There’s a common belief that the work of historians consists of collecting ‘the facts’ from documents and writing them down. And that’s it. I mean, how hard can it be?
I can’t even begin to express the issues I have with such thinking. But let’s start with the obvious one – ‘the facts’. It’s something that hits us in everyday terms, even outside history. And the solution is the same in both senses.
The problem is that ‘facts’ aren’t always literally accurate. The issue – assessing the context and value of data – highlights the requirements of history as a technical study, although on my experience it’s something that few autodidiact history enthusiasts know because of the Dunning-Kruger effect.
Take the casualty figure of the Hawke’s Bay earthquake of 1931, for example. This has never been pinned down. A little while ago it was unofficially reduced from the official but long disputed figure of 256, down to 253. That still doesn’t tell us too much about the true quake figure. But it’s illuminating about something much more important – how do we handle data, historically?
The new figure was obtained by a painstaking analysis by Ancestry.com specialists of available geneological records, including death certificates. However, that wasn’t a definitive record: after the figure was published, someone pointed out that his uncle was removed from the casualty list because his death certificate listed ‘intestinal obstruction’, therefore the quake hadn’t killed him. The family story is that he was tipped off a fence by the quake and fell on a spike, catching him in the stomach – so, in all probability, he WAS a quake victim.
That raises it back to 254. And I am sure this won’t be the only story. See what I mean? Context is everything for history.
At this remove in time, of course, the specific number can probably never be definitely obtained. The work of Ancestry.com specialists has refined the picture, but there is no substitute for actually being there in 1931. So do official documents really give a picture? We can’t ignore the possibility of error at time of writing. On my own research into my family, for example, I’ve found discrepancies in death and birth certificates and in census records. And that’s apart from the fact that these documents don’t give context.
The issue highlights the point that history is a lot more than just collecting facts or listing numbers. Sure, that’s where you start – but the issue then is what do those facts mean? And are they, in fact, facts? The official figure was provided in the 1934 government report on the quake. The tally of 256 seems to have been a miscount – there are 258 people named on the memorial in Napier’s Park Island cemetery. But it’s been long disputed anyway because problems facing people at the time included:
- In Napier, fire followed quake – immolating casualties caught under the rubble before they could be dug out. Every effort was made to locate remains, but the issue complicated the identification and some people reported as missing were never found. There is a persistent story of a woman lost in the Masonic Hotel – where the fire was most intense – who was never found, and whose remains are thought to have been scooped up with the rubble and dumped on the foreshore (under the modern Soundshell).
- Some citizens took advantage of the confusion to deliberately disappear to avoid debts. One was tracked down in Auckland.
- There wasn’t an official enquiry into the casualty numbers in Napier.
- Some of the injured, who died later, were taken out of the district and missed being recorded. My family have an example, an injured man who was looked after. A doctor from Palmerston North arrived and took him away to care down south. Was he captured by the official record? We don’t know his name and can’t check.
- Nobody has agreed on a framework for identifying the casualties. By ‘deaths’ should we mean those killed by direct seismic action, either at once or who succumbed to injuries in days afterwards? Or should we include those killed during subsequent events – such as those who died in an air crash near Wairoa while reconnoitring post-quake? How can we distinguish between death by natural causes on the day and quake injuries? And do those need distinguishing? Did anybody die, for example, of a heart attack – now known to be one potential consequence of earthquake trauma, even if no other injuries occur on the individual. Depending on how the numbers are identified, various tallies could be made. So is it better to have several lists that include all of these under different categories?
The point gives perspective to those who imagine ‘history’ as a data-copying exercise. As for the quake issue? There is room for a new coronial investigation even today – if for no other reason than enabling a new official tally to be established. However, because of the uncertainties an absolute ‘final’ number of deaths will likely always be a matter for argument (and that, of course, is something the coroner can officially point out).
Copyright © Matthew Wright 2017