History

This website has sometimes been said to be over-critical and negative about the industry that is its subject. That isn't the intention, and I thought it might be useful to outline my personal philosophy on the nature of history, and of knowledge in general.

Our species is called homo sapiens. Central to our self-image is the idea that it is our inexorable pursuit of knowledge that sets us apart from the dumb beasts. And yet true knowledge is an elusive thing - so elusive, in fact, that our pursuit of it may be said to be always unrequited. Little, if anything, in our experience is genuinely knowable. Professions of knowledge are always questionable and often fraudulent (Note 1). The pragmatist might ask - "without knowledge, how could we have sent people to the moon?" The correct response to this is that, to get to the moon, knowledge is not required.

Cursed by their inability to attain true knowledge, human beings create narratives to explain the world they live in, and invest these with the regalia of knowledge. Such narratives might be the Book of Genesis, the Phlogiston Theory, or an account of the sequence of reactions that takes place in the bowels of a cement kiln. The narrative remains inviolable as long as its predictions remain sufficiently well in accord with practical experience. Successful narratives are propagated at the expense of those that fail.

That narratives are inviolable is a political statement, and narratives, in collaborative human ventures, are sacrosanct - that is, contradictions and criticism of them are in practice discouraged and suppressed. Every human organisation has self-preservation - and preservation of its narrative - as its primary sine qua non function, to which all other functions are subordinate. Accounts of actual experience are selected and distorted in order to fit with the accepted model. The adoption of a new, more successful, narrative only takes place by disruption, in which the priestly proponents of the old are not converted to the new, but simply rendered irrelevant.

In surveying the historical sources, particularly on a technological subject, one finds that people got things wrong all the time. Their errors not only limited their own success, but also that of succeeding generations. They were not necessarily at fault for this - ignorance is after all the default human condition, and learning follows the arrow of time (although sometimes things are unlearned as well). The application of a now-superseded narrative was not qualitatively different from the application of the current narrative. However, the process of writing history here, now, always consists of interpreting the past in terms of the current narrative.

What emerges from this theory is that knowledge (in the true sense) is a rare commodity and the best one can do is to create a convincing narrative that fits the available data. This applies to any human enquiry, whether scientific or in "the liberal arts". The writing of history is conventionally assigned to the latter.

I spent a lot of my time in the 1990s (when not making cement) reading and re-reading A. S. Byatt's Possession, and in the next decade her The Biographer's Tale. Both novels have as their central theme the near-impossibility of reconstructing history from limited, fragmentary and imprecise data. One is left wondering how any history has ever been written, given the limitations of the sources. However, history has been written - in great quantity - often with little data included, apart from a meagre scattering of dates. Perhaps, when Henry Ford said "History is more or less bunk", he was expressing the view that the professional historian has a license to talk bollocks. However, this is a philistine view. The writing of history is the purest form of narrative creation. When there is a paucity of historical data to work with, this allows an unconstrained inventiveness in the historical account.

It occurs to me that the "liberal arts" approach and the scientific approach are not so different. Both, typically, involve creating a narrative from limited, fragmentary and imprecise data. And for the scientist, this is not a daunting prospect - in fact, it's all in a day's work. The only difference is that scientists apply an additional constraint to their narrative - that of Occam's Razor.

My favourite example of scientific use of scanty data is the elucidation of organic molecular structures (from 1930 onwards) using x-ray diffraction - studies first introduced to me by Professor Mansel Davies. From an array of dots on an x-ray photograph, the geometries of complex molecules were obtained with sublime precision. Furthermore, the positions of the dots show only the shape and size of the unit cell: the molecular geometry was distilled solely from the intensities of the dots. These, at the time, were extremely imprecise - measured simply by "looking snake-eyes" at each dot and assigning a value on an arbitrary 1-100 scale. But there were lots of dots! The final geometry was adjusted so that its calculated diffractogram would most nearly agree with the array of intensities measured.

The message here is that something useful - even by Henry Ford's exacting standards - can be derived from a shoe-box full of unpromising snippets of data, if it is treated effectively.

Not every approach is effective. The British historian Nennius (said to have written the Historia Brittonum) famously said "Ego . . coacervavi omne quod inveni" - "I have made a heap of all that I have found". A recurrent theme in this website is that there are many such heaps lying around, and while they remain as a heap, they contribute nothing. So how do you deal with a heap of stuff?

I have often thought that historical research is a bit like doing a jigsaw puzzle. In fact, it's a lot like doing a jigsaw puzzle - one you found in the loft, certain to have pieces missing, and probably also containing pieces from other puzzles. You have a vague guide to what the final picture should look like. You look at each piece, and get an impression of what it might be part of. At this point, various strategies are possible. On the one hand, you might analyse every detail of the piece, and scan the guide picture until an exact fit is found. The piece is then placed at its precise Cartesian coordinates on the table, and you start on the next piece. Etcetera. On the other hand, you might sort the pieces into broad categories of colour or texture, occasionally finding chance pairs of pieces that that fit together, until broad clumps of complete image begin to coagulate, and the process proceeds rapidly from there on.

Needless to say, the first strategy is hopeless. This is the approach of those who would not begin to write history until every figure and statement is validated from multiple sources, duly referenced, with more footnotes than text. The approach fails because it proceeds at a snail's pace, or not at all. A picture never emerges. But the picture is the desired outcome! And the bigger the picture, the better! In my x-ray diffraction example, the data are poor, but there are lots of dots!

The big picture is important here. As I put together a history of the British and Irish cement industry, I frequently come in contact with local projects, preparing meticulous accounts of one plant, or one process, and presenting them finely inscribed on a little bit of ivory. The results, of course, are worth their weight in gold to me, but generally, they don't weigh much. This is because they never see the Big Picture. They re-invent the wheel, and as often as not design a square one. This dispersed treatment is an unfortunate result of the dispersal of the industry - a dispersal that becomes more pronounced as the remnants of Blue Circle are divided into bite-sized chunks.

There are many good examples of the Big Picture effect. The Barton plant is the subject of local studies. The plant at Stockton, several groups seem to be having a go at. Both these plants started with static kilns, but installed rotary kilns just before WWI. It emerges that the rotary kilns were of similar dimensions, and built simultaneously by the same supplier - Edgar Allen, so they are probably identical. Both installations are irritatingly poorly documented: there are some dimensions for Barton: there are some production statistics for Stockton. There are no maps for either. But undoubtedly most information about them has been lost, as their respective companies were taken over and their cultural effects were buried. Then one day, a picture of the Stockton kilns emerges. It shows layout, ancillary equipment etc. Now, not only do we know what the Stockton kilns looked like; we also know what the Barton kiln looked like! Only by a global approach can these insights be obtained.

Of course, this is just another little cluster of pieces in the jigsaw: the emergence of a global picture is still a long way off. The status quo is still largely a table of scattered pieces, some rightly placed and ripe for bits to be attached, and some only very vaguely located, perhaps wildly wrong. But they won't get any less wrong by being put back in the box.

So much for the general approach. Another aspect of the website that might be subject to criticism is my tendency to editorialise. Nothing can be more pointless in a historical account than to present past errors uncritically. Bearing in mind the need to preserve the organisation's cultural memes, it's obvious that whenever the results of an innovative experiment are published, the experiment is claimed to be a success. A retrospect of decades allows a pragmatic assessment of the truth of the claim to be made.

As an example of this, there arose in the mid-1920s the practice of introducing slurry into the upper end of cement kilns by means of sprays. Every account of this pronounces it to have been a dazzling success. A large proportion of the dried feed was entrained in the exhaust gases, and because the kilns of the time had little in the way of dust filtration, dust was emitted over the neighbourhood in volcanic quantities. However, accounts of the time said that dust emission was not an issue, or even that slurry sprays improved the situation. Published accounts a few years later said that dust had been a problem initially, but this had now been fixed! At no time (of course) was there any suggestion that an installation was a failure. Using only the published accounts as sources of information, one might conclude that the process was so successful that it was installed everywhere and became standard practice. In fact all installations were removed within a very few years. Only by looking at newspaper accounts of the cacophony of dust complaints that took place at the time does one get any idea of an alternative point of view. Of course, the people at the "sharp end" of the industry who were required to implement this system were very well aware of its shortcomings, but had to soldier on. Those who dared put their heads above the parapet and point out that the "King's New Clothes" were not as claimed, were quickly "disappeared". To be on the team, you have to be a true believer.

I leave it to those still in the thick of it to say whether this style is still a feature of the industry.

The point that emerges is this: anyone versed in the current narrative would say without having to think for long that slurry sprays were a really stupid idea - probably emanating (as in fact was the case) from someone with little knowledge of the industry. A historical account should be capable of making this point.

I have said that this requires a retrospect of decades. It follows that current events can't be subject to the same critical approach. However, one can at least avoid adopting the current memes, and this I have tried to do. And if the King's still naked, I'd rather say nuffin than join in admiring his robes. It's difficult. But (as a colleague used to say ad literal nauseam) no-one said it would be easy.

NOTES

Note 1. The argument goes something like this. A necessary condition for a claim to knowledge is that the thing known should in fact be the case. Another way of stating the principle is that "One can't know a thing that isn't so". A statement such as "I know that the Earth is flat" is not valid because the Earth isn't flat. A logical inductive extension of the principle is that one also can't know a thing if, although its truth is at present unchallenged, it is capable of being challenged. A claim to know such a thing is not valid because the claimant is merely chancing his luck on a statement that may be struck down at any moment, by a mechanism that is undefined. A little consideration proves that few propositions survive this test, and most claims of knowledge are invalid at best and fraudulent at worst.