�鶹��Ƶ

By 1200, the overwhelming majority of English parish churches had been established on sites they were to occupy until the Reformation and beyond, and a raft of papal decretals, episcopal ordinances, and dispute settlements had significantly circumscribed their character.¹��Moreover, if documentary evidence for the division of responsibilities for the fabric of the church is lacking before the 1220s, evidence from elsewhere in Europe suggests that by the end of the twelfth century the principle that in default it was the rector’s duty to maintain the chancel while the parish should maintain the tower and nave was gaining acceptance.²��Certainly, it is possible to talk of the parish church in 1200 and distinguish it from other types of church, even if there is a blurring at the margins between secular minsters that continued as small-scale collegiate churches and single-priest parish churches per se.³��One might also add that in the course of the twelfth century many local churches had been conveyed to ecclesiastical proprietors and were no longer held by the founder’s family or local landlord.⁴��As this had the effect of sharpening distinctions between renders owed to the parish and those owed to the manor, it seems likely that it heightened consciousness of the parish as an entity and a community. The pastoral chaos implied in the story of Stori—an English layman who owned land in Derbyshire under Edward the Confessor and is recorded in Domesday as being able to ‘make himself a church on his land and in his jurisdiction without anyone’s permission, and dispose of his tithe where he wished’—would have been unthinkable one hundred and fifty years later.⁵

The following paper explores how the parish church of c.1200 differed from its predecessors and seeks to evaluate its architectural repertoire in relation to other types of building. Were the choices exercised as to materials, plans, elevations, and surface detailing conditioned by an appreciation that the parish church called for a certain type of architecture? One that created clear divisions between the area set aside for the priest and that occupied by the laity, for instance, or which sought to shape the architecture to contrive distinctive and recognisable silhouettes? Secondly, on the understanding that the period under review wasn’t so constrained by restrictions on finance and expertise as to make the question irrelevant, did the norms of suitability change in the course of the twelfth century?

Long Sutton (Lincolnshire) is a good starting point, given the survival of a celebrated and unusually-informative charter dated to around 1180. By this, William, son of Ernisius, transferred his church at Sutton together with a site for the new building to Castle Acre Priory, giving them ‘three acres of land in Sutton in the field called Healdefen next the road to build a parish church there. And my wish is that the earlier wooden church in the same vill, in place of which the new church will be built, shall be taken away and the bodies buried in it shall be taken to the new church’.⁶��This transfer of proprietorship and site—and change of material from wood to stone—lies towards the end of a period which saw masonry parish churches come into being and lay owners transfer ownership to ecclesiastical institutions.⁷��It is also notable that the charter specifies that burials ‘in the church’ should be transferred to the new site (as might have happened in other contexts, but for which we do not have documentary confirmation).⁸

The church which emerged from this—with a seven-bay nave, two-storey elevation, and elaborate arcade—is a good example of a type of parish church that emerged over the second half of the twelfth century (Figs.��1 and 2). It is one that specifically singled out the nave for development, both its arcade and its exterior clerestory, though when Long Sutton was first built on its new site the aisles themselves would have been relatively narrow and dark, hence the subsequent overbuilding of the late-twelfth-century clerestory so as to roof the usual tall and wide late-medieval aisles.⁹

The date at which Long Sutton was built—the last quarter of the twelfth century—is a date by which an understanding of parishes and their churches had solidified.¹⁰��Parochialisation as a process had largely happened by 1180, even if there remained plenty of anomalies,��of what might��be regarded as leftovers from a minster-based system of pastoral care. At one extreme, there were still superior churches, whether collegiate or not, which administered pastoral care centrally through the thirteenth century. At Beverley Minster, for example, David Palliser has shown how a large parochia was administered from the minster into the late Middle Ages, with four town chapelries and three chapelries outside the town (all with dependent chapel buildings but reliant for baptism and the provision of priests on the Minster).¹¹��Or Pocklington (East Riding, Yorkshire), which to all intents and purposes functioned like an Anglo-Saxon minster without formalised collegiate status: a super-parish with a parochia consisting of ten dependent chapels until 1252, when Archbishop Walter de Gray ordained a vicarage at Pocklington and amalgamated the dependencies into four parishes, though the ultimate supremacy of Pocklington was maintained.¹²��All but three of those chapels retain evidence of twelfth-century work. Many of them are quite substantial buildings, though none, significantly, is larger than Pocklington.¹³��Bampton (Oxfordshire) was not dissimilar. Described by John Blair as ‘crypto-collegiate’, Bampton claimed jurisdiction over a large number of chapels and in addition to a rector enjoyed the services of three beneficed clergy, variously described as a ‘prebend-farmer’ and two ‘prebend-portioners’.¹⁴��Yet another variation is the post-Conquest prebendal church where more than one parish was served from a single principal building—a twin-rector, twin-parish church—as was the case at Grantham (Lincolnshire).¹⁵��William Rufus had granted the churches of Grantham to Osmund, bishop of Sarum, in 1092 and Bishop Jocelyn turned these into two prebends around 1150: Grantham Australis (with dependent chapels at Gonerby, Harlaxton, Cotsworth, Stoke, and Welby) and Grantham Borealis.¹⁶��Each prebend held a respective rectory and appointed a vicar, though the two parishes shared a single church—St.��Wulfram—a church that, as it happens, was rebuilt from a probably late-eleventh-century core around 1200 and given a vast new elongated late-twelfth-century chancel of a sort encountered below.¹⁷��The period between the Conquest and c.1160 does seem to have been one in which the institutional and architectural status of churches was particularly mutable.¹⁸

Given these variables, the extent to which status, ownership or history is embodied in the architecture of parish churches by 1200 isn’t easy to assess, though the question merits a different answer to what one might say of the situation at the beginning of the twelfth century. Then, at least as far as we can judge, the wealthiest post-Conquest parish churches were former minster churches that re-emerged as parish churches (rather than as monastic or Augustinian re-foundations of minsters), and churches established by bishops on their manors.¹⁹��Sometimes these overlap, as at East Meon (Hampshire), but good examples of the former would be the churches at Shifnal (Shropshire) or Castor (Northamptonshire) and good examples of the latter would be North Newbald (East Riding, Yorkshire), or the church that Roger of Sarum or his successor established in the new town of Devizes (Wiltshire).²⁰

All of the above were initially built or reconstructed as aisleless cruciform churches. It is a form that frequently leaves ghosts or residues when it comes to later medieval development. Take Bampton (Fig.��3). This clearly shows signs of its twelfth-century configuration as an aisleless cruciform church. The exterior silhouette remains strikingly cruciform, while on the interior it is evident that the crossing tower stair turret was an external feature, and that the later medieval transformation began with the construction of a vestry off the north side of the chancel c.1200.²¹

Needless to say, looking at parish churches from a plan-form perspective—with aisleless cruciform churches at the top of the pecking order—has its limitations. In part this is because we don’t have a sensitive grasp of the nuances of parish church design, of what meanings, if any, may be associated with linear plans nor, more importantly, as to the speed with which such meanings may have been diluted. There are also concrete objections. The rebuilding of former minsters to cruciform plans was not invariable.²²��The church at Ledbury (Herefordshire) had minster status before the Conquest but was incorporated into a fluidly-developing parochial system at some point thereafter and was reconstructed to a two-cell design with a partly-aisled chancel.²³��Bibury (Gloucestershire) shows no evidence of having been through a cruciform phase, though Bibury was an important Anglo-Saxon minster.²⁴��Norham (Northumberland) is a particularly grand example of a twelfth-century parish church in an episcopal town that didn’t conform to a cruciform plan.^25��Turning the relationship around, parish churches with no evidence of minster status could be built to aisleless cruciform designs, as with the ambitious de Braose financed parish church at Old Shoreham (Sussex).²⁶��The plan was also widely endorsed by smaller-scale collegiate churches and, most conspicuously, was used in early Augustinian and Cistercian churches.²⁷��Nonetheless some level of prestige does seem to have been attached to the aisleless cruciform plan that could be endorsed by high-status post-Conquest parish churches. The question is: when does this start to break down? It may be significant that Ledbury’s impressive part-aisled chancel, and the surviving south nave arcade and chancel at Norham, are works of the third quarter of the twelfth century. John Blair’s date of 1160 as the high-water mark of the cruciform plan surely puts this a little late.²⁸��Among grander parish churches, the alternatives were proliferating by then, and the most important of the new alternatives involved aisles; aisles and what goes with them, namely arcades (obligatory) and clerestories (optional). However, before moving on to consider aisles it would be as well to examine another desideratum that had already made an appearance in the eleventh century: the parish church tower.

Towers in aisleless cruciform churches are sited over the crossing, as long as there is a four-arched crossing, which is to say in the overwhelming majority of cases.²⁹��As this is immediately west of the chancel, it is questionable whether the tower belongs with the lay or the rector’s portion of the church.³⁰��Towers attached to the west end of the church are unambiguously associated with the nave, though it does not follow that their introduction was a lay initiative. David Stocker and Paul Everson have argued that the proliferation of west towers in Lincolnshire between c.1075 and c.1125 was ‘an expression of the new doctrinal order’, one that reflected the introduction of ‘the new, dramatic, symbolic liturgy of the reforming Roman papacy’ to the parish church.³¹��They claim that Bishop Remigius instituted a new form of burial service at Lincoln Cathedral that laid emphasis on a liturgical performance and vigil over the deceased body prior to burial, as Archbishop Lanfranc had done before him at Canterbury.³²��This was to take place in a separate space adjacent to the cemetery and was accompanied by the ringing of bells.³³��The result, they argue, was the addition of western bell-towers to a large number of pre-existing churches over a period of around fifty years. All of these have an upper bell-stage and can be entered from the nave, while virtually all have, or had, a west portal (Fig.��4).³⁴

There is no denying that the sixty towers identified by Stocker and Everson are sufficiently consistent to be considered a group, nor that they are post-Conquest in date and were built over two or three generations at most. Their social and economic context is notably varied, with around a third of the sample attached to churches built on public open spaces, and therefore likely to have been financed by groups of parishioners and local sokemen.³⁵��Others may have been built at the expense of the manorial tenant. Their distribution is also remarkable, consisting of local clusters with very few outliers. The overwhelming majority are in Lincoln and the north of the county, with a small group in the south-east.³⁶

This regional, or sub-regional, quality to the distribution of parish church towers in Lincolnshire can be found elsewhere. In the diocese of Exeter, for example, there are a significant number of churches with towers attached to their north or south sides.³⁷��They are scattered throughout Cornwall, but in Devon they are overwhelmingly concentrated in the north.³⁸��In East Anglia, a majority of eleventh- and twelfth-century parish church towers are attached to the west end of the nave, but differ from those of Lincolnshire in being round and having no external entry.³⁹��Most are in Norfolk, and even there, there are far more in the east than in the west of the county, with a notable concentration in the Waveney valley. The buildings of England estimated there is evidence for one hundred and forty round towers attached to the west end of parish churches in Norfolk, forty-two in Suffolk, and thirteen in Cambridgeshire and Essex combined.^40��The Round Tower Churches Society confine themselves to surviving examples and use post-1974 county boundaries, but they list one hundred and twenty-six in Norfolk, forty-two in Suffolk, six in Essex and two in Cambridgeshire.⁴¹��A cogent and detailed argument has been advanced by Stephen Heywood to date the East-Anglian round towers to between c.1070 and c.1200 and relate them to broader patterns of building around the North Sea.⁴²��This seems reasonable, as does the assertion that the peculiar mixing of Norman geometric ornament with so-called ‘Overlap’ forms is so ubiquitous that it might be regarded as ‘an autonomous cultural signature for the region’.⁴³��However, although studies of the East-Anglian round towers examine their form, date, position, and the materials from which they are built, none address how they were used. Unlike the Lincolnshire towers, East Anglia’s round towers lack west portals and can only be accessed from the nave. Thus, if one accepts the argument that the Lincolnshire towers are organised so as to facilitate the procession of the corpse directly to the cemetery, the East-Anglian towers either have nothing to do with funerals, or the burial rite, as presumably introduced to the diocese of Norwich along with the customs of Fécamp by Bishop Herbert de Losinga, was in practice different to that prescribed by Lanfranc for Canterbury.⁴⁴��If form and function were related, and it is the positioning of portals and, by extension, processional use that is at issue, there must have been significant variations in regional practice.

There are other examples of regional groupings of parish church towers, though these are less concentrated that those of Norfolk or Lincolnshire. In the East Riding of Yorkshire, square west towers seem preferred. In the south midlands there are a significant number of parish churches without transepts but with a tower bay between the chancel and nave, as at Iffley (Oxfordshire) or Stewkley (Buckinghamshire) (Fig.��5). In the decades to either side of 1200, the Fens witnessed substantial investment in the construction of freestanding bell towers.⁴⁵ The one constant is that parish church towers carried bells. But even here a tower was not a sine qua non. A bell-cote would suffice, even in such expensively finished parish churches as Kilpeck (Herefordshire).

Aisles are less susceptible to regional variation. The date at which these first appear in parish churches is unclear. There are numerous examples of narrow aisles known from excavation, though they are notoriously difficult to date unless there is evidence of the corresponding arcade. In most cases the suppressed aisle walls have been reduced to footings and, as their recovery is usually the result of nineteenth-century restoration, they tend to be poorly recorded.^46��Standing examples suggest aisles become popular from around 1140, though there are a handful of earlier examples. There is, for instance, a group of three churches in Buckinghamshire and Hertfordshire whose institutional status at the time their naves were aisled is uncertain, but whose arcades appear eleventh-century. These include Wing (Buckinghamshire), St.��Albans (Hertfordshire) and Walkern (Hertfordshire).⁴⁷��All are pre-Conquest foundations and retain elements of Anglo-Saxon fabric. At St.��Albans and Walkern these are single aisles. At Wing, aisles were added to both sides of the nave. If they were parish churches at the time aisles were added, then aisles were a feature of English parish churches before the Conquest. Otherwise, the parish church at Ickleton (Cambridgeshire) is probably the earliest standing example of an aisled English medieval parish church, dating from the last quarter of the eleventh century.^48��Ickleton formed part of the honour of Boulogne when the church was constructed, eventually passing via Eustace III of Boulogne’s daughter, Maud, to King Stephen, so it’s reasonable to assume high-level patronage.⁴⁹��But although the status and wealth of the patron may have facilitated the building, they don’t explain it, and if the counts of Boulogne had intended the church at Ickleton to do no more than serve the parish it would seem to have been exceptional.

By 1140–50, there is no mistaking the popularity of aisles. Former minster churches might be rebuilt on cruciform plans with integral nave aisles, witness Hemel Hempstead (Hertfordshire) (Fig.��6).⁵⁰��And there is a particular type of clerestory, designed on the half-beat and sunk into the spandrels of the nave arcade, that may have parochial associations. It is popular in the west midlands, as at Clun (Shropshire), and is the form used in the nave at Ewenny Priory, where a wooden roofed nave with a single north aisle was juxtaposed with a vaulted chancel and transepts (Fig. 7). Since there were parochial rights over Ewenny’s nave, the design highlights the different uses to which the two parts of the church were put.⁵¹��It is even possible this was a contrast in the service of architectural decorum, deliberated and understood in terms of acoustics and lighting, at least at Ewenny. Otherwise, the syncopated clerestory seems confined to west midlands parish churches. Ripple (Worcestershire) is a thirteenth-century example of the design type.

Though there are other desiderata—as with the towers discussed above—the aisle seems to be the critical area of development in the late-twelfth-century parish church. With aisles and lateral arcades, the highly-articulated longitudinal sequencing of space that characterises most late-eleventh- and early-twelfth-century parish churches starts to dissolve. Romanesque parish churches are almost by definition cellular. They consist of discrete spatial compartments in which the chancel is separated from the nave by an arch, or intermediate bay.⁵²��In most cases these are staggered, so the chancel is narrower than the nave. Where roof lines survive, chancels or apses are invariably lower than naves and are frequently vaulted. Kilpeck (Herefordshire) is a nice example (Fig.��8), or Kempley (Gloucestershire).⁵³��Lesser churches and chapels continued to be built to aisleless linear designs through the second half of the twelfth century. Indeed, there are instances of former minsters being rebuilt to relatively large two or three-cell linear plans as late as c.1180, like Blockley (Gloucestershire).⁵⁴��However, the half century between c.1130 and c.1180 saw a move away from the simple and the smaller-scale, and it is notable that the generation of parish churches that invests in aisles, like Hemel Hempstead, coincides with the generation that invests in large unvaulted chancels. Among early examples of the latter are at Cliffe (Kent) and Norham (Northumberland), both with aisled naves and twelve-metre aisleless chancels, and both likely built between c.1150 and c.1170 (Fig.��9).⁵⁵

Our first intimation that the parish church was not necessarily cellular comes at a similar date—around 1140–50, to judge by the style of the sculpture—in St.��Peter’s at Northampton.⁵⁶��This is the earliest example known to me of an aisled parish church without structural division. The church originally consisted of nine bays, with a rectangular east end that projects beyond the aisles (rebuilt in 1850–51) and an axial west tower that cut into the westernmost bay of the nave when it was reconstructed in the seventeenth century. The aisles were refenestrated in the late Middle Ages, though the clerestory remains substantially as originally built. Scott’s restoration respected the twelfth-century forms, reusing many of the original exterior clerestory capitals and corbels.⁵⁷��Notwithstanding slight variations in the distance between windows, this clerestory takes the form of a continuous arcade on the exterior, while the internal openings are cleverly arranged to both play against the rhythm of the arcade bays and maintain the autonomy of the clerestory (Figs.��10 and 11). In other words, there is an aesthetic dimension to St.��Peter’s, a concern to create a correspondence of parts that draws on an established Romanesque tradition of grouping supporting elements in twos and threes and integrating these with the elevation. It is an aesthetic hitherto unexplored in the English medieval parish church, precisely because it requires arcades and multi-storey elevations and prior to c.1140 these were largely confined to Anglo-Norman great churches.^58��Needless to say, the level of architectural detailing at St.��Peter’s is outstanding and is lavished across tower, nave, and chancel alike. Many of the columns are three-part monoliths with moulded annulets, the arcade voussoirs��mix limestone with ironstone, and the capitals are the work of one of the most distinguished groups of sculptors active in the midlands during the twelfth century.⁵⁹��Even the arch that leads into the tower is like a chancel arch, displaced and exiled westwards. This not only points to high-level patronage, it also points to the involvement of an architect with previous experience of the recondite world of Anglo-Norman pier design.⁶⁰

Despite the emphasis on open vistas at Northampton, the internal divisions are easy to read. The nave consists of three double bays, outlined by half-columns that rise to the top of the wall from quatrilobe piers (which in turn alternate with columns). The chancel is marked by three bays supported on columns and an absence of vertical articulation.^61��Whether the chancel and nave were originally separated by a timber screen is unknown.⁶²��There is no evidence either way. But if not, the subtle guidance provided by the elevation will have been reinforced by orientation and liturgical furnishing alone, for the steps that now lead into the chancel are post-medieval and the nave pavement has been lowered. The mid-twelfth-century floor was level throughout.

Quite what is driving this—why a hall-like building was created for St.��Peter’s at Northampton—remains an unanswered question. The church was unquestionably of high status. It was the successor to a minster known to have had a large parochia before the Conquest, with an advowson that swung between the Cluniac priory of St.��Andrew at Northampton (founded by Simon I of Senlis, Earl of Northampton) and the king in the twelfth and thirteenth centuries.⁶³��It was also built immediately to the west of what may have been a Mercian royal palace.⁶⁴��However, despite the pointers to elite patronage and an understandable tendency to ascribe the church to the second earl of Northampton, Simon II de Senlis (d.��1153), neither the immediate circumstances of its construction nor the intentions of its patron are known. Nor have its architectural sources been satisfactorily explained. It is unlike an antique basilica in that there is a western tower and the apse has been squared off. It is unlike a secular hall in that the aisles are low and there is a clerestory. It may be like a monastic building, or hospital, but none survive from the eleventh or early-twelfth century in this particular format.

Northampton has many of the features one associates with the most precocious late-twelfth-century parish churches—particularly those in the Fens—with a clerestory, varied pier designs, a long nave, and sculptural enrichment. Tilney All Saints (Norfolk) of perhaps 1180–90 comes closest, and like Northampton it dispenses with a chancel arch in favour of uninterrupted rooflines and continuous volumes (Fig.��12).⁶⁵��Tilney also has the virtue of retaining evidence for the original arrangement of windows at the east end of the chancel. Where the central vessel projected beyond the ends of the aisles, the clerestory continued and was joined by two additional windows at the level of the arcade (Fig.��13).⁶⁶��The arrangement compares with that found in East-Anglian Perpendicular churches such as Blythburgh or Southwold (both Suffolk), and as with these ensured that the eastern projection was the most brightly-lit area within the church.

However, in assessing the origins of the aisled parish church without structural division we may be missing a step. If one subtracts the bell tower, the prototype for at least one type of Fenland parish church, the type which retains a chancel arch, seems to be the monastic infirmary. The relationship is clearest at Walsoken (Norfolk), built in perhaps the 1170s with an aisled chancel, seven-bay nave, columnar and octagonal piers which alternate both along and across the nave, and a stupendous arch marking the chancel entry (Fig.��14).⁶⁷��The original twelfth-century clerestory even survives in the chancel, albeit now blocked. It relates directly to the new monastic infirmary at Ely, for which there is a terminus ante quem of 1169 thanks to Nicholas Karn’s publication of an episcopal charter granting the monks timber for work on their infirmary (Fig.��15).⁶⁸��Walsoken even quotes the Ely sanctuary piers (Figs.��16 and 17). This relationship between the late-twelfth-century parish church and monastic infirmaries is one that merits further exploration. Current understanding is that there was a boom in English monastic infirmary construction over the 1150s and 1160s, and that the linear three-part monastic infirmary, with a two-storey aisled hall, chapel, and sanctuary was a new building type, one that displaced earlier cloister or courtyard-based monastic infirmaries.⁶⁹��However, there may have been examples of in-line hall-chapel infirmaries before the middle of the century. The infirmary built at St.��Albans by Abbot Geoffrey de Gorron (1119–46) was described as an aula ‘cum capella versus orientem’ and may have been influential.⁷⁰��Hospitals and monastic guesthouses may similarly have been built as aisled halls with eastern chapels in the second half of the twelfth century, though the evidence is poor and points in too many directions to suggest there was a preference for any one plan type.⁷¹

Notwithstanding the origins of the building type, the reason for labouring Northampton, Tilney All Saints, and Walsoken—and one can extend this to Long Sutton—is that despite differences in their treatment of the relationship between chancel and nave, all four can be described in ways one might describe thirteenth- and fourteenth-century parish churches. They are over thirty metres long, with relatively thin walls, slender piers, aisles, two-storey elevations, and wooden roofs. Their nave arcades are perfectly adaptable and simply served as a platform for new clerestories and roofs. The difference between the parish church in 1200 and the same in 1400 strikes home when one looks at their exteriors, and is largely down to the status accorded to the aisles and the size of the windows.⁷²

These two features—that is, aisles and large windows—are, I believe, unrelated. At least, they don’t arrive together. Large windows are a rare feature in twelfth-century parish churches. The surviving Romanesque window opening at, say, Thornham Parva (Suffolk) is reasonably representative of a twelfth-century parish church window (Figs.��18 and 19).^73��And on the few occasions when large Romanesque windows survive they are invariably applied to aisleless spaces. The surviving examples at Iffley (Oxfordshire) or Stewkley (Buckinghamshire) are about as big as they come (Fig.��5).⁷⁴��The stunning aisleless chancel at Cliffe (Kent) would be another example. One certainly never finds original large window openings in twelfth-century parish church aisles. Romanesque parish church aisle windows are always small, as in the case of Compton (Surrey).⁷⁵��They are also low-set, because the aisles are low; so low, in fact, that where original aisle walls are retained from pre-thirteenth-century arrangements the aisles are frequently heightened. Compton rather nicely captures the difference, where the south aisle was heightened but the north was not (Fig.��20). The classic demonstration of this is the development of Castor (Soke of Peterborough) from a post-Conquest aisleless cruciform church (Fig.��21).⁷⁶��When the nave was aisled, probably around 1220–30, the upper nave wall was retained, but it was necessary to take out the lower masonry and cut through the twelfth-century transept west window, leaving a trace of the earlier roof line. By the fourteenth century, the desire for tall traceried windows was irresistible and the aisle was heightened.

So common was the practice of expanding aisles that exceptionally few late-twelfth- or early-thirteenth-century aisles survive in their original state as simple lean-tos with low parapets. Where they do, as at Little Faringdon (Oxfordshire), they tend to be narrow, though it’s clear that not all were (Fig.��22).⁷⁷��The important parish church at Redbourn (Hertfordshire) was rebuilt in the middle of the twelfth century with wide aisles, which retain a characteristically low-set Romanesque window head.⁷⁸��Width is much less of an issue than height.

So, what motivates the proliferation of aisles in English parish churches from around 1140 onwards? When looking into this question, scholars tend to return to the same potential wellsprings: altars, burials, chantries, population growth, great church envy, architectural grandeur.⁷⁹��Population growth must have been a factor, although a related phenomenon—the increased use of parish churches by parishioners for purposes other than attendance at the Mass—may also have been significant. In a case study based on parish churches in Warwickshire, Lindsay Proudfoot drew attention to a correlation between ‘the pattern of nave and aisle extension in parish churches and known national and regional population trends’.⁸⁰��The greatest growth in the combined floor area provided by nave and aisles took place between c.1250 and c.1340, when parochial (non-chancel) floor space increased by 48.6 percent.⁸¹��The increase between c.1200 and c.1250 was 8.7 percent. The appearance of aisles in parish churches from c.1140 onwards may therefore have been an early, and initially limited, response to the upswing in population. By c.1250, when a new type of aisle had been developed—one that was both taller and wider than twelfth-century aisles—it was easier to effect significant relative expansion in floor area. It may also reflect the hardening-up of the parochial map. Far fewer new parish churches were created between 1250 and 1340 than had been created in the twelfth century, so increased populations had to be accommodated in existing churches.

There obviously is a practical advantage to the addition of aisles, though there are significant qualifications to the statistical method.⁸²��Differences of spatial provision in churches in adjoining parishes of comparable size and economic potential imply there will be examples of both the over- and under-provision of space. And surplus wealth in the hands of individual parishioners, if channelled into the parish church, could lead to the large-scale expansion of a church, regardless of population size. There are examples of this in the later Middle Ages.⁸³ It may similarly have been a feature of the largely undocumented addition of aisles to parish churches in the decades around 1200.

In addition to the simple provision of space, aisles accommodated a variety of functions by the fourteenth century. Were these considerations in the adoption of aisles in parish churches in the second half of the twelfth century? Chantries one can rule out, at least as an instigator of the fashion. Chantries were often established in aisles in the later Middle Ages, but aisles come into being before there is evidence for chantries in parish churches, which on existing knowledge is around 1200.⁸⁴��As for burials, their impact is arguably best understood from Warwick Rodwell’s excavations of St.��Peter’s at Barton-upon-Humber (Lincolnshire). These encompassed the most comprehensive examination of parish church burials ever undertaken in England and enable one to compare the incidence and location of burials between the tenth and nineteenth centuries on a century-by-century basis.⁸⁵��If one examines the situation with regard to the addition of narrow aisles over the later twelfth and thirteenth centuries, there is a small group of burials in the first south nave aisle, but in comparison to the situation prior to the creation of the nave aisles it is hardly a step change.⁸⁶��As Rodwell remarks, ‘the incidence of indoor burial was low before the mid-thirteenth century’.⁸⁷��The later thirteenth-century widening of the south nave aisle resulted in many more south aisle burials than was prompted by the initial twelfth-century expansion. This suggests that burials certainly colonised aisles once they had been built, but that before the later thirteenth century the overall number of burials within the church remained roughly constant, with or without aisles.

Altars are more promising, and aisles are often equipped with them. It is clear that at Little Faringdon, there was an altar with accompanying piscina at the east end of the north aisle, where a vestry door was subsequently inserted (Fig.��23). It is equally clear that there was not an altar at the east end of the south aisle at Castor.⁸⁸ Aisles will certainly provide an altar emplacement that faces in the same direction as the chancel altar and doesn’t hinder movement by sitting west of the chancel arch, though technically there is no reason why an altar could not have been set against an aisleless nave wall.

Among formal, allusive, and symbolic criteria, great church emulation is, I think, a non-starter. Notwithstanding such oddities as New Shoreham or Hythe with their three-storey chancels, one mostly searches in vain for late-twelfth- or early-thirteenth-century vaulted nave aisles.⁸⁹��Parish church builders, or lay parishioners, do not seem to have been looking at great church aisles, or great church elevations for that matter. Indeed, the dialogue with great church architecture exists at a wholly different level of detail, and is seen in moulding profiles, pier designs, and sculptural embellishment. The c.1200 nave piers at Grantham are variations on Lincoln Cathedral designs. Its section or elevation emphatically is not.⁹⁰��The same can be said for many early-thirteenth-century parish churches in Nottinghamshire.⁹¹

Symbolic explanations, along with considerations of changes in liturgical practice, tend to founder on particularism, or the lack of it. There are both too many and too few parish churches that had adopted aisles by the first quarter of the thirteenth century for them to be the result of multiple individual patronal desires to create a church as an image of some other paradigm—like the Constantinian basilica—or a universal perception that a church without aisles was metaphysically deficient.

Finally, looked at from a formal and visual perspective, what is striking is not the aisle as seen from the aisle, but the aisle as seen from the nave. Aisles introduce arcades, and arcades introduce a level of architectural complexity. There may have been many reasons behind the burgeoning popularity of aisles in the run-up to 1200, but their capacity to reformulate an elevation by delivering an architecturally-distinguished backdrop must count high among them. One sees any number of examples of this: Kelmscott (Oxfordshire), Little Faringdon (Oxfordshire), the north nave arcade of St.��Mary at Barton-upon-Humber (Lincolnshire), St.-Nicholas-at-Wade (Kent) (Fig. 24).⁹²��As an approach to architectural renewal, a new aisle lies between refenestration and the significantly greater expense that would be involved in the complete replacement of the nave, but can still achieve a lot in visual and spatial terms.

None of the considerations aired above is exclusive, and it is highly unlikely that there was a single overriding reason behind the introduction of aisles. That may seem an unsatisfactory conclusion, but in the context of ecclesiastical architecture more generally it is hardly surprising. From as early as the fourth century one can see different functions allotted to aisles: the separation of the sexes; the provision of processional routes, particularly in funerary contexts; the enhancement of the central vessel with an arcaded frame, in turn enriched with or without textile hangings. Though the uses change over time, aisles never seem to have been reduced to a single primary purpose.

The aim of this paper was to evaluate the architectural repertoire of the parish church in 1200, particularly in relation to other types of building, and the evidence suggests that, by the end, a distance from the great church had opened. In part this was by default. The parish church espoused forms that great churches had left behind. Clerical communities no longer entertained aisleless and multi-cellular churches, while once prestigious forms—such as axial western towers—became the sole preserve of the parish.⁹³��More radically, the adoption of aisles brought new permutations into play. That this was an independent initiative is clear from the types of aisles that were introduced—lightweight, wooden-roofed, lean-to structures with low outer walls-aisles—which cannot be understood on great church terms. It is easy to overlook this last point, since it is rare for the outer walls of aisles to survive in their twelfth-century state. Their external elevations, the architectural potential of the aisle, was only realised in the thirteenth century with the arrival of tall windows. The status accorded to the aisle is the crucial difference between the parish church in 1200 and its late-medieval offspring. Nonetheless, as aisles brought a new spatial definition to the nave in the middle decades of the twelfth century, so a possible avenue for architectural development was opened, an avenue that seems to have been most enthusiastically explored in eastern England. With thin walls, slender piers, two-storey elevations, and wooden roofs, the late-twelfth-century churches of the economically-vibrant Fens might be regarded as the late-medieval parish church in embryo.