Round Tower Churches

Stephen Hart


  Exterior of All Saints, Frostenden
  All Saints, Frostenden, Suffolk (Photo: David Striker; all other photos: Stephen Hart))

The round church towers of East Anglia are intriguing. Apart from a handful of exceptions in Berkshire (two) and Sussex (three), this form is found nowhere else in England. Why were they built in this way? What was their purpose? And why are there so many in this area? Including those in a semi-ruinous condition and visible remnants of fallen examples, Norfolk has 127 round church towers, Suffolk has 43, Essex six and Cambridgeshire two. Of these, about 160 are of medieval origin, dating from the 11th to the 14th centuries.

Controversy has long surrounded their original purpose. Until the last quarter of the 20th century, when detailed ongoing studies began to examine their construction and how they are structurally integrated with their churches, it was widely believed that they were built as free-standing defence or refuge towers and that only later were churches added to them. However, it has now been established that, except where a nave was rebuilt later, almost all round towers were either contemporary with their churches or were built onto them at a later date. The round tower at Bramfield in Suffolk is the only one that was built as an independent structure from its church. Although Little Snoring tower is also freestanding, it was originally attached to a church which, little more than a century after it was built, was demolished and rebuilt a short distance to the north. Other lone round towers at Wolterton, Ringstead and St Benedict’s, Norwich, are all that remain of lost churches.

  Freestanding round tower at Bramfield, Norfolk
  Bramfield, Norfolk. This is the only round tower built as a freestanding structure from its church.
  Round tower church at Haddiscoe, Norfolk
  Haddiscoe, Norfolk. This Saxo-Norman tower has Caen stone dressings and Saxon style belfry openings with Norman mouldings.
  Round tower church at Little Saxham, Suffolk
  Little Saxham, Suffolk. This is a fine Norman tower with blind arcading in stone between the belfry openings.

It is now widely accepted that round towers were built as bell towers, a conclusion supported by the fact that the belfries of many towers, even the earliest, can be shown to be contemporary with the lower stages of the tower and the belfry openings were self-evidently not built for military or observation purposes. Claims that the towers were built to command strategic positions are part of the now-discredited defence-tower theory, and assessment of the sites does not suggest that their locations are significantly strategically superior to those of other churches.

One possible explanation for their form lies in their construction. Almost all are built of flint, except for a few in East Anglia that are built of dark brown carstone or similar-looking puddingstone, more technically known as ferruginous conglomerate or ferricrete. East Anglia has no sources of freestones of the quality of those of the Lincolnshire limestone formations further to the west, and although the local carstone and conglomerates, and indeed flints, were sometimes used as quoin stones for the corners of early church naves and chancels, there are only three square western church towers in East Anglia with flint corners that may pre-date the 12th century: Hethel, Warham St Mary and Little Bardfield. (Beeston Regis tower, which has flint quoins, is probably 13th century and Heigham is later still.) This suggests that, in the absence of good stone for quoining, the circular shape for early structures of tower height was preferred. Because of the difficulty and expense of obtaining suitable limestone, it would have been logical to build towers of a shape that did not require corners. This has been widely accepted as a functional explanation for the adoption of the circular shape and for the high concentration of round towers in the region.

An alternative theory is that the round tower form may be of continental origin. This supposition derives from the fact that there are about 30 round church towers in regions of northern Germany, Poland and southern Sweden bordering the Baltic Sea, most of which have been dated to the 12th century, although the earliest, at Heeslingen (now demolished), may have been built in the late 10th or early 11th century. This proposition sees the English round towers as a cultural legacy of trading links between these Baltic regions and the East Anglian ports, and holds that the circular shape was adopted as an aesthetic choice inspired by the continental examples rather than for functional reasons.

Whether the circular shape was a practical or aesthetic choice, and whether or not it was inspired by continental precedents, it was a logical solution to the technological challenge of building towers in a region lacking suitable stone for the quoins of a square tower. It will probably never be known for certain whether it was a local phenomenon or an imported cult. What is significant is that it was adopted quickly and universally in East Anglia as a form suited to an area without freestone. Natural conservatism ensured that round towers continued to be built concurrently with square towers long after freestone had become widely available in the area: there are probably more Norman round towers than square ones in East Anglia. This suggests that what may have originated as a functional solution persisted as an economic and aesthetic choice.

As it can be shown that virtually all the features of East Anglian flint buildings that are recognised as characteristic of Anglo-Saxon architecture were also used in the post-Conquest period, sometimes well into the 12th century, buildings that have these features cannot be dated with certainty as pre-Conquest. There are about 40 round towers of the possible Saxon and Saxo-Norman overlap periods, with no evidence to suggest that any may be earlier than the 11th century. A total of 44 Norman round towers have been identified, while post-Norman medieval round towers number about 80. It can therefore be said that (excluding post-Reformation and 19th and 20th century revivals) the building of round towers occurred from the 11th to the 14th centuries.


The belfry openings of the earliest round towers were probably single-light with arched heads and unsplayed reveals and were formed in flint, as at Great Hautbois. The twin-light type, in which a pair of semi-circular or triangular headed arches passes through the full wall thickness, centrally supported by a throughstone carried on a single column, was probably introduced in the 11th century. This type, as seen at Bessingham, has come to be regarded as a hallmark of Anglo-Saxon workmanship and style. However, it continued to be built after the Norman Conquest, and examples at Haddiscoe and Herringfleet have Norman billet mouldings and are built of Caen stone (which is not thought to have been imported into East Anglia before the construction of Norwich Cathedral began in 1096). The Normans also introduced their own method of constructing twin belfry openings: they dispensed with the throughstone and spanned the full width of the two lights with a single arch and a pair of smaller, thinner sub-arches supported on a column directly beneath, placed at the outer face of the wall, as at Gissing, or recessed, as in their finest example at Little Saxham.

Whether arising from structural deterioration or from a desire to conform to contemporary architectural taste, there seems to have been a trend in the late 13th and 14th centuries for rebuilding or adding belfries with lancet openings. At Tasburgh, where the lower part of the tower is clearly earlier than the belfry, truncation of the upper tier of its encircling blind arcading suggests that the present lancet belfry replaces an earlier one, and at Blundeston and Fishley new lancet belfries were built on top of existing Norman ones.

Concurrently with these belfry rebuildings and additions, entire wholly circular towers with lancet belfries were built. Syleham’s tower, with its square panels of knapped flints forming a chequer pattern around the base of the belfry stage and the belfry openings, is a beautiful example. This use of knapped flints represents a further stage of development. It was not until the end of the 13th century that flint walls were faced with cleft or split-faced flints, and the early 14th before skilfully knapped flints were used, as in the wholly circular tower of Tuttington and later still at Wolterton. Before that, walling flints were used ‘as-found’ and the presence in a wall of any with a severed face would have resulted from the reduction of larger flints to manageable sizes, the breaking off of awkward projections or the inclusion of broken pieces.

The tower at Syleham also provides an instance of perhaps one of the earliest uses of medieval brick in a round tower’s fabric and in its putlog holes. After the departure of the Romans, brick-making in England ceased for about 900 years and was not revived in East Anglia, it is now generally believed, until the late 13th century. The discovery of medieval bricks as distinct from Roman ones is therefore a useful indication that the fabric cannot be earlier than post-Norman. In towers of the 14th century and later, bricks are often found in the flint fabric, usually in putlog holes.

  Round tower with octagonal belfry, Ilketshall St Andrew, Suffolk
  Ilketshall St Andrew, Suffolk. The two-stage 14th-century octagonal belfry is contemporary with the circular lower stage.
  Round tower with octagonal belfry, Bexwell, Norfolk
  Bexwell, Norfolk. The lower half of the circular stage is Saxon with a Norman former belfry surmounted by a 15th-century octagonal belfry.


A major innovation in the architectural development of round towers that occurred in the later 13th and early 14th centuries was the octagonal belfry built on a circular base stage. However, because of a persistently held belief in the antiquity of round towers, it has long been widely supposed that all octagonal belfries on round towers were additions to Saxon or Norman structures, and no serious consideration was given to the possibility that they could be contemporary with the circular stages. However, of the 55 round towers with a medieval octagonal belfry stage, many hold evidence to suggest that a circular lower stage and an octagonal belfry are contemporary.
Such evidence includes continuity of the internal circular shape into the octagon with no variation in the internal flintwork quality at the level of the external change of shape, as at Thorpe Abbotts and Ilketshall St Andrew. In several other cases, the circular stage has no evidence that can be reliably dated as earlier than the octagon but contains features that are probably contemporary with it, such as medieval brick in its fabric or a pointed tower arch. West Somerton and Wramplingham are examples of such cases.

In the 14th and 15th centuries the popularity of an octagonal belfry on a circular lower stage found further expression in the number of towers that had octagonal belfries added to them. In some, the octagon was built directly on top of the original belfry, as at Beachamwell, Bexwell or Needham, while in others a former belfry must have first been demolished, as at Bedingham where the upper part of the circular stage seems also to have been rebuilt. Virtually all the added octagons were given two-light belfry openings at the cardinal faces and some had flushwork replicas in the alternate facets.


Access to the upper levels of some round towers must originally have been by means of ladders and a trap door (as is often still the case), but in many churches there is an upper door from the nave in the tower’s east wall above the tower arch. This was presumably accessed via a ladder in the nave, perhaps until the advent of seating made it an inconvenience. The upper door can still be seen in the nave in many churches, while in others it has been blocked but remains visible inside the tower.

In a few of the later round towers of the 14th century, a newel stair was built as an integral part of the tower’s original construction like those at Shimpling, Rockland St Peter and Bardfield Saling. Subsequently in the 15th and 16th centuries permanent staircases of brick and flint were built onto some towers as external turrets, as at Needham or Stody, or formed internally in the tower, as at Yaxham, or in the nave, as at Witton. In some towers, a spiral stair appears to have been formed entirely within a tower’s wall thickness; in the Norman tower at Wissett the stair is clearly a later insertion because the apex of the Tudor arch of the stair entry in the south reveal of the Norman tower arch cuts into the tower arch impost. At Lound, the whole section of the wall enclosing the stair appears to have been rebuilt partly encroaching within the tower space. At Haddiscoe, the north arcade’s west respond and the tower wall were simply broken through to provide stair access from the north aisle.


After the Middle Ages many churches and towers suffered neglect and deterioration, and like those at Appleton, Burgh St Mary and Kirby Bedon St Mary, fell into ruin. Stumps of fallen towers still attached to their churches include those at St Julian Norwich and Feltwell. At Denton, following the collapse of the round tower in the late 18th century, a square tower was grafted onto a surviving curved section of wall still attached to the nave. Among many others to have undergone major restorations, Clippesby, Freethorpe, Roydon, Gresham and South Elmham All Saints have new upper stages, and the wholly circular towers at Ashmanhaugh, Weeting, Belton and Spexhall were entirely rebuilt in the 19th and 20th centuries. At Sidestrand, the whole church was rebuilt further inland with materials from the original cliff-top church which had been lost to the sea. Welford in Berkshire and Higham in Suffolk have new Victorian churches with round towers, the latter designed by Sir George Gilbert Scott.


The round tower church has demonstrated both longevity and versatility. Efforts over the past 30 years or so to examine the round tower churches of East Anglia more systematically have illuminated, albeit incompletely, the rich history of this fascinating and diverse body of English church architecture. By investigating the round tower church both separately and as part of a wider phenomenon we have been able to learn much about how and why they were constructed and how they were subsequently extended or adapted. Built into the very fabric of these structures is a record of shifting architectural tastes and styles, developing skills and technologies, changing methods and materials: an elaborate history of change and renewal.



Billet mould A Norman decorative moulding of alternating small cylinder shapes
Blind arcading Recessed blank arches as decoration on a wall
Flushwork The combination of knapped flint and dressed freestone laid flush with each other to create decorative patterns on flint walls
Impost The capital on a respond from which an arch springs
Knapped flint Flints that have been split and or trimmed to provide a flat face
Putlog hole A hole left in a wall during construction to accommodate scaffolding timbers
Quoin Cornerstone
Respond Termination of a length of wall from which an arch springs
Reveal The side face of an opening in a wall
Through-stone A long stone set horizontally on a central column, that supports the paired arches in the middle of a two-light Saxon belfry opening
Tower arch The interior archway between the nave and tower of a church

Recommended Reading

  • S Hart, The Round Church Towers of England, Lucas Books, Thorndon, 2003
  • S R Heywood, 'The Round Towers of East Anglia', in Minsters and Parish Churches: The Local Church in Transition 950-1200 (ed J Blair), Oxford University Committee for Archaeology, Monograph No 17, 1988
  • D Shreeve, The Round Tower Churches of Norfolk, 124 line sketches with
    descriptions by E M Stilgoe, Canterbury Press, Norwich, 2001

This article is reproduced from Historic Churches, 2008


STEPHEN HART BArch RIBA has made a study of round tower churches since retiring from his practice in 1989. His published works include Flint Architecture
of East Anglia
(2000), The Round Church Towers of England (2003), and Flint Flushwork: A Medieval Masonry Art (2008). He is a regular contributor to The Round Tower, the quarterly magazine of the Round Tower Churches Society (

Further information


Churches (general)


Advisory bodies and associations


Knapped flint
Site Map