignment argument.

It is worth recalling, too, that in Rogers v Hosegood (which is referred to in both Re Union of London and Smith’s Bank Ltd’s Conveyance and Newton Abbot Cooperative Society Ltd v Williamson & Treadgold Ltd, in each case with seeming approval), the covenant concerned identified the land to be benefited by the covenant, for the purposes of annexation of that covenant, as follows:

“enure to the benefit of the said [vendors], their heirs and assigns and others claiming under them to all or any of their lands adjoining or near to the said premises…” (emphasis supplied).

There are two points to be made at this stage about this description of land. The first is that it cannot be understood on its own. A factual enquiry must be made outside the conveyance, to see what other “adjoining or near” land was owned by the vendors at the time. The second is that it uses the words “adjoining” and “near”. The former is reasonably precise (although there might be an argument about a plot which is separated from another merely by a narrow pathway, or which touches that other only at a point and not along a line). Yet the latter term “near” is far from precise. How near is near? A few yards, a mile, ten miles? And yet the Court of Appeal in Rogers v Hosegood held that the benefit of the restrictive covenant had passed by annexation to Rogers and to the trustees of Sir Millais’s will. Moreover, in saying what they did about the test for land to be benefited for the purposes of annexation, the judges in Re Union of London and Smith’s Bank Ltd’s Conveyance and Newton Abbot Cooperative Society Ltd v Williamson & Treadgold Ltd must have intended that those words would satisfy any test that they might lay down, for they could hardly be supposed to have intended to reverse the decision in Rogers v Hosegood.

In my judgment, therefore, when Romer LJ says “the covenant must be intended to enure for the benefit of that particular land”, and Upjohn J says “the land for the benefit of which it is taken must be clearly identified in the conveyance”, both are accepting that the phrase “lands [of the vendor] adjoining or near to the said premises” sufficiently complies. It is a test of linguistic certainty, that is, do we know what the parties meant by the words which they used? The fact that it may be necessary in some cases to resort to extrinsic evidence to work out the application of the words used in the conveyance is not an objection. That is a question of evidence. What would be objectionable is a case where the conveyance does not provide any formula or other basis for the external enquiry, and where only surrounding circumstances provide an answer (as with Devonia in the Newton Abbott case).

For completeness, I mention also that in Crest Nicholson Residential (South) Ltd) v McAllister [2004] 1 WLR 2409, Chadwick LJ (with whom Auld and Arden LJJ

But Mr Dray for the claimant went on to make a further submission in this connection. He argued that for the purposes of the doctrine of annexation, even in the context of a pre-1926 covenant, the land to be benefited had in addition to be “easily ascertainable”. He based this argument on two decisions of the Court of Appeal, Marquess of Zetland v Driver [1939] Ch 1, and Crest Nicholson Residential (South) Ltd v McAllister [2004] 1 WLR 2409, each of them dealing with post-1925 restrictive covenants. Both are decisions of the Court of Appeal, and therefore they are binding on me for whatever they decide.

In the first of these cases, the covenant on a conveyance made in 1928 out of a settlement provided that

“The purchaser to the intent and so as to bind as far as practicable the said property hereby conveyed into whosesoever hands the same may come and to benefit and protect such part or parts of the lands in [Redcar], now subject to the settlement (a) as shall for the time being remain unsold or (b) as shall be sold by the vendor or his successors in title with the express benefit of this covenant, covenants with the vendor that the purchaser will at all times observe, perform and keep the said stipulations contained in the Second Schedule hereto.”

The plaintiff as successor in title to the vendor sought to enforce the covenant against the defendant as assignee of part of the land so sold. The judge at first instance, Bennett J, dismissed the action on the basis that the plaintiff had no right to enforce the covenant, as the benefit of the covenant had not been effectually annexed to any land. The matter was however taken to appeal, and the appeal succeeded.

Farwell J, giving the judgment of the Court of Appeal (Sir Wilfred Greene MR, Luxmoore J and himself) said (at page 7):

“Covenants restricting the user of land imposed by a vendor upon a sale fall into three classes : (i.) covenants imposed by the vendor for his own benefit; (ii.) covenants imposed by the vendor as owner of other land, of which that sold formed a part, and intended to protect or benefit the unsold land; and (iii.) covenants imposed by a vendor upon a sale of land to various purchasers who are intended mutually to enjoy the benefit of and be bound by the covenants.”

On page 8 he dismissed cases of the first and third classes as inapplicable to this case, and continued:

“If, therefore, the appellant is entitled to sue on this covenant it must fall within the second class above mentioned.

Such covenants can only be validly imposed if they comply with certain conditions. Firstly, they must be negative covenants. … Secondly, the covenant must be one that touches or concerns the land … Further, the land retained by the vendor must be such as to be capable of being benefited by the covenant at the time when it is imposed. Thirdly, the land which is intended to be benefited must be so defined as to be easily ascertainable, and the fact that the covenant is imposed for the benefit of that particular land should be stated in the conveyance and the persons or the class of persons entitled to enforce it” (emphasis supplied).

The court concluded (at pages 8-9):

“The covenant is restrictive; it is expressly stated in the conveyance to be for the benefit of the unsold part of the land comprised in the settlement and such land is easily ascertainable, nor is it suggested that at the date of the conveyance the land retained was not capable of being benefited by the restrictions, and lastly the appellant is the successor in title of the original covenantee and as such is the estate owner of part of the land unsold which is subject to the settlement. That being so, the appellant is the person now entitled to the benefit of the covenant and, prima facie, is entitled to enforce it against the respondents who, although not the original covenantors, took their land with notice of the restrictions and are, therefore, bound by them.”

It is right to mention that there was no reference in this case to any possible effect of the Law of Property Act 1925, section 78. In the event the court reversed the decision of the judge at first instance, and granted an injunction to restrain a breach of the covenant. It is notable that the Court of Appeal gave no sign that it considered it was differing from its earlier decisions in Rogers v Hosegood and Re Union of London and Smith’s Bank’s Conveyance, the latter only six years before. It may well be therefore that all that that court meant by the phrase “easily ascertainable” was that the terms of the conveyance, taken with the surrounding circumstances, enabled the court to identify the land intended to be benefited. If so, it made no change to the law.

In the later case of Crest Nicholson Residential (South) Ltd v McAllister, a company divided land into building plots and in 1928 sold three of them off, subject to restrictive covenants limiting the purposes for which the properties could be used. Three further plots were sold in 1933 to the same purchasers, one to each, to enlarge their properties. The benefit of the covenants was stated to be annexed to “the property of [the company] or the part thereof for the time being remaining unsold” (emphasis supplied). Subsequently a fourth plot was sold in 1936 by the company, subject to different covenants, and on which three further houses were built.

The claimant bought part of the gardens of four of the six houses, and the whole of the property of a fifth, intending to build further houses thereon. The owner of the sixth house (built on the 1936 land) objected. The claimant sought a declaration against the owner of the sixth house as to the non-enforceability of the covenants. At first instance it was conceded that the benefit of the covenants in the 1928 conveyances had been annexed to the defendant’s land, and Neuberger J held that the defendant was entitled to enforce the covenant against the claimant. The claimant appealed to the Court of Appeal. At the hearing itself it successfully applied to withdraw the concession of annexation of the benefit of the covenant to the defendant’s land.

Chadwick LJ (with whom Auld and Arden LJJ agreed) briefly discussed (at [23]) the legal position for the annexation of covenants entered into before 1926. I have quoted the substance of what he said about this above. In contrast, he then turned to post1925 covenants, and said:

“24.

In relation to covenants imposed in instruments made after 1925 (as were the covenants with which we are concerned in this appeal) the position is governed by the provisions of section 78 of the Law of Property Act 1925 …

25.

The effect of section 78 of the 1925 Act was considered by this Court in

It can be seen, therefore, that, in relation to the green land but not in relation to the red land, the plaintiff was able to rely on a chain of express assignments. …

… The effect, as Lord Justice Brightman observed (ibid, 603B), was that the covenant was plainly enforceable for the benefit of the green land. He said this:

“Having reached the conclusion that the restrictive covenant was capable of assignment and is not spent, I turn to the question whether the benefit has safely reached the hands of the plaintiff. The green land has no problem, owing to the unbroken chain of assignments. I am disposed to think that that is sufficient to entitle the plaintiff to relief, and that the plaintiff’s right to relief would be no greater at the present time if it were held that it also had the benefit of the covenant in its capacity as owner of the red land.”

Nevertheless, he went on to consider annexation, which was, of course, material to the question whether the density covenant was enforceable for the benefit of the red land. He explained that:

“An express assignment of the benefit of a covenant is not necessary if the benefit of the covenant is annexed to the land. In that event, the benefit will pass automatically on a conveyance of the land, without express mention, because it is annexed to the land and runs with it.”

[ … ]

It is clear that the Court approached the question of annexation in the Federated Homes case on the basis that the density covenant was taken for the benefit of retained land which could be identified in the 1971 conveyance. Lord Justice Brightman expressed his conclusion in these terms (ibid, 605A-C):

“If, as the language of section 78 implies, a covenant relating to land which is restrictive of the user thereof is enforceable at the suit of (1) a successor in title of the covenantee, (2) a person deriving title under the covenantee or under his successors in title, and (3) the owner or occupier of the land intended to be benefited by the covenant, it must, in my view, follow that the covenant runs with the land, because ex hypothesi every successor in title to the land, every derivative proprietor of the land and every other owner and occupier has a right by statute to the covenant. In other words, if the condition precedent of section 78 is satisfied – that is to say, there exists a covenant which touches and concerns the land of the covenantee – that covenant runs with the land for the benefit of his successors in title, persons deriving title under him or them and other owners and occupiers.”

The decision of this Court in the Federated Homes case leaves open the question whether section 78 of the 1925 Act only effects annexation when the land intended to be benefited is described in the instrument itself (by express words or necessary implication, albeit that it may be necessary to have regard to evidence outside the document fully to identify that land) or whether it is enough that it can be shown, from evidence wholly outside the document, that the covenant does in fact touch and concern land of the covenantee which can be identified.

It is clear from Lord Justice Brightman’s reference (ibid) to Rogers v Hosegood [1900] 2 Ch 388 that it is sufficient for the conveyance to describe the land intended to be benefited in terms which enable it to be identified from other evidence. …

The question left open in the Federated Homes case had, I think, already been answered in the judgment of this Court in Marquess of Zetland v Driver [1939] Ch 1, a decision not cited in Federated Homes. The applicable principles were restated in the following passage, (ibid,7-8):

“Covenants restricting the user of land imposed by a vendor upon a sale fall into three classes: (i) covenants imposed by a vendor for his own benefit; (ii) covenants imposed by a vendor as owner of other land, of which that sold formed a part, and intended to protect or benefit the unsold land; and (iii) covenants imposed by a vendor upon a sale of land to various purchasers who are intended mutually to enjoy the benefit of and be bound by the covenants: Osborne v Bradley [1903] 2 Ch 446, 450.

Covenants of the first class are personal to the vendor and enforceable by him alone unless expressly assigned by him. Covenants of the second class are said to run with the land and are enforceable without express assignment by the owner for the time being of the land for the benefit of which they were imposed. Covenants of the third class are most usually found in sales under building scheme, although not strictly confined to such sales. It is not suggested that the present covenant falls within this class. Nor will it assist the appellant if it falls within the first class, since he was not the original covenantee or an express assignee from him. If, therefore, the appellant is entitled to sue on this covenant it must fall within the second class above mentioned.

Such covenants can only be validly imposed if they comply with certain conditions. Firstly, they must be negative covenants. . . . Secondly, the covenant must be one that touches or concerns the land, by which is meant that it must be imposed for the benefit or to enhance the value of the land retained by the vendor or some part of it, and no such covenant can ever be imposed if the sale comprises the whole of the vendor’s land. . . . Thirdly, the land which is intended to be benefited must be so defined as to be easily ascertainable, and the fact that the covenant is imposed for the benefit of that particular land should be stated in the conveyance and the persons or the class of persons entitled to enforce it. The fact that the benefit of the covenant is not intended to pass to all persons into whose hands the unsold land may come is not objectionable so long as the class of persons intended to have the benefit of the covenant is clearly defined.” [emphasis added]

In its later decision in the Federated Homes case this Court held that the provisions of section 78 of the 1925 Act had made it unnecessary to state, in the conveyance, that the covenant was to be enforceable by persons deriving title under the covenantee or under his successors in title and the owner or occupier of the land intended to be benefited, or that the covenant was to run with the land intended to be benefited; but there is nothing in that case which suggests that it is no longer necessary that the land which is intended to be benefited should be so defined that it is easily ascertainable. In my view, that requirement, identified in Marquess of Zetland v Driver remains a necessary condition for annexation.”

There are three important points to make about the case of Crest Nicholson. The first is that it is to be noted that the claimant actually succeeded because, on the terms of the relevant covenant, once the land was sold to the defendant’s predecessor in title, it ceased to be “the property of [the company] … remaining unsold”, and the covenant could no longer apply. It was in effect a covenant for the benefit of the company alone, which it retained in order to control the land sold until it disposed of the remaining land, and to enhance the value of its remaining land when it came to sell it (see at [48]). A company has no heir, and by its own terms the covenant could never enure for the benefit of a purchaser from the company. That would require an assignment of the benefit of the covenant to the purchaser. So all the discussion about the test for annexation was in fact obiter dicta. There could never be an effective annexation in favour of a third party.

The second point is that there is no doubt that in this case Chadwick LJ was dealing, and intending to deal, only with the position for covenants after 1925. He was concerned with (and deciding) an aspect of the effect of section 78 of the 1925 Act (which ex hypothesi could not apply to any covenant entered into before 1926) on a post-1925 covenant. As a decision, therefore, it is not binding in relation to a pre1926 covenant.

The third point is that, in paragraph 31 of his judgment, Chadwick LJ said that it was

“sufficient for the conveyance to describe the land intended to be benefited in terms which enable it to be identified from other evidence.”

Strictly speaking, therefore, those points are enough to deal with Mr Dray’s argument. But I think I should add this. Although both decisions of the Court of Appeal refer to the expression “easily ascertainable”, there is no explanation given in either of them as to what the phrase actually means, and how it differs from (say) the formulation used by Romer LJ in Re Union of London and Smith’s Bank’s Conveyance (“if the conveyance shows that the covenant was intended to enure for the benefit of that particular land”) or that used by Chadwick LJ himself (“it is sufficient for the conveyance to describe the land intended to be benefited in terms which enable it to be identified from other evidence”).

Nor is there any consideration of how easy it would have been to ascertain the land of Cubitt & Co “adjoining or near” the burdened land in Rogers v Hosegood. If that formulation is sufficient (as everyone accepts it is), must there then be a further inquiry on the facts, to see how easy the exercise is? If so, there is no trace in the books of such a thing actually happening. Apart from Crest Nicholson itself (where the point is obiter), I am not aware of any case where the court has said that the formula in the conveyance is fine but it is not in fact “easy” to ascertain the land to be benefited by reference to it, and held on that account that the covenant is not enforceable.

In Coventry School Foundation v Whitehouse [2014] 1 P & CR 4, CA, a post-1925 covenant case, the judge at first instance found that the benefited land was “easily ascertainable” from the conveyance and extrinsic evidence. The Court of Appeal did not have to decide the point. Although it was cited to me at the hearing, in my judgment it turns on its facts, and for present purposes I respectfully derive no assistance from it.

In Crest Nicholson, Chadwick LJ however refers to good reasons for preferring the formulation “easily ascertainable”. He says:

“34.

… It is obviously desirable that a purchaser of land burdened with a restrictive covenant should be able not only to ascertain, by inspection of the entries on the relevant register, that the land is so burdened, but also to ascertain the land for which the benefit of the covenant was taken – so that he can identify who can enforce the covenant. That latter object is achieved if the land which is intended to be benefited is defined in the instrument so as to be easily ascertainable. To require a purchaser of land burdened with a restrictive covenant, but where the land for the benefit of which the covenant was taken is not described in the instrument, to make enquiries as to what (if any) land the original covenantee retained at the time of the conveyance and what (if any) of that retained land the covenant did, or might have, ‘touched and concerned’ would be oppressive.”

I accept, of course, that those are good arguments in favour of a rule that requires the land to be benefited to be set out in terms in the instrument (and therefore in the extract placed on the register). But, once you accept that what is described in the instrument can be construed by reference to surrounding circumstances or extrinsic evidence, and that “the land retained by the vendor adjoining or near to the purchased land” is sufficient (as is certainly the case for pre-1925 covenants), the degree of ease with which you can ascertain the land must depend on the quality of the extrinsic evidence and the ease of access to it.

Suppose you have two cases with identical covenants in the terms set out above, and in relation to one there is a full conveyancing file available, in one place, easily accessible, which sets out exactly what other land the vendor retained in the vicinity, whereas in the other there is no comprehensive file, and a list of the vendor’s retained land can only be obtained, if at all, by chasing down copies of old deeds from friendly neighbours, careful conveyancers for others and local archives. Does the law really say that the first covenant is enforceable and the second not? And if the comprehensive file in the first case were subsequently destroyed, would the covenant in that case cease to be enforceable? Or, conversely, if a comprehensive file in the second case were to come to light, would that covenant become enforceable? That would seem to me to confuse substance with evidence, that your right exists only if for the time being you have a certain sort of evidence of it. In my judgment, the substance must be judged as at the time of the transaction: the rest is evidence.

Accordingly, I proceed on the basis in this case that, at least in relation to pre-1926 covenants, the test to apply is as formulated by Chadwick LJ in light of Brightman LJ’s judgment in Federated Homes, namely, “it is sufficient for the conveyance to describe the land intended to be benefited in terms which enable it to be identified from other evidence”, and it is not necessary to show further that the land to be benefited is “easily ascertainable”. It is therefore necessary to look at the 1922 conveyance itself, judged in the light of the surrounding circumstances, and decide whether it sufficiently identifies the land to be benefited by the covenant in such terms.

The terms of the covenant in this case have already been set out above. For our purposes, the structure of the conveyance falls into two parts. In clause 1, there is a conveyance of the Rec to the purchaser, subject to an important reservation in favour of the vendor and his successors in title for:

“the free and uninterrupted passage and running water and soil from the other buildings and land of the vendor and his tenants adjoining or near to the said hereditaments hereinbefore described through the sewers drains and water courses which are now or may hereafter be in or under the said premises” (emphasis supplied).

Then clause 2 contains the covenant with the vendor and his successors in title (omitting unnecessary words):

“The Purchasers for themselves their successors and assigns hereby covenant with the Vendor his successors in title and assigns … that no workshops warehouses factories or other buildings for the purpose of any trade or business which may be or grow to be a nuisance and annoyance or disturbance or otherwise prejudicially affect the adjoining premises or the neighbourhood shall at any time hereinafter be erected upon the said hereditaments and premises … and that nothing shall be hereafter erected placed built or done upon the said hereditaments and premises … which may be or grow to be a nuisance and annoyance or disturbance or otherwise prejudicially affect the adjoining premises or the neighbourhood…” (emphasis supplied).

I should say that Mr Dray argued that, if the benefit of the covenant was annexed at all to land of the vendor, this was for only so long as it remained in the settlement. Thus, once the land had been sold out, the benefit ceased to be annexed. I do not accept this. The words of the conveyance to which I have referred do not contain any such limitation. Nor would that be sensible if, as I have held, it is to prevent prejudicial effects on the land itself.

The covenant is one not to do certain things to the prejudice of “the adjoining land or the neighbourhood”. In substance that means that the covenant is for the benefit of the land so described. ‘Adjoining land’ is clear enough. ‘Neighbourhood’ means the area nearby. Reading clause 2 together with clause 1, I am in no doubt that the phrase “the adjoining land or the neighbourhood” in both places where it appears in clause 2 is a reference back to the phrase “the other buildings and land of the vendor and his tenants adjoining or near to the said hereditaments hereinbefore described” in clause 1. Accordingly, in my judgment, the land for the benefit of which the covenant was imposed is sufficiently described in the deed itself. The formula used, that is, land of the vendor “adjoining or near” the land conveyed, is in substance the same as the formula successfully used in Rogers v Hosegood.

In my judgment, given that the seventh and eighth defendants have been shown to have sufficient interests in land which in 1922 formed part of the “land of the vendor and his tenants adjoining or near” the Rec, it must follow that they at least have the benefit of the covenant by annexation. But I need to go on and consider the effect of any uncertainty as to which property is to attract the benefit of the covenant.

Earlier, I referred to litigation concerning the trusts on which the Rec was held: Bath & North East Somerset Council v AG [2002] EWHC 1623 (Ch). In the judgment of Hart J, there is a reference to land of Captain Forester adjoining the Rec. The judge, referring to the covenant in the 1922 conveyance, said this:

“7.

It is clear from internal evidence in the 1922 Conveyance that Captain Forester owned adjoining land, and that the portion of the land occupied by Bath Artcraft Limited consisted of a skating rink. This latter area appears to be that which was conveyed to Bath Artcraft Limited by a conveyance dated 8 April 1922 and by Bath Artcraft Limited to the Corporation by a conveyance dated 27 March 1930. This area is not therefore comprised in the 1956 Conveyance.”

The first part of the first sentence was not a statement necessary to the decision in that case. But it might be thought to be a comment relevant to a question arising in the current proceedings. Yet it must be borne in mind that that case did not involve the present parties, and moreover was not a claim under section 84(2) of the 1925 Act. It is therefore not authoritative in the present proceedings. The decisions of the First Tier Tribunal in 2014, and the Upper Tribunal in 2015, about the charitable scheme affecting the land, are not so far as I can see relevant to the question of enforceability of the restrictive covenant, and neither party suggested otherwise.

On the other hand the admirable research done by the claimant’s solicitors has shown that in 1922 the Bathwick Estate still contained a significant number of properties, mostly let, which would therefore fall within the words “land of the vendor and his tenants adjoining or near” the Rec. I accept that absolute precision as to the properties then still forming part of the Bathwick Estate is not possible on the evidence currently available, and may never become possible in the future. But I decline to hold that, because there is some uncertainty at the fringe, the owners of those properties which can be demonstrated to have formed part of the Estate at the time of the 1922 Conveyance, and which can properly be said to be adjoining or near the Rec, cannot enforce the covenant the benefit of which was in my judgment annexed to their properties. If a person comes forward claiming that the benefit of the covenant has been annexed to his or her property, the burden will lie on that person to show that that is so. For the reasons given, I am satisfied that the seventh and eighth defendants fall into this category. And, because of the existence of documents listing properties that were sold to the Bathwick Estate Company in 1924, there will be others too who can show that their property remained in the Estate in 1922 and was adjoining or near to the Rec.

I turn finally to consider the questions of assignment of the benefit of the covenant and the possible application of a building scheme so as to confer the benefit on subsequent owners. Given my conclusions on annexation, I can deal with this shortly.

In her witness statement of 23 October 2019, at paragraphs [32]-[40], and in her witness statement of 31 January 2020 at paragraph [33], Ms Preist gives evidence of the complete lack found by her of any chain of assignments of the benefit of the covenant in favour of the owners of any property which formed part of the Estate in 1922. Although she did not examine the title of every such property, she did examine the title of the properties of the defendants named in this case, those of other persons who had asserted that they had the benefit of the covenant, and also a selection of other properties from the streets surrounding the Rec.

She also made enquiries with her colleagues who specialise in residential conveyancing in Bath, who confirmed that in their professional experience they do not recall any reference to the assignment of the benefit of this covenant. A similar enquiry was made of conveyancing solicitors at the claimant’s previous solicitors, with a similar result. Finally, no respondent to the claimant’s mailshotting asserted a right to rely on the benefit of the covenant by express assignment or provided any evidence to support any such assertion. This evidence is not challenged by the seventh and eighth defendants, and I accept it. In my judgment, there is no reasonable possibility of the benefit of the covenant having passed by assignment in this case.

Turning then to the question of a building scheme, the legal requirements are set out in the decision of the Court of Appeal in Birdlip v Hunter [2016] EWCA Civ 603, where Lewison LJ (with whom Laws and Gloster LJJ agreed) said :

“2.

The characteristics of [a building] scheme are that:

i)

It applies to a defined area.

ii)

Owners of properties within that area have purchased their properties from a common owner.

iii)

Each of the properties is burdened by covenants which were intended to be mutually enforceable as between the several owners. iv) The limits of that defined area are known to each of the purchasers.

v)

The common owner is himself bound by the scheme, which crystallises on the occasion of the first sale of a plot within the defined area, with the consequence that he is not entitled to dispose of plots within that area otherwise than on the terms of the scheme.

vi)

The effect of the scheme will bind future purchasers of land falling within the area, potentially for ever.”

In her witness statement of 23 October 2019, at paragraphs [41]-[67], and in her witness statement of 31 January 2020 at paragraph [34], Ms Preist gives evidence that TLT solicitors wrote to the claimant in April 2013 (after the initial mailshotting exercise by the claimant) on behalf of 45 local residents, asserting that a building scheme had been created in the late 18^th century. She then gives evidence of examination of historic plans of the Estate, from 1793, 1803 and 1845, which appeared to show development or potential development in the late 18^th and 19^th centuries. She visited the Bath Record Office and inspected the selection of deeds relating to properties which appeared to have been constructed at that time.

These showed that the owners of the Estate (originally the Countess of Bath, and then the Earl of Darlington) granted long leases of the building plots but did not sell off freehold estates. When the leases fell in, the owners (by now including the later Dukes of Cleveland and – as tenant for life – Captain Forrester) renewed them or granted new ones. Accordingly they retained control of the properties on the Estate by means

However, some properties were sold off in the early 19^th century, and many of these reserved annual fee farm rents. Again, however, an examination of sample conveyances showed no covenants similar to that in the 1922 Conveyance. The first bulk sale of individual properties from the Estate took place following the auction of 1921. There is no evidence that any covenants were imposed at this time. The same is true on the sale of the residue of the Estate in 1924. In summary, there is no evidence of any intention to create a reciprocal scheme of covenants mutually enforceable between the various owners or purchasers of properties on the Estate. In the circumstances, any suggestion of a building scheme in my judgment is quite hopeless.

For the reasons given above, I will dismiss this claim for a declaration that the covenant contained in the 1922 Conveyance is not enforceable by anyone. In my judgment, on the materials before me the covenant is enforceable, by virtue of the doctrine of annexation of the benefit of the covenant, by the seventh and eighth defendants, and it will be enforceable by virtue of the same doctrine by others with properties forming part of the Bathwick Estate at the time of the 1922 Conveyance and which are adjoining or near to the Rec.

I should not like to leave this case without expressing my thanks to all the lawyers involved in this highly technical exploration of conveyancing and property law. Although covenants from before 1925 are now increasingly rare, the need for stability in property law means that the law applicable to that period is still important to us today. I am grateful to counsel for their cogent, coherent and measured submissions, and to the solicitors for their diligent and indeed essential research, which has been so clearly presented to the court.