>’…and as many more in the adjacent Isles…’


The stimulus for this piece came from the ‘Parliamentary Abstracts; Containing The Substance Of All Important Papers Laid Before The Two Houses Of Parliament During The Session of 1825′.
In a table introduced by the sentence; ‘The following list shews the places at which churches have been directed to be built; most of them absolutely, a few provisionally:’ , I noticed that in the Parish of Harris on ‘Berneray Isle’ a church was to be built for the population of 500:
And as many more in the adjacent Isles of Pabbay and Killigray.’
Reading that, in 1825, the population of these three islands in the Sound of Harris was estimated to be 1000 souls I wanted to investigate further. Although a decennial census had been introduced in 1801, the first four of these only provide a figure for the population of the whole Parish.
For Harris, these figures were:
1801 2996
1811 3569
1821 3909
1831 3900
Our year, 1825, lies neatly between two censuses in which the population, despite all the displacements that were occurring, remained remarkably stable at circa 3900 people.
Thus the 1000 estimated to be living on our three islands were about one-quarter of the parish’s people reminding us that ‘Prior to the nineteenth century, the majority of the population of Harris lived on the machair of the west coast and on Pabaigh and its neighbouring islands (Berneray/Beàrnaraigh, Ensay/Easaigh and Killegray/Ceileagraigh)’ http://www.paparproject.org.uk/hebrides2.html
As an aside, we have this communication from the 18th of July 1832 which I think is illuminating.
The later censuses do provide figures for each island in the Parish of Harris and those for the years 1841-1871 are given below. I have shown the number of males and females and computed the average ‘people per hearth’ for each island with the trio of isles that are our focus shown in bold:
1841 – 7th June
Anabich 18 males and 23 females in 7 houses (41/7 = 5.9 people per hearth)
Bernera 335 males and 378 females in 130 houses (713/130 = 5.5pph)
Ensay 7 males and 9 females in 2 houses (16/2 = 8pph)
Hermitray 5 males and 3 females in 1 house (8/1 = 8pph)
Killigray 3 males and 2 females in 2 houses (5/2 = 2.5pph)
Pabbay 179 males and 159 females in 61 houses (338/61 = 5.5pph)
Scalpay 14 males and 17 females in 4 houses (31/4 = 7.8pph)
Scarp 60 males and 69 females in 23 houses (129/23 = 5.6pph)
Tarrinsay 38 males and 50 females in 16 houses (88/16 = 5.5pph)
There were 1056 living on our three islands which was almost 23% of the total of 4646 people in the Parish of Harris.
Five years later the first of the Potato Famines occurred and the response of the Factor can be seen in his letter of the 21st August 1846 to the Countess of Dunmore.
1851 – 31st March
Anabich 63 people in 12 houses (63/12 = 5.3pph)
Bernera 452 people in 89 houses (452/89 = 5.1pph)
Ensay 14 people in 3 houses (14/3 = 4.7pph)
Hermitray Uninhabited
Killigray 7 people in 1 house (7/1 = 7pph)
Pabbay 29 people in 6 houses (29/6 = 4.8pph)
Scalpay 282 people in 48 houses (282/48 = 5.9pph)
Scarp 145 people in 29 houses (145/29 = )
Tarrinsay 55 people in 11 houses (55/11 = 5pph)
Only 488 living on our three islands which was less than 12% of the Parish total of 4254.
Nine out of every ten people from Pabbay and one-in-three of the population of ‘Bernera’ had gone.
Just four days after the census, on the 4th of April 1851, the Factor John Robertson Macdonald in ‘Rodil’ was being ‘interrogated’ by Sir John McNeill and an earlier piece analyses his account.
We should also note the dramatic increase in the population of Scalpay that had occurred, the reasons for which are to be seen in this investigation.
1861 – 8th April
Anabich Not listed
Bernera 130 males and 185 females in 64 houses (315/64 = 4.9pph)
Ensay 10 males and 5 females in 2 houses (15/2 = 7.5pph)
Hermitray Not listed
Killigray 2 males and 3 females in 1 house (5/1 = 5.0pph)
Pabbay 10 males and 11 females in 4 houses (21/4 = 5.3pph)
Scalpay 199 males and 189 females in 71 houses (338/71 = 4.8pph)
Scarp 72 males and 79 females in 27 houses ( 151/27 = 5.6pph)
Tarrinsay 25 males and 30 females in 12 houses (55/12 = 4.6pph)
There were just 341 living on our three islands or about 8% of the 4174 people of Harris.
Once again, almost one third of the remaining people of Bernera had gone leaving just under half the hearths from the 130 of two decades earlier.
1871 – 3rd April
Anabich Not listed
Bernera 169 males and 204 females in 75 houses (373/75 = 5.0pph)
Ensay 4 males and 2 females in 1 house (6/1 = 6pph)
Hermitray Not listed
Killigray 3 males and 6 females in 1 house (9/1 = 9pph)
Pabbay 3 males and 5 females in 2 houses (8/2 = 4pph)
Scalpay 222 males and 199 females in 82 houses (421/82 = 5.1pph
Scarp 78 males and 78 females in 33 houses (156/33 = 4.7pph)
Tarrinsay 35 males and 33 females in 12 houses (68/12 = 5.7pph)
A small increase to 390 living on our three islands but still only just reaching double-figures again at 10% of the the people of the Parish.
Bernera’s population had risen by 18% but the island trio would have needed nearly three times as many residents to regain the proportion of the population that had led to the church being built there only four-and-a-half decades earlier…
Note: I have left all spellings as they appear in the original sources, except that those for the census lists are ‘standardised’ from the 1841 census rather than reflecting the variations that appear in some of the subsequent decades.

Kerr Occupations 1841-61 – A Regional Comparison

The figures in brackets are the number of Kerr folk found at each location:

1841 Harris (65) Other Inverness (29) Ross (28) Argyll (192) Sutherland (242)
Carpenter or Joiner 2 1 0 1 1
Mason 1 0 0 0 0
Shoemaker 2 1 0 5 0
Tailor 1 0 2 0 0
Weaver 0 0 0 1 1
(The one Tailor on Harris appears as a ‘Tenant’ on the census but I know my great, great, great grandfather
John was in fact a Tailor at this time – it demonstrates that this data is subject to all manner of ‘noise’ – and the same is true of the one Mason who was Peter, the Dry Mason who moved to
Argyll between 1851 and 1861)
1851 Harris Other Inverness (46) Ross (23) Argyll (255) Sutherland (171)
Carpenter or Joiner 2 0 0 4 1
Mason 1 0 0 2 0
Shoemaker 2 0 0 4 0
Tailor 1 0 1 0 0
Weaver 1 0 0 2 0
(Also a ‘Teacher of English’ on Barra and a Farmer’s Widow in Borve)
1861 (52)

Harris Other Inverness (69) Ross (43) Argyll (251) Sutherland (225)
Carpenter 1 0 0 1 0
Mason 0 0 0 1 (b. Harris) 0
Shoemaker 2 0 0 1 1
Tailor 1 0 0 1 1
Weaver 2 0 0 1 0
I think it is clear that the concentration of these particular occupations in the small population on Harris demonstrates a cluster compared to the wider region.
The table also suggests that, if these records in any way can be used to reveal a trail left as crafts were passed from father to son (although I have included one Weaveress from 1861!) then Argyll is the most likely source for that trail.
I know this contradicts my recent focus upon Sutherland but I hope it also demonstrates why I am looking
outside Inverness and Ross as likely origins for the Harris folk?
My quest continues…

The Age Gap – or, where are all the 60 & 70 year-olds?

In 1841 there were no Kerr folk on Harris in their 60s or 70s. None.
In the surrounding areas there were between 8% and 12% of such people in their populations which, if translated to the situation on Harris where there were 65 Kerrs, means that we would expect to see perhaps half-a-dozen people of that generation.
Instead, there is a gap between a few in their fifties and one lady, Chirsty of Taransay, who is said to be 80.
If they were a settled population going back a couple or more generations then we would surely expect at least one person in this age group to have been present? I think this is clear evidence that those in their 40s and 50s (born 1781-1801) amongst the population of 1841 were in fact the first generation to be born on Harris and that it was their parents who came to the island attracted by the call for craftsmen that went out when Macleod began his Improvements in the 1780s.
These parents would have been of Chirsty’s generation (she herself may have been the mother of Roderick of Rha and others) and thus it is less surprising that they were no longer present by 1841. My own ancestors, Malcolm Kerr and Effie Shaw, parents of John and Angus, were of this generation and are the only ones whose children survived, or remained on Harris, beyond 1855 and the introduction of Statutory Registration. Hence my being able to identify them from their sons death certificates. If they had any daughters then, sadly, I would have to examine every female death of those born, say, 1780-1820 from 1855 onwards in the hope of discovering them beneath their married names. The same is true for the others of this ‘original’ generation.
A task too far!

If anyone can suggest another explanation for this unusual feature then I would be delighted to hear from you, but I think that mine is the simplest to fit the known facts, including the activities being undertaken on the island at the time?
“Captain Macleod also encouraged shoemakers, weavers, turners, wrights and masons to settle down at Rodel.” (John Lanne Buchanan quoted in ‘Travels to terra incognita’, Martin Rackwitz 2007)

“he alfo encouraged a great many artificersas fhoemakers, weavers, turners, and wrights, and mafons” (John Lanne Buchanan in his ‘Travels in the Western Hebrides: from 1782 to 1790’)

Note: There are two records from the 1851 census that show ages of 74 and 77 (and therefore people who would have been in their 60s in 1841) but the one who survives until the 1861 census resorts to a birth in the late 1780s so I think the ‘gap’ as described is real rather than merely an artifice of the records.

Population Table – Kerrs of the West Coast

Note: Columns are for Inverness-shire then Harris within that county, Ross-shire and Lewis within that county; for Inverness-shire & Ross-shire combined and the number & proportion deriving from Harris & Lewis. These are followed by the numbers for Argyll-shire and Sutherland-shire and the total for all four counties:

Census Inverness-shire Harris Ross-shire Lewis I&R H&L % Argyll-shire Sutherland-shire Total













































































(I have prepared charts from this data but am struggling to get ‘Blogger’ to accept them!)
I have restricted my examination to these four counties because, as can be seen on this map ,  they form the mass of the (Gaelic-speaking) West Coast of Scotland. Caithness did support a small population and their occupational range is quite similar to that on Harris but the pattern of names is different. A similar result was found for Argyll-shire.
The firsr feature that strikes me is that we start with over half the population of Inverness-shire and Ross-shire being due to those on Harris. This strongly suggests to me that the name was not native to those counties and hence we need to look further afield.
The figures exhibit greater synchronicity between Sutherland and the Inverness/Ross population than do those for Argyll which form a cyclical pattern. There is also greater similarity amongst names, such as Angus and Roderick, in the three counties with others such as James and Peter appearing in Argyll. (Interestingly, there was a Peter in Harris in 1841 but he later moved with his family to Argyll). If male names form a chain back into history, then  there is stronger evidence suggesting that it snakes back to Sutherland than anywhere else.
All this has to be taken with a ship-load of salt but I am leaning towards the idea that a family or two moved from Sutherland to Harris (including Taransay) by the early 18thC and settled there but that a range of circumstances (Clearances, Emigration, non-marrying males & a lack of male heirs) led to the name’s demise on Harris.
Those who left the island, whether to Canada, America, Australia, England, Argyll or, indeed, Lewis, were slightly  more successful in continuing to add links to the chain…

Update: This PDF makes the point rather well – it ‘feels’ like the Harris families, but in Stoer (and a few other places in Assynt, Sutherland) and with a larger overall presence in the population: http://rogart.fileave.com/HMD%20MARRIAGES.pdf

Sutherland, perhaps?

In attempting to discover where the Kerr families of Harris may have originated, I have been re-examining the records for the four counties of Inverness-shire, Ross-shire, Argyll & Sutherland and that remains very much a work in progress as I have yet to analyse the data from Christian names.

However, this afternoon a friend alerted me to some evidence given to the Napier commission by Murdoch Kerr, 55 a Crofter’s Son, formerly a Fisherman, of Auchmelvich in Sutherland.
His testimony paints the usual bleak picture which Murdoch presented to the commission in a written document titled ‘Auchmelvich Township Grievances’ but it is the following exchange that caught my attention (Murdoch’s replies are in italics):

27631 You name is rather uncommon? – It is a strange name in the place.
27632 Are there several more in this place of Auchmelvich? – Yes.
27633 Do they all belong to the same clan? – No, they are separate families.
27634 Are they long here? – My ancestors have been here for seven hundred years, the Kerrs to whom I belong.
27635 What family were in possession of Assynt at that time? – I cannot tell.
27636 Do you, and the people in your place, look with favour upon the large sheep farms? – We would rather not see any in the country.

(I included 27636 because of its comprehensiveness – they did’t want such farms anywhere in the land!)

So, what does this brief exchange have to tell us?
Firstly, two years earlier the 1881 census recorded 32 families in Sutherland headed by a man called Kerr and  3 of these were Lowland families which accords with Murdoch’s comment that they were not all of the same clan but represented different families.
Secondly, at least one branch of the Sutherland Kerrs had been there since the 12th or 13thC, a remarkable fact that is testament to the oral tradition in Gaelic communities.
I don’t know if the Gaelic Harris Kerrs originally were from Sutherland or Argyll, Ross-shire or Inverness-shire (or indeed elsewhere in the Highlands) but Murdoch offers the tantalising thought that a left-handed Highlander may have excelled himself on the field of battle in the Medieval period and established a line that leads all the way down to the left-hander typing these words…

(Note: I apologise for that brief lapse into Romanticism. It may happen again…)

Update: Almost two-thirds of the Kerr folk in the censuses living in Sutherland were living in Stoer and some 80% of households in the county headed by a male Kerr were there too.
Stoer is less than 3 miles from Murdoch’s home in Auchmelnvich.

Note: Full Transcript can now be read here.

Widow’s Row(s) of Stornoway

Having stumbled upon this excellent letter in the Stornoway Gazette of the 15th April 2004, I felt a return to the topic of Widow’s Row(s) in Lewis might be of interest:

1851 90 in 27 (3.3)
1861 56 in 16 (3.5)
1871 97 in 31 (3.1)
1881 38 in 14 (2.7)
1891 None Listed
1901 22 in 9-Newvalley (2.4)

The correspondent states that in 1851 there were 22 families at the Mossend ‘Widow’s Row’ so it may be that 5 of the 27 shown above for that year were elsewhere for, as he says, “The Sandwick Widows’ Row was but one of a number of such named settlements in Lewis.”

All the ‘Widow’s Rows’ were in the Parish of Stornoway but the only one given a specific address is the Newvalley one in the 1901 census, and there is no mention of any others in Stornoway in that year.

The figures which strike me are the average number of people per household. They range from 3.5 down to 2.4 which is well below the norm and a sad reminder that, in many cases, these were people living-out their final years far (both geographically and socially) from the heart of their communities.

Fishers of Harris

Another somewhat crude interrogation of the censuses, this time looking at males recording ‘fisher’ or ‘fisherman’, but removing those who were a fisher or fisherman’s son.

1841                 49
1851 230-56=174
1861 340-66=274
1871 517-44=473
1881 630-64=566
1891 492-06=486
1901 498-12=484

1841, it should be remembered, is unreliable in recording all the occupations of occupants at an address so we can only properly examine the second-half of the century.

The numbers of fishermen increased by 57% from 1851-1861 and by a staggering 73% in the following decade. Growth slows to a more serene 20% during the 1870s but the result is that for every 4 fishermen in 1851 there were 13 by 1881.

The 14% decline from this peak to the figures of 1891 and 1901 (which is the only decade to demonstrate stability) might reflect, in part, the effects of re-crofting on Harris but I cannot be sure of the importance of this factor. If nothing else, these figures echo the phrase from the evidence to the Napier Commission of men ‘turning their backs on the land to face the sea’…