How to use this blog...

This blog allows you to explore some archaeological themes, periods and places. You can do this by: clicking the dates on the left to select particular posts; enter a term (e.g. Newgrange) in the search box below; scroll down and visit the Archaeology News section on the left.

Search This Blog

The Early Medieval Landscape - Ringforts and Crannogs in Ireland

Ringforts, Crannogs and the Early Medieval Landscape
(remember you can find lots of additional information at
• Due to the survival of extensive law tracts and historical texts we have detailed information on Irish society post-400 AD.
This means we can approach the archaeology of the period in various ways.
This lecture will look at the organisation of the landscape and its most visible settlement forms alongside the documented status of different social ranks.
This will begin by looking at the organisation of the landscape from open space (faithche) through to the ringforts and crannogs themselves.
The term faithche is difficult to translate but means, roughly, a ‘green’.
All of the main royal residences have a faithche.
The faithche may contain a bile (sacred tree) and a lawn (airscar) which was used for ball-games (liathróite).
It is usually extended as far as a bell or the crowing of a cock could reach.
The faithche could contain tilled fields (gort), hills (tulcháin), roads (slighti), margins (imli), recesses (cúla), hollows (cabána) and dark places (inada diamra).
Tilled fields named in text include corn (ith), woad (glaisen), vegetables (lub), orchard (aball) and hay (fer).
Achad and clúain refer to fields used as pasture.
Its not necessarily enclosed.
It was normal practice for four low grade farmers to employ a co-operative agreement (comar) to contribute an ox each to a plough team. This implies large open fields.
Below is a wooden plough from Oakbank crannog in Scotland.
Field boundaries
Four types of field boundary are distinguished in law:
Corae (stone walls)
Clas (bank and ditch), with a bank (múr) and trench (clas)
Nochtaile (literally the bare fence)
Dairimbe (literally the oak fence), probably a hedgerow
Field boundaries
Corae (stone walls) are described as being of ‘three stones’ (probably meaning three courses) with a width of three feet and a height of four.
Another text implies three stones at the bottom, two above and one on top.
The tool used to build it is described as a socc (an iron lever).
The bank and ditch of a clas should be three foot high and three foot wide and deep.
Field boundaries
A nochtaile was a fence that was strong enough to keep out animals, both by its strength and also by closing gaps to prevent, e.g. young pigs, gaining entry.
It appears to be mainly woven panels of wicker or wattle, apparently toppped with blackthorn (like barbed wire).
Field boundaries
Part of agreements between different social grades would include the requirement to contribute to public labour project such as building trackways (such as the one below) and maintaining fencing and roadways.
Field boundaries
There do not appear to have been formal gates (geata is a late borrowing from English).
Some fences had styles of some sort (céim)
Other types of fence are wall are described in law tracts but their nature isn’t clear.

Cush, Limerick
Fields are described as having two long and two short sides.
The long side is described as three forrachs and the spear cast of a youth.
Its implied that crops are grown on a raised bed up to eight feet in width, or seven sods of the plough, with a drainage trench in-between.
Cush, Limerick
The real size of fields is unclear as the term forrach is not clearly understood and it’s value may have changed over time.
Calculation vary greatly from around 144 feet by 72 feet to ten times that size.
This makes it difficult to determine the standard farm size.
Ballyutoag, Antrim
The airlise
In An Críth Gabhlach, the airlise is defined as the distance of a spear-cast on every side of a les (lios or fort).
In land-holding inheritance, property was divided equally except that the oldest heir received the houses, les and airlise.
Cush, Limerick + The Spectacles, Limerick
In the case of a kings residence the airlise also contained an area of assembly.
A synod of clerics could also meet in his airlise.
Corofin, Co Clare
An airlise is generally considered as containing enclosures for grazing and cultivation.
A bóaire was a rank of farmer who was supposed to own a kiln (áith) and a barn (saball), both within the airlise.
Some of the work carried out on this land (and beyond) employed slave labour (dóer).
A male slave was a mug and a female slave was a cumal.
St. Patrick herded livestock in woods and mountains in snow, frost and rain.
Male slaves are often associated with wood-cutting and carrying.
Female slaves are often associated with milking, churning, other dairy-related activities and preparation of grain.
The Bethu Brigte (although later) indicates that as the daughter of a slave she was a hereditary slave (even though her father was actually her master).
Sanas Cormaic on female slaves
The tenth century Sanas Cormaic gives the origin of the term cumal as coming from a Latin phrase ‘cum mola’ meaning a quern.
He then states that it refers to ‘… the woman who grinds at the quern, for this is the work which female slaves used to do before water-mills were made.’
The collar on the right is from Lagore in County Meath.
Horizontal Mill: Raystown
Base of mills, as excavated at Raystown.
Reconstruction of Raystown c. AD 900 (Simon Dick for CRDS Ltd)
Ringforts and social grades
The various early laws that deal with status record three grades of king, four of nobility and either five or six grades of free independent farmer, according the Críth Gablach or Cáin Aicillne.
To this can be added the fuidir, in a legal sense a ‘semi-freeman’ or ‘tenant at will’.
Effectively the fuidir were also slaves as there was no limit on the work they must provide to a lord and did not have any meaningful legal status, but, unlike slaves, they could severe the relationship.
A bothach (cottier) had similar status but may have had their own residence (rather than one provided by their lord).
Both grades became senchléithe after 3 generations (permanently bound to the lord).
Freemen: ranks
Fer midboth – generally appears to be a male teenager with some legal status
Ócaire - farmer
Bóaire – a wealthier farmer
Mruigfer – the richer bóaire
Fer fotlai – an aspiring noble
Relationships are defined between lords and clients (céile) who provide services such as direct labour or food-rent in return for land, legal protection, livelihoods (e.g. cattle) etc. Sometimes a clients fief became his property after 7 years (when he became ‘free’ client rather than a ‘base’ client).
But which grade built the ringforts and crannogs?
The flaith (cheiftains, nobles, lords etc) did not carry out any manual work and were divided into seven levels
Aire déso
Aire echta
Aire ardd
Aire tuise
Aire forgill
Tanaiste rí
King (Rí).
Kingship furthermore is divided into three classes:
of one tuath (tribe or kingdom)
three to four tuaths
Ócaire: lowest grade of farmer:
Expected to own land equivalent to seven cumals.
A cumal was a unit of measurement used in the early Irish law texts (e.g. An Críth Gabhlach, Uraicecht Becc etc). Technically, a cumal was a female slave, but the value equalled three milch (i.e. dairy) cows.
Had only one oxen and needed to make co-operative arrangements for ploughing (and needing later legal recourse to his lord to resolve disputes).
Carried out all the regular farm-work himself with his family.
May have also joined in co-herding agreements.
His food-rent included live animals, meat, grain, malt, bread, milk, milk-products and vegetables (a king and the church also had certain rights).
He also had to provide fixed tasks, such as joining reaping parties, wolf patrols, road works or helping build ramparts.
A man could have contracts with up to three lords.
An ócaire is unlikely to have been able to afford to build or live in a ringfort.
The construction of ramparts appears to be connected to the status of clientship.
Thus only freemen who had clients are likely to have had ringforts built as it was a duty of clientship to construct the bank and ditch which were the privilege of a ‘lord’.

Bóaire (and similar grades like the mruigfer):
Expected to own land equivalent to 21 cumals.
Possessed his own plough team (and so didn’t need recourse to his lord to resolve disputes).
He should own his own kiln and barn and have a share in a horizontal mill.
He still provided all of the same dues to his lord as part of his clientship, although he may have been able to substitute others for him in direct labour if he was rich enough (e.g. a mruigfer).
Bóaire (and similar grades like the mruigfer):
Should have a sheep-pen (sometimes implied to be a square structure).
A calf-pen
A pig-sty
Given the requirements on a bóaire it seems quite likely that they may have lived in a ringfort, although some commentators would contend that only actual nobles could construct a ringfort (due to the issue of vassalage).
Ringforts, a brief introduction
Normally considered to be dwelling places
Circular earth and stone structures that houses and huts were built within
Some had defensive features but they were residences and were not primarily built to withstand sieges or attack.
Only built to withstand small raiding parties until nearby help arrived
Ringforts (and cashels)
Banks made from ditch material – or ‘cashel’ walls from stone
Could have been used for controlling and protecting livestock – Rustlers – Predators
Enclosure was clearly a status symbol – Early Irish law indicates that a base client had to build and maintain the ‘Dún’ of his flaith (lord)
Enclosing the home: a territorial and privacy statement
Some protection from the elements
Ringfort distribution
Up to c. 50,000 ringforts known from maps and field survey
Most numerous archaeological monument in Ireland
Many still highly visible
M Stout believes broadly representative of Early Medieval settlement patterns in Ireland
Survival might have been greater in medieval Gaelic regions
Tradition, beliefs and superstition: ‘fairy forts’
Some studies of cropmarks in Leinster demonstrate destruction prior to OS c. 1840
Some of the best soil areas have low density of ringforts
Less common in mountainous areas
Ringforts: size matters
What is the due of a king who is always in residence at the head of his túath?
Seven score feet [43m] of perfect feet are the measure of his [internal] stockade on every side.
Seven feet [2m] are the thickness of its earth—work, and twelve feet [4m] its depth.
It is then that he is a king, when ramparts of vassalage surround him.
What is the rampart of vassalage?
Twelve feet [4m] are the breadth of its opening and its depth and it measure towards the stockade.
Thirty feet [9m] are its measure outwardly.
From Irish law text Crith Gablach c. 700AD

NB This may only be relevant for nobles not lower grades.
Ringforts: size matters
Irish law was schematic and probably aspirational rather than being strictly adhered to.
These measurements do conform [broadly] with the archaeological evidence
The average ringfort has an internal diameter is c. 30m
Some of those at royal centres like Rathcroghan are very much larger
Cashels (stone ringforts) tend to be smaller
Crannogs tend to be smaller, like cashels, while producing quite high status finds.
Univallate (i.e. one enclosing bank).
Example here is from Killyliss in Tyrone.
May be referred to as lios (also les) or dún in early sources.
It’s unclear which grade built these univallate sites.
Carn More, County Louth (from
Typical univallate ringfort (with souterrain) during partial excavation.
Bivallate (i.e. two enclosing banks).
Example here is from Lisnageeha in Tipperary.
Note the scale of the central area.
These are probably more clearly identified with term dún in early sources and are noble residences.
Multivallate (i.e. more than two enclosing banks).
Example here is from Garranes in Cork.
These are undoubtedly noble or royal residences.
Platform or raised ringfort
Example here is from Rathmullan in Down.
Note that the top of the ringfort is quite small.
Some have been excavated but their status is unclear.
How do they fit in?
Name derives from Irish word ‘crann’ meaning a tree (referring to the wooden elements of a ‘crannog’).
The terms ‘inis’ and ‘oilean’ are also found in texts (usually pre-9th century). Both mean island and often refer to crannogs.
Crannog dates
Ringfort/Cashel/Crannog chronology
Radiocarbon and dendrochronology dates from various sites used to provide general date range but all three of these site types are in use at the same time.
Most ringforts and cashels were constructed between 600-900AD (i.e. mainly pre-Viking)
Some were occupied through to the end of the Middle Ages
Status of crannogs?

Defence or Status Residence?
Ringforts: interior - Caherconnell
Hard to determine status of cashel sites.
Moynagh Lough
Moynagh Lough crannog
Occupied from as early as the Mesolithic
Major phase of occupation from around the 7th or 8th century AD.
Includes evidence for rich finds including metalworking areas.
Two houses found were 24 and a half feet and thirty-six feet in diameter.
Moynagh Lough
Various texts describe what building should be within the farmyard (lios) which we should probably equate to a ringfort or crannog.
The information given for a mruigfer and bóaire includes a house with diameter of 27 feet and an outhouse with a a diameter of 17 feet.
These correlate to the known dimensions of some early medieval houses.
Moynagh Lough
A mruigfer and bóaire should also own many objects including:
A cauldron with spit
A vat for brewing beer
Kneading troughs
A tub
Various farming tools
Farming equipment
He should always have a candle in a candle-holder and a fire burning all the time.
Craigywarren, Co. Antrim
Typical finds from the residence of a mruigfer or bóaire?

Craigywarren crannog, Co Antrim
Moynagh Lough
The houses of lords are only slightly larger than that of a mruigfer.
An Aire Déso had a house of 27 feet similar to a mruigfer.
The highest non-king grade of noble (Aire Forgill) had a house of thirty feet in diameter (with eight cubicles).
The house of a king should be thirty-seven feet in diameter (like Moynagh Lough) and with twelve bed cubicles.
Lower grade housing
An ócaire was supposed to have a house of nineteen feet diameter with an outhouse of thirteen feet (where his food-rent was divided up).
Texts also distinguish another grade of house, the tech nincis, which (in the instance described) was provided for a fosterson who undertakes the work of an elderly landowner who had no family support.
Lower grade housing
The tech nincis was:
Seventeen feet in diameter
Constructed of wattle from the ground level to the roof pinnacle
From the level of the lintel (i.e. top of the door) there was a layer of feathers between every second band of wattling (apparently between the wattling and the reeds or thatch)
There were two doorways – one to the outside and one to a larder (cuile).
Half of the interior seems to have been a bed cubicle whilst the other half was paved.
Deer Park Farms, Antrim (during excavation)
Some houses were around 17 ft in diameter
‘…constructed of wattle from the ground level to the roof pinnacle’
Deer Park Farms
Collapsed wattle extended above the height of the door and became the roof.
Deer Park Farms
Note wall cavity between inner and outer wicker panels.
Central hearth is also present.
Deer Park Farms
These houses were dated to the seventh century AD and appear to represent a tech nincis as described in the early texts.

An ócaire was supposed to have a house of nineteen feet diameter with an outhouse of thirteen feet (where his food-rent was divided up).

Deerpark Farms, Co. Antrim
Ringfort 25m in diameter (i.e. relatively low status)
Stone-paved entrance at E directly to door of main 7m building from 8th century AD. The size of the building is quite small for a grade like mruigfer.
Visitors to the lios are directed into a main public space like a parlour.
Figure of 8 layout: also known from ritual sites and conjoined Iron Age burial sites but does it represent the two distinct houses of differing diameters required by freemen of particular grades.
Inside the Ringfort: Case Study of Deerpark Farms, Co. Antrim
Deerpark Farms: unusual preservation of internal structures
Site became waterlogged: anaerobic preservation of organic materials – especially wooden structures
Evidence for c. 25 houses within central area (but only a handful standing at any one time)
Collapsed old house roofs, frames, walls etc. were left in situ and new ones built above these
Site became elevated as a result of this: like a ‘tell’ site
Water table rose and preserved organic material similar to what is normally uncovered at a crannog.

An ócaire was supposed to have a house of nineteen feet diameter with an outhouse of thirteen feet (where his food-rent was divided up).

Or is it a tech nincis:
There were two doorways – one to the outside and one to a larder (cuile)?

Not the residence of a mruigfer or bóaire (the houses are two small) or an ócaire (too rich).
Is it occupied by a number of ócaire?

Bivallate Ringfort at Hughes’ Lot East, Cashel, Co. Tipperary
Approximately 18% of ringforts are bivallate
Whilst the status of the univallate sites is unclear, multivallate sites, starting with bivallate ringforts are undoubtedly the preserve of nobles.
They also tend to be spaced further apart (as you would expect if the rely on local ‘clients’).
Early medieval landscapes
Study of overlapping ‘visual territories’ in Braid Valley in County Antrim
All ringforts were in this study had visual ‘contact’ with at least one other – Some with as many as 17 others
Does this suggest their positioning is co-ordinated?
Ringforts, landscape and society
South-west midlands model based on results of detailed study (by Matt Stout).
High status bivallate ringforts lie close to church centre
Provides focus and possible defensive support for smaller upland (univallate) ringforts
Large but lesser status multi-functional ringforts are focus for other clusters of low status ringforts
And might provide resources or specialist services (e.g. metalworking) for high status site
Ringfort size, number of enclosures, distance from others, proximity to important facilities (e.g. church or road), artefacts produced etc. said to represent hierarchical society
Ringforts, landscape and society
Study of Clogher ringforts (Tyrone) and environs again noted proximity of ‘royal’ ringfort to church, major roads, and rivers
Like Early Irish Law: very schematic model
But accords well with early texts and with ringfort and church distribution studies
Prior to Vikings Irish economy and settlement was overwhelmingly rural
And dispersed farmsteads rather than villages
Or ‘dispersed villages’