CHURCHES WERE the most ambitious buildings in the medieval landscape. Their designers aimed to provide an appropriate architectural setting for a sacred and theatrical liturgy and a glimpse of what heaven might be like. In pursuit of these goals a restless search took place for new architectural effects. As a result, the architectural motifs they used constantly evolved. These details tend to develop in broad phases known as styles, each with its own distinctive overall aesthetic quality. Knowledge of these styles can be extremely rewarding, enabling one to better appreciate these beautiful works of architectural art; to give an approximate date to entire buildings and parts of buildings; and to understand them better as monuments to the changing times and tastes of the distant past.
The aim of this book is to enable beginners to recognise these styles as they appear in England. To this end, I pick out in particular those details that are most foolproof and easy to remember, and that can thus be described as diagnostic when identifying a given style, while paying less attention to those that are harder to distinguish from each other. Learn these, and you are halfway there. The most straightforwardly diagnostic elements are often ornamental: the way foliage is carved, for example, or the patterns of tracery seen in windows. Other features, such as the shapes of arches and the stone patterns made in roof vaults, are also helpful for diagnosis; the mouldings with which all these things are articulated can, by contrast, be challenging for the beginner, and only the more straightforwardly diagnostic developments are addressed here. Indeed, some very simple mouldings are possible at any period. Definitions of technical terms, picked out in bold when they are first used in the text, will be found in the glossary where they are not self-explanatory.
These stylistic phases were first identified in 1817 by Thomas Rickman, and the names he gave them – for the Gothic era, Early English (EE), Decorated (Dec) and Perpendicular (Perp) – remain widely used. However, the masons who created these buildings were living in the present moment, as we do, simply attempting to create the most up-to-date and impressive buildings they could. They would not have recognised our stylistic categories, which depend on over five hundred years of hindsight. So a final and subsidiary aim of this book is to give the reader a sense of how these styles evolved ‘in the present’.
Two Norman windows at Berkswell, West Midlands, mark this church out as twelfth-century Norman; in the fourteenth century, a window in the Decorated style was inserted into the same wall. Such collisions of style and date give old churches a distinctive character and charm.
I also attempt to describe the overall ‘feel’ of a style, for the process of identification has a subjective dimension. Apparently subtle changes of detail can have a marked influence on the stylistic effect of a building, and the way in which motifs are used – the context of other forms with which they sit – can be as diagnostic as the naming of individual motifs in isolation. One needs to develop a sense of the flavour of a style.
It is a myth that it took centuries to build a medieval church. A small parish church took only a few years to build and usually went up in a single campaign, whereas a cathedral could be constructed in forty to sixty years. But progress might be halted in mid-build if funds ran out or wider events intervened. The longer such interruptions lasted, the more likely they were to have a visible impact on the design of the building.
Individual parts of any church might be rebuilt, or new parts added, however, and, when this happened, the style then current was usually followed. Windows (as well as tombs and other fittings) might be inserted into ancient walls, making a structure appear younger than it really is.
As a result, an old church might have developed over a long time, reaching its current form in a complex process that is intellectually stimulating to unravel. Ultimately this story can be fully understood only by studying any surviving documentary evidence and by bringing to bear a type of structural analysis known as buildings archaeology. This seeks to identify disjunctures in the fabric of a building, such as blocked arches or signs of one wall having been joined to another, and to study them to see whether they can be interpreted stratigraphically, that is, whether one part can be proved to be older than another. No matter how persuasive the stylistic or documentary evidence, the story it tells is likely to be flawed if it is contradicted by the stratigraphy of the structure itself. This important subject is beyond the scope of this book, but a list of further reading can be found at the end.
The dates given for the architectural styles in this book should be treated as rules of thumb. Ten years’ variance either side should be allowed for any date given as a circa (c.), for example, and one should be aware that an individual building may not fit the stylistic categories straightforwardly. In general, however, large ambitious buildings – great churches, see below, page 9 – as well as those smaller churches that happen to have been funded by exceptionally well-connected patrons, tend to be progressive, displaying early use of new motifs; out-of-date styles hold on longest in poor or isolated locations. But major buildings are also those for which documentary evidence of dating most often survives, and our chronologies have thus depended disproportionately on them. Some masons held on to old motifs (or even, on occasion, deliberately evoked the architecture of the past, or developed hard-to-classify new ideas); and in some parts of the country a given stylistic innovation might be ignored (see page 65). Exceptions to the rule are thus always possible. To confuse things further, Victorian restorers could produce deft, but misleading, versions of medieval style (see page 78).
In spite of this, the system outlined in these pages works surprisingly well. The science of dendrochronology is supplying us with accurate dates for ancient timbers used in smaller or undocumented buildings, and tends to confirm the accuracy of the broad stylistic phases that have been recognised for almost two hundred years.
It is essential to understand the plan of a church early in the process of exploring it. Though it rarely provides diagnostic dating information in its own right (for exceptions, see pages 16, 21 and 31), a church’s plan helps orientate the viewer and reveals much about a building’s development.
This parish church plan combines cruciform and basilican elements. Only the high altar is shown, but in the medieval period there would have been many more: a chapel in each transept, an altar in front of the screen which divided nave from chancel, and often others, too. The narrow windows in the chancel, the element sticking out to the right, suggest that it is the oldest part of this church. In this church, the tower might have risen over the centre of the building.
Almost all church buildings in England are axial, that is, they are longer than they are broad. The long axis runs east, towards the high altar, which is the building’s raison d’être. The arm of the church that contains this altar is thus called the east end, though its constituent parts have various names. Other parts are named accordingly: north and south transepts, north and south aisles, west front.
A typical great-church plan, combining axial and cruciform elements on a grand scale, with towers over the centre and at the west end, as well as a complex east end. A substantial stone screen, the pulpitum, separates choir from nave and transepts. As with the parish church, there are many side-altars, but here a dozen or so are set in separate chapels, positioned along the east-facing walls of the transepts and east end. No two buildings are the same, and not all have every feature shown here; in Romanesque great churches, screens are sometimes used to extend the space in which the choir sits under the crossing and some distance into the nave. The area west of the crossing is often still called the choir, though strictly speaking most of it will be the presbytery.
At its simplest, such a building may be merely a box, subdivided by an arch or a screen so that the area containing the altar, in this situation called the chancel (with the area immediately around the altar known as the sanctuary), is separated from the larger space, the nave. Such two-celled plans may suggest an early date and warn the observer to look for Anglo-Saxon or Romanesque (also known as Norman) features, especially if the chancel has a semicircular end, an apse (see page 21), which is almost always Anglo-Saxon or Norman in date.
Two more complex plans are also common: the most frequently encountered is known as basilican after the Roman meeting-halls used as a model for early Christian churches. Here, aisles run on both sides of the nave, and sometimes extend either side of the chancel, either stopping partway down it or extending as far as its eastern wall. Altars usually stood at the end of each of these as well as in other parts of the building.
Roman basilicas usually had an apse at the end of the main central space, explaining the early popularity of apsidal chancels. Their aisles were separated from the central space by rows of vertical columns, which by the Christian era supported arches, forming arcades. Turning briefly to the resulting elevation of the interior, the central space, often called the central vessel, was usually higher than the side aisles, so as to create space for clerestory windows: each arch had a window above it, the two together comprising a bay. This is the basic architectural unit of the interior of a church building.
The next most common plan-type is known as cruciform because a north–south arm called a transept cuts across the central east–west vessel of nave and chancel, resulting in a cross-shaped plan, with altars against the east wall of each transept. If these transepts are very long and there are no aisles, this type of plan is almost centralised – that is, the focus of the plan is in the middle of the church rather than at the end of one axis. However, the main altar was always at the end of the east–west axis, and this almost always predominates architecturally: the most truly centralised medieval churches in England are circular, and even these had a chancel extending from one side.
Duntisbourne Rouse, Gloucestershire, is a two-cell building with a stumpy tower and a porch at the west end. Most of this simple building was constructed at one time, probably in the late Anglo-Saxon or early Norman period.
The grandest churches often combine cruciform and basilican forms, combining them with a range of other features. The transept may have an aisle or aisles; there may be a second set of transepts further east, part of an east end that, in such buildings, can be very complex. Here, there will be many side-chapels, grouped in a variety of ways around the high altar (see page 7), and, along the main axis, such elements as (from east) a Lady Chapel; a hall-like retrochoir or ambulatory aisle; a feretory; a presbytery/sanctuary; and a choir. Such churches are often distinguished architecturally from ordinary parish churches by being called great churches.
There is a hierarchical quality to medieval art, vividly illustrated by the idea of the ‘great church’. It is important to be familiar with this concept as, in addition to the elements of their plans just mentioned, a great many stylistic motifs and other architectural features are most commonly seen in these churches, which in any case usually led the way in terms of the invention and dissemination of new motifs. It also helps make sense of the vast panoply of structures with which one is faced.
York Minster reached its current form in stages, resulting in a building with major works of the Early English, Decorated and Perpendicular periods. With three towers and a set of eastern transepts, and a combined cruciform and aisled (or basilican) plan, it is emphatically a great church.
The term ‘great church’ describes a range of architectural attributes that were possessed only by those churches that were home to wealthy religious institutions: in other words, the cathedrals (a church that is the seat of a bishop; such churches were also home to a community of monks or priests), as well as the larger monastic churches (inhabited by monks or nuns) and collegiate churches (the base for a community of priests). The architectural template for the great church developed over time (see pages 21, 31 and 74), but by about 1200 – that is, the dawn of the Gothic era – such buildings possessed or aspired to possess, in addition to a complex plan, stone vaults that covered the interior spaces of the entire church, and three-storey internal elevations, with an arcade at the lowest level, a gallery or triforium in the middle, and an upper clerestory. These were arranged into complex architectural compositions by the use of arches, mouldings, attached shafts and carved decorations. They also had complex east ends (see above, pages 7–8); elaborate façades, especially on the west fronts; and more than one tower (typically, two at the west end and one at the centre). While not every great church possessed all these features, many did, and most others had at least two or three of them. Great churches were usually designed by top-flight master masons who travelled widely in their work.
With its niches for statues and clerestory windows of alternating design, the elevation of the Decorated-era church at Cley next the Sea, Norfolk (left), is unusually elaborate for a parish church. Yet it could never be confused with Lincoln Cathedral (right), with its blank wall arcades, three-storeyed elevation and stone vaults.
In England, this window at St Nicholas, Galway would be called Curvilinear and dated to the 1330s–40s. But late medieval Scottish and Irish designers largely ignored English fashions, and this window is in fact early sixteenth century in date.
Parish churches possessed a very different character from great churches. This was the case even when their designer was clearly au fait with the latest ideas (he may indeed have come from a great-church workshop), or the church itself very large and elaborate; I have here sometimes called them ‘lesser churches’. In these buildings, stone vaults are rarely seen (if there are any, they will be in the tower, the entrance porch or the east end); and the internal elevation has at most two storeys: a clerestory and an arcade. The resulting composition is usually simple: a well-designed arcade with a row of windows above. The east end is also simple: a chancel, and chapels at the east end of any aisles. Façades are rarely elaborated and there is only one tower (sometimes at the west end, sometimes in the centre of the building, over the space between the transepts, or crossing).
Originally the distinction between greater and lesser churches was not as clear-cut as it is now. A huge number of smaller and mid-ranking monastic houses sat on a scale between the two poles, but these have been lost in disproportionate numbers. Some (Temple Church, London, is an example) created sophisticated great-church architecture, but on a small scale; others (such as Dorchester Abbey, Oxfordshire) made episodic attempts at work of great-church character, but could not afford to sustain such efforts, resulting in memorably idiosyncratic buildings. Nevertheless, that an architectural ‘glass ceiling’ indeed existed is demonstrated by the fact that, though more than eight thousand medieval parish churches have survived in England alone, only one – St Mary Redcliffe, Bristol – was built purely as a parish church and yet possesses almost all the defining features of a great church. Other medieval parish churches with great-church attributes were almost always originally the base of a religious community, and are often called ‘abbey’, ‘priory’ or ‘minster’ to this day. Likewise, some buildings that are today cathedrals were not built as such: for example, the cathedrals of Gloucester and Newcastle were originally a top-flight medieval monastery and a parish church respectively, and their architectural character reflects their original status rather than their current one: Gloucester is the equal of any cathedral; Newcastle has the very different character of a large urban parish church.
All these buildings looked very different in the Middle Ages. To most medieval people, bare, unpainted stone surfaces looked unfinished. Churches were usually whitewashed and brightly painted, especially on the inside. Religious scenes filled walls, especially around altars, and bright colours and simple geometrical or floral patterns enriched architectural features such as mouldings, columns, and capitals. There were many altars, all richly furnished, and most separated from the rest of the interior by screens. A vast number of images filled niches, covered wall surfaces and looked down from stained glass windows. Such features complemented the rich sounds and scents that accompanied the liturgy. Reconstructing what fittings, images and paintings originally filled a given church is a major project in its own right, in which painting, stained glass and sculpture each have a stylistic history of their own.
The Beauchamp Chapel (1439–62) at Warwick, though heavily restored, is one of the few places where one can experience the rich images and vivid colours that filled medieval churches.
Such features have largely disappeared, mostly as a result of iconoclasm in the 1540s and 1640s. Incidental sculpture and tomb effigies, neither of which could be interpreted as idolatrous, survive in larger numbers, if usually scraped of their colour; medieval fonts (and to a lesser extent screens) quite commonly remain in use. By contrast, a vividly coloured, life-size image of the rood filled the chancel arch and thus dominated the nave of every church in the land; not one has survived in its entirety. Most images in churches today are Victorian; medieval sculpture, painting, stained glass and woodwork are rare and precious.
A fifteenth-century rood screen at Cullompton, Devon, in the Perpendicular style. Such screens, topped by an image of the crucified Christ with St Mary and St John, dominated the interior of a medieval church; further screens subdivided the building, demarcating areas around altars. The screen was repainted in modern times, but the colours give a general impression of the original effect.
OTHER BUILDINGS, OTHER PLACES
This timber window of about 1460 at Gainsborough Old Hall, Lincolnshire, a grand secular residence, is Perpendicular in style. The surrounding timber-framing evokes the tight rectilinear patterns in which the style delighted.
Medieval buildings of all kinds, from tithe barns to castles, shared a single ornamental vocabulary. As a result, it is possible to apply the styles in this book to many non-religious buildings. Likewise, the stylistic phases outlined in this book are also seen in Wales and, more generally, and only until the later fourteenth century, in Ireland and Scotland – and, to some extent, across western Europe. By about 1400, however, many parts of Europe were developing distinctive national styles (see page 11), and the Scots and Irish variants were far more influenced by Continental developments than by English ones.
There is dynamism in Anglo-Saxon style, with its rows of windows separated by squat balusters, its bold patterns formed of pilaster strips, and its long-and-short work running in steps up the corners of a building: eleventh-century Earls Barton, Northamptonshire.