/ UiT Norges arktiske universitet

Insights from inferences: census taking and social dynamics in Tromsø 1865

By Michael Drake

(This article was presented at the seminar of the Institute for History, the University of Tromsø, and is published for the first time here.)

The census of Norway, taken in 1865, has two notable features. First, it marked a return to the nominative mode of census taking, where characteristics (sex, age, occupation etc.) of each named inhabitant are collected. This was not the first census of this type in Norway. That had taken place - successfully - in 1801, but the method was then abandoned The second feature of note was quite new. For in the towns the 'census forms were distributed to all houseowners and completed by them' (NOS :1). It was thought the innovation would have 'two benefits, to arouse greater general interest in the census, and, at the same time, reduce the work-load on the enumerators' (NOS :1). Three questions suggest themselves. First did the innovation go according to plan? Second, if it didn't then why not? Third what can we infer about the social dynamics of a town, by what occurred?

Some 390 census forms were completed for the town of Tromsø. On the first side of each form was the signature of the person who had completed it. Table 1 shows an analysis of these signatures. It is apparent that Tromsø's experience was quite different from that in Norway's capital, Christiania. There most of the forms were completed by the relevant houseowner or household head. In the comparatively few instances where this was not the case, the task was undertaken by a police constable, or one of the town's inhabitants who helped with the collection of the forms (NOS :1).

Table 1 Signatories of completed census forms, Tromsø, 1865.




No signature



Household head



Member of household other than head









Source Census Tromsø 1865. Photocopies of original forms, Statsarkivet, Tromsø

In Tromsø we know of only two houseowners who completed the form. In the one case the form was signed by J.H. Ludwigsen as 'owner of the house' for a house where the 'manager of Ludwigsen's spirits' store' was the household head. In the other case a master bricklayer signed the form for a house he owned in Nordre Strandgata, and of which he was also the household head (RHD, 17.29 and 67.11). But what of the others who signed? Were they spread haphazardly around the town or was there some sort of pattern to this distribution. The answer seems to be - a bit of both.

Map of Tromsø in 1875

The main settlement area for people who completed the census form, lay around the Church to the north side of Søndre Strandgata, the south side of Storgata from Strandskillet to Søndre Kirkegata and Søndre Kirkegata itself. In that part of Søndre Standgata which lay to the east of the Church, all the forms were completed by the head of the first household listed in each house. North of the Church, that is to say the eastern side of Nordre Kirkegata and the western side of Søndre Standgata, only one third of the household heads signed the forms. The north side of Storgata from the Grand Hotel to Elvegata (quite a long way) and the east side of Nordre Strandgata, were two other parts of the town where many household heads signed to indicate they had completed the forms. By calculating the percentage of those completing the form in each section of the town (as delineated by the town development plan), a more precise view of the situation could be obtained.

Table 2 Taxable income (1866 ) of 'household heads in the first household' of each house in Tromsø in 1865, according to whether or not they completed the household schedule.

Taxable income(spd) 1865 census schedule completed by:
First household head in house Non-member of house
3000+ 4 4
2000-2999 5 5
1000-1999 12 12 4 2
500-999 30 30 8 3
200-499 35 35 53 23
80-199 12 12 112 49
Under 80* 2 2 52 23
100 100 299 100

*This is a guess on my part as I do not know if people in this category earned less than 80 spd or did not appear in the tax lists by chance, even though they may have earned more than 80spd.

Source Ligningsprotokollen, Tromsø byarkivet b.831, Statsarkivet, Tromsø

Those who signed the forms had a considerably higher income that those who didn't (Table 2). From Table 2 it would appear that only 14 per cent of the household heads who completed the form had an annual taxable income of under 500 spd, (1) as against 72 per cent who didn't. On the other hand there were 12 - or 5 per cent - of household heads who did not complete the form and yet had a taxable income of more than 500 spd. in 1866, and, of these, 4 had a taxable income of more than 1000 spd. Who and where were these people? Some were possibly out of town. The forms were distributed on 6 January 1866, but they were 'to be completed according to the situation pertaining at the end of 1865 without regard for changes that had occurred in the meantime' (NOS :3). The census was to be of people who 'resided permanently in the house.

Those temporarily absent, e.g. at sea or on a trip within the country or abroad, to be included in the family from which they are absent. Those who are temporarily resident e.g. a traveller in lodgings, is not to be included, but should be enumerated - if he lives in the Kingdom - where his home is. (NOS :3)

Such an instruction was open to varying interpretations. As it happens, it is interesting that 7 of the 12 who fell into the category of 'permanent residence' in Tromsø were merchants and two were ships' captains. A macabre example of one who was absent - for good! - was the 'district doctor' Hendrik Nissen. He was born in Holstein, had lived in Norway 41 years and in 1865 was 64 years of age. He hanged himself between 31 December 1865 and 8 January 1866, and so - rightly - appeared in the census (RHD, 23.4). Age possibly played a role in another case. I.C. Dreyer was a merchant aged 70 years. He was born in Denmark and had lived in Norway for 49 years. Another member of the family (merchant Andreas Dreyer) who lived about 100 yards away completed the form for him (RHD 16.29 and 14.18).

Three of the group we are currently considering were dissenters. Did this play a role? There were not many dissenters in Tromsø in 1865. Of the 28 who were household heads, 10 completed the census schedule, somewhat above the percentage for householders as a whole. There were also four catholics amongst the 'first households' in a house. None completed the schedule. But one cannot say this was particularly remarkable. Two of the four had low incomes - respectively 160 spd. and 180spd.; another was a widow - and no women completed a form (see below) and the final one was one of the town's two catholic priests. He had been born in Bavaria and had lived in Norway for 10 years. The form was completed by the town magistrate, Ludwig Andreas Krogh, who also lived in the house (RHD 24.1).

If we look more generally at those who completed the census forms for others, we can divide them into two groups. One group of 18 lived in the house where they did the work, whilst another group of 26 did it for people who lived elsewhere. Of the 18 as many as 9 were related to the household head: in the majority of cases - possibly all - they were sons. A couple - Figenschou and Lie: see below - completed forms in several other houses, together with magistrate Krogh who we've already met. Finally, there were two teachers. One was a teacher at the Grammar School. He was a lodger of the headmaster (Johannes Steen) who for some reason had handed over the job to him.

Table 3 Taxable income of household heads in houses where another household member completed the census forms

Income (spd) Men Widows Total
3000+ 1 1
1000-1999 4 3 7
500-999 2 2 4
200-499 3 1 4
Under 200 2 2
Total 12 6 18

Source: see Table 2

Table 3 shows that those who handed over the job of completing the census forms to others were not always amongst the poorest. Quite the contrary, some two-thirds had incomes in excess of 500 spd.. A third were widows and half of these had incomes of 1000 spd. or more. It is unlikely, therefore that they were not in a position to complete the forms.

We will now look at the large group of household heads (229 in all) who had another person, who did not live in the same house, complete the census forms for them. As we saw in Table 2, a large proportion of these had relatively low incomes. A proportion of these, one might assume, were unable to complete the forms because their reading/writing skills were not up to it.

What of the 26 people who completed the forms for these 229 household heads? Table 4 tells us something about the people who did the bulk of the work. Of the 26, some 14 completed between one and six forms, with a mean of 2.5. Here we find a master tailor who possibly completed the form for his sister (RHD, 61.42 and 26.22); another who did it for several neighbours (RHD, 27.23; 27.2; 27.11; 29.8); another (Krogh), who we have seen already, completed the form for one of his own class (RHD, 23.4 and 24.1). Possibly several others completed the forms because they owned the property. We don't know.

If we look at the 12 people who completed 9 or more forms, we discover that they come from only three of the 13 groups into which those with a taxable income were placed (Table 4); namely 5 from Group 1, 5 from Group 4 and two from Group 7.

Table 4 Groups of taxpayers in order of appearance in the tax assessment lists, Tromsø, 1866
1. State servant, lower officials, state pensioners
2. Merchants
3. Shop clerks
4. Other members of the business community
5. Ships' captains
6. Ships' mates
7. Craftsmen
8. Journeymen
9. Chimney sweeps
10. Carters
11. Dock workers and labourers
12. Carpenters
13. Seamen, workmen and fishermen.

Source: See Table 2

I've put the town's tax collector in the first of these three groups, although (as a shoemaker) he should, perhaps, be placed in the third. In terms of age they were relatively homogeneous (from 28-46 years); their income spread was, however, relatively wide (from 180spd. to 1200 spd.). Almost all were married and 5 were Tromsø born.

Why did precisely these men take on the task?

Table 5 Biographical details of those completing a large number of census forms, Tromsø 1865.

Name Age Civil status1 Occupation2 Income 1866 Birth-place No. of forms completed
E.I. Brostrøm 44 S 1 500 Trondheim 14
O Christiansen3 38 M 2 250 Trondheim 27
S.Danielsen 41 M 3 300 Hammerfest 24
E.A. Dørum 34 M 4 180 Oppdal 16
A Dreyer 41 M 5 1000 Tromsø 21
A Ebeltoft 46 M 6 1200 Tromsø 17
P.H. Figenschou 32 S 1 1000 Tromsø 12
H.A. Givær 31 M 1 800 Tromsø 10
A.M. Hagen 28 M 7 350 Trondheim 10
H.W. Holmboe 40 M 8 1200 Tromsø 14
E.C.B. Mejer 43 M 9 440 Hammerfest 18
H.B. Theting 34 M 7 250 Røde 9
14 Others

1. S = Single: M = Married

2. 1 = merchant: 2 = taxcollector: 3 = master smith: 4 = master shoemaker: 6 = solicitor: 7 = teacher: 8 = bookseller: 9 = junior customs' officer.

3. An 'Ole Christiansen' was the town's collector of taxes. I believe he was a master shoemaker, as there was no other person in the census who fitted and because he lived in the middle of the area in which he completed 27 forms including his own.

Source: Census, Tromsø 1865 (RHD edition supplemented by photocopies of original forms: Ligningsbok 1866, Tromsø town archives, B832, Statsarkivet, Tromsø.

It doesn't appear to be associated with income, the spread was too wide for any meaningful correlation. On the other hand, social status does appear to have played a role, with the upper reaches of the town's hierarchy taking a disproportionate share of the task. For the social hierarchy see Table 4. Were the form-fillers recruited formally by the town magistrate, who had overall responsibility for the census, or the town's tax collector, who had the operational responsibility? Again we don't know. One clue suggesting the recruitment was more accidental than formal comes from the addresses of the form-fillers, and the areas they covered. For instance Brostrøm and Ebeltoft lived next door to each other, as did Theting and Hagen. Figenschou and Holmboe lived in the same house in Sjøgata. Both of these last two completed forms in Bakkegata, which was not far from where they lived. But most did the work for households in the streets where they lived, or but a few yards away.

What effect would the above have had on the accuracy of the census? To attempt to answer this we need to look again at the instructions provided by the census authorities. Sent to the magistrate and tax collector they read:

For use in the census there follows, what is reckoned to be, a sufficient number of blank forms, so that for each property and occupied house, a list of occupants can be drawn up. Before distributing the forms, the census administrator should give each a number in consecutive order. In addition, the tax number of each occupied house in the town, should be entered on the first page of the form. It is not necessary that the tax number be entered consecutively, so long as the census administrator finds another more appropriate way of proceeding. He must, however, be careful that no house is missed. (NOS :3)

It would seem, therefore, that we should find two numbers on the first page of each form: a tax number and a serial number. And, for the most part, this is the case. One would also expect that the serial numbers should begin with 1 and continue to 390, the latter being the total number of houses in the town. However this is not the case. Several numbers are missing, which would suggest the census is deficient. This is what one might expect if there was, as I've suggested above, a gap between how the census was supposed to be carried out and how it was carried out in practice. I do not, however, think the situation was as bad as it might, at first, appear. There were far too many missing serial numbers. But this was not the case with the tax numbers. By mapping these we find only a few missing.

One final anomaly is of some general interest. Henrik Martin Henriksen was a goldsmith's apprentice, unmarried and aged 19 according to the 1865 census. We find him listed along with Carl Nicolai Pedersen, a married goldsmith, aged 27, who lived in Grønnegate. But we also find Henriksen at the home of Frederick Waldemar Hvoslef, who was archdeacon and vicar, and who lived in Storgata. Henriksen is here described in the census as being 'apprenticed to goldsmith Edelsten. The houseowner is goldsmith Edelsten who has a workshop here, which is where Henriksen sleeps'. This appears to show quite a complicated situation: a craftsman who rents out a home to an archdeacon, who in turn has, on the property, at least at night, an apprentice to a goldsmith who lives elsewhere. To complicate matters there is no goldsmith Edelsten in the census! I believe, in fact, that goldsmith Edelsten and goldsmith Pedersen were one and the same. For, in 1884, a goldsmith C.N. Edelsten went bankrupt (Tromsøposten, Nr 83, 18, October 1884. I am grateful to Astri Andresen for this reference). Note the initials are the same as those of goldsmith Pedersen's in 1865. Finally, in the latter's house in 1865 we find listed a Pauline Amanda Edelsten, described as Edelsten's wife's daughter aged 2 years. (RHD 43.23 and 22.21. See also 13.16 and 18.32).

I would argue that this very parochial study of census taking has raised a number of issues of more general import. First it has shown that, with almost 60 per cent of household heads not being able - or perhaps willing - to complete the census forms, the authorities seriously underestimated the problems facing their 'new' method of census taking. From the point of view of the census as, a whole this was not a serious setback. For the 'official' urban population (i.e. settlements formally granted urban privileges) at this time only amounted to some 16 per cent of the country's population. And in the rural districts the forms were completed either by or under the eye of, the enumerators - usually teachers. The reason for the difference was that, given the dispersed nature of the rural population, to leave the census forms and return for them on another day would have been very time-consuming for the enumerator.

A second point of some general interest is the reaction to the failure of so many to complete the forms. We can only infer that this was ad hoc, a conclusion based on who eventually did the job, and where they lived in relation to the household heads whose forms they completed. That the better-off helped the poorer members of the community is not unexpected. That it was, in the main, members of the business community, rather than state employees (including the clergy), is, perhaps, somewhat more surprising.

Norway's nominative censuses are particularly rich, allowing much more detailed analysis than is usual elsewhere, at least outside Scandinavia. It would be worthwhile, I suggest, to pursue exercises like this both in other towns and on other matters.


No.5 (Norges Officielle Statistik, Resultaterne af Folketællingen i Norge, 1 Januar 1866, C. No. 1
RHD (Registreringssentral for historiske data (1990), Folketellinga 1865, Tromsø, University of Tromsø. The pairs of numbers refer to the page and line on which the reference can be found).

1. Spd is an abbreviation for spesidaler. At this time £1 = 4.5spd.