At the same time as the population census in 1801, a survey was carried out
of parish registers. This was updated in census years up to 1841. These
were published alongside the census reports as "Parish register
abstracts". You can find them on the Histpop website (seems to be down at
the moment). The reports detail which parish registers were available -
rather more then than now. Rickman was a good statistician - I'm sure that
he was well aware of the limitations of his data.
I can't share your faith in the Cambridge Group in relation to Welsh
demographic studies. Much of their early work was based on family
reconstruction from parish registers - a process which is impossible to
apply to Wales because of patronymics and the resultant paucity of surnames
- and also the shortage of early registers. For whatever reason,
subsequent work tended to concentrate on England. A quick glance at the
titles in their publication list shows that "Wales" only appears as
"England and Wales" whereas "England" appears alone many more times.
I'm not necessarily criticising them for this - we all know that for a
variety of reasons research is more difficult in Wales. But it's a pity
that there has been so little academic interest in Welsh demography, in
Welsh or English universities.
On 29 May 2017 at 13:31, Brian Swann <bps(a)norvic8.force9.co.uk> wrote:
Very interesting – one of those things that you think must have been
tackled to modern standards, but in fact it becomes very difficult to track
down any methodology deployed.
Thanks for the reference to Richman’s work, of which I was unaware. Not
too many of us read the Full Report on the 1841 Census. Obviously from
1801 onwards the statistics are easy as Census figures survive, the
challenges are the assumptions used to produce numbers before 1801. The
link you supplied leads to the following Footnote in Table 1.1 – which is
the only Table that says anything useful about the population of Wales
Richman's estimates, published in 1841 Census (Enumeration 1 pp 34-37 of
preface). The estimates were
obtained by assuming in the years 1570, 1600, 1630 etc. the same
proportion of baptisms, burials
and marriages to the existing population as in 1800 and 1801. For each of
the years given, the figures
were calculated by averaging the entries in the parish registers over a
period of three years (e.g. 1570 was
based on the average number of entries for 1569, 1570 and 1571)
If John Davies’ book really does have no reference as to where he obtained
his figures – then according to the criteria we would use regarding
acceptable standards of genealogical proof, we could rubbish his figures
immediately – no acceptable methodology provided, so you are unable to
argue or discuss its merits.
I get a bit concerned also if the assumptions used by Richman are correct,
or indeed how he gathered the information. For example, how many parish
registers survive for Wales for 1569-1571; 1599-1601; 1629-1631 and
1669-1671? Could he have actually accessed this numerical information in
1841, when the parish registers for Wales would have still been located in
the individual parishes?
It all reads very well – until you start to dig into where the figures
actually come from and what this last sentence actually means in practice.
And who would have ever queried such a statement in 1841 anyway? It was
all too new and population history and the use of statistics as a science
was in its infancy.
The work of the Cambridge Group on England should clarify this assumption
too: The estimates were obtained by assuming in the years 1570, 1600,
1630 etc. the same proportion of baptisms, burials and marriages to the
existing population as in 1800 and 1801.
It would involve translating their results for the English populations
they examined into the Welsh population, but they will have carried out
these measurements for sure – one of the principal reasons they were set up
was to answer questions like this.
*From:* Anna Brueton [mailto:firstname.lastname@example.org]
*Sent:* 29 May 2017 12:10
*To:* Brian Swann
*Cc:* RHODRI DAFIS; Robert Williams; dyfed(a)rootsweb.com
*Subject:* Re: [Dyfed] primary sources of information.
In my experience the Cambridge Group have shown little interest in
population studies in Wales (academic gossip attributes this to a falling
out between them and some Welsh academics). I can't find a source for John
Davies' figures, but they are consistent with those calculated by John
Rickman, who carried out the first British census. You can find these here:
These estimates have been criticised, particularly at the county-level,
but as far as I know, no-one has come up with anything better!
Rhodri, your county figures look a bit low in respect of Cardiganshire.
The Census count in 1801 was 42,956.