Lack of Environment

A blog on the politics and psychology underlying the denial of all our environmental problems

A riotous assembly of data leads to a ground breaking conclusion

with 36 comments

Or should that have been: ‘Ground-breaking data leads to riotous conclusion’…?

One of the incidental benefits of being a Fellow of the Geological Society and a Chartered Geologist is that I get the society’s monthly Geoscientist magazine. This month’s edition includes an article written by Alan Watson – a Chartered Civil Engineer and a member of the Institute of Civil Engineers – who has published an eBook entitled ‘Gravity and Mind – Human Response to Tectonic Stress‘.

Astonishingly, Watson’s book and his article question the scope and reliability of the notion ‘free will’.  Leaving aside the question of crowd psychology (i.e. the fact that individuals behave differently when in a group), Watson’s analysis of data suggests that civil disturbance can be linked to tectonic disturbance.  That is to say, riots are more frequent in the days immediately preceding earthquakes.

With the permission of the editor of Geoscientist, I publish extracts of the article below with graphical representation of the results of Watson’s analysis, upon which his startling conclusions are based.  However, many of the images in the original article are Copyright of the British Geological Society (BGS).  Therefore, if you have the time or inclination, please view the whole thing on the Geological Society website.

Here is how Watson begins this article:

Reports of ‘unusual animal behaviour’ before earthquakes became common during and after the 1960s – snakes coming unseasonally out of hibernation, dogs deserting their kennels, birds sensing impending quakes and, most recently, insects not resting. But, as anyone who has lived with animals will know, animals ‘behave strangely’ all the time; which means evidence of this kind suffers from a huge and possibly unresolvable ‘false positive’ problem. The trouble is, nobody bothers to record animals’ ‘normal’ behaviour. And even if you do watch them all the time, the quality of their behaviour is extremely tricky to quantify.

Humans, though are different. We have the media. I believe that comparing news reports with seismicity data provides compelling evidence that we humans may be responding to the effects of seismicity shortly before earthquakes.

I think it is worth repeating, here, that Watson is no liberal-minded social worker seeking to excuse irresponsible or criminal behaviour.  He is a well-respected geotechnical engineer with a professional reputation to look after.  Therefore as you will see, especially if you read the whole article, he has been meticulous in consideration of all the reasons why correlation might not imply causation.  As he says:

My intention in this article is to summarise the facts about the relative timing of earthquakes and riot, and let you make your own mind up.

We all know the adage about correlation not necessarily meaning causation; but the first step must be to determine whether there is at least a correlation there. I believe the statistics show there is.

As such, Watson investigated two hypotheses regarding earthquakes of magnitude greater than 2.5 on the Richter Scale (i.e. ‘2.5ML’), namely:

  1. That there is a significantly higher incidence of riot and disorder shortly before earthquakes of 2.5ML or greater, compared with the same period afterwards.
  2. That there is a significantly lower incidence of riots and disorder after more than 140 days has passed since the last most recent earthquake of 2.5ML or greater, compared with the incidence that would be expected by chance.

One of the many potentially complicating factors that Watson acknowledges is the reality that, as happened in the UK in August 2011, one riot can often be the trigger for so-called “copy-cat” riots by those you might call opportunists.  Hence Watson says:

A substantial number of cases of riot appear in clusters with a common initial cause. My dual studies have therefore included both a full appraisal of these cases, including ‘tails’ of clusters as well as excluding them. One would imagine that copy-cat rioting in these tails of riot clusters would be influenced to a lesser extent by seismicity than might be the case for the initial onset of violence. The dual study therefore removes the uncertainties resulting from such potentially contaminating ‘sociological’ effects.

As stated in my introduction, many of the Figures in the original article are Copyright of the BGS.  However, with my thanks to the editor of the Geoscientist magazine, I am able to reproduce here the graph demonstrating the correlation of the ‘without tails’ data.


Watson summarises the results of his analysis by making the following six statements:

  1. There is a significantly higher incidence of rioting and disorder in the 14 day periods prior to earthquakes compared with the 14 day periods after earthquakes.
  2. The ratio of riot frequency before to after earthquakes falls off from a peak of 3.2 (with tails and 2.5 without) within 14 days to a lower ratio of 2.5 (with tails and 1.67 without) within 7 days of the shocks.
  3. There are substantially fewer instances of rioting and disorder when more than 140 days have passed since the last most recent earthquake of at least 2.5ML.
  4. These findings will provide support to other earth science studies about interactions between the biosphere and the lithosphere. There have been reports of unusual behaviour exhibited by birds, snakes and insects, among other species, prior to earthquakes. This project widens the scope of influence between the lithosphere and biosphere and asks the question: are humans influenced by the behaviour of the lithosphere in ways not yet understood?
  5. The statistics of riot and earthquake incidence serve to re-affirm seismology research known as ‘the new geophysics’ that tectonic stress may vary on a regional scale prior to earthquakes.
  6. The occurrence of riots, in certain circumstances, may provide one further factor to consider, when assessing the risk of an impending earthquake.

In his personal communications with me, the editor of the Geoscientist magazine expressed his own surprise at the conclusions of Watson’s analysis, to which I responded as follows:

On the fundamentals of the statistics and/or the plausibility of the hypothesis, I think it much more credible than many other commonly-held beliefs about the nature of reality (and I am not talking about religion).

For the avoidance of any doubt, one of the ‘many other commonly-held beliefs about the nature of reality’, to which I was alluding here, is the startlingly-persistent, unduly-optimistic, counter-factual, and/or ideologically-prejudiced belief that humans are not primarily responsible for the unprecedented warmth and accelerating change through which we are now living.

As the pre-eminent film director, James Cameron, says in this trailer (below) for a new television series, ‘Years of Living Dangerously‘,  to be screened next year:

“If 99 doctors say you sick and need an operation, would you seek another opinion?”

36 Responses

Subscribe to comments with RSS.

  1. Disclaimer: Martin, I haven’t read the full paper and am commenting only on what you have presented so this may be completely irrelevant but all I can say is…There’s nothing like an earthquake to break up a riot.

    Was there anything in the methods or any caveats to account for the effect of a quake itself on behaviour because preumably the before and after were the same place?


    8 November 2013 at 03:34

    • Did you mean break up a riot of for riots to break out? If you are suggesting people riot in response to earthquakes, that does indeed appear to be borne out in the statistics. The surprising thing is the more-than-expected result before the earthquakes (and less-than-expected in absentia). I do not see how riots stopping on day 0 would affect these kinds of conclusions.

      However, one of the more obvious defects in the the analysis is the failure to link earthquake and riot location (or the fact that there is no link).

      Martin Lack

      8 November 2013 at 10:42

      • It seems a bit odd to me. Why riots? Are they using the word riot as an umbrella for other activities or do they literally mean bunches of people running around breaking stuff? Anyway, in the words of my teenage son…. meh.


        8 November 2013 at 21:37

        • Watson has compiled his own database of what he calls “widely reported riots” from trawling the internet and news media archives. As such, civil unrest serious enough to get reported on the national news – on the street and inside prisons – seems to be included. He invites people to inspect his data and tell him if he has missed any significant civil unrest.

          Maybe you and/or your teenage son should consider Rachel’s remarks about Christchurch in NZ (i.e. below)? The overall correlation in Watson’s UK data would appear to be undeniable but that does not mean that people are actually responding to tectonic stress. I think the latter may well be plausible but only in a geographically limited way (and such a correlation is noticeably absent).

          Martin Lack

          9 November 2013 at 12:05

        • lol.You know me Martin and you know my passion for science and as I said I didn’t read the paper so any criticism I have of methods and whatnot is only based on what you guys are saying. My comment “Meh” isn’t a criticism of anything inparticular with the paper, but is more a case of me wondering what the point of it is? To me, science should have a purpose other than looking for obscure correlations. Is the author hoping to develop some sort of early warning system for Earthquakes based on human behaviour? Perhaps law enforcement organisations can use information from seismologists to predict when there is a greater risk of riots? My suspicion is that this is science for science sake and apart from contributing something rather meaningless to overall human knowledge it doesn’t seem to have any real value. It would appear to me to be a waste of resources that could have been spent researching something important with tangible benefits? That is just my personal value judgement.

          I do have a question about methods……. of course. Why would he rely on media reports for his data and not go directly to law enforcement organisations that would have very accurate records? Why riots and not some other aspect of human behaviour? Why not domestic violence or public drunkenness? How did he classify a riot? Is it more than n people acting in x manner causing y amount of damage resulting in n police attending etc? Again I don’t have the paper in front of me to read the methods but on the surface of what you guys have said it all seems a bit vague…..and odd.


          9 November 2013 at 21:18

        • Your cynicism is understandable, Mike. The author has indulged a personal interest and written a book; and is now trying to publicise it. Something I would never do, of course. Oh, no, wait a minute, may be I did…😉

          Martin Lack

          10 November 2013 at 09:29

        • Mike,
          Just a quick response to something you’ve said here about science needing a purpose. Really? I disagree. Some areas of science/mathematics are so obscure that there is no purpose or practical application of the research but this is not to say that one day the research will not prove useful. The British mathematician, G.H.Hardy, once described his work in number theory as completely useless and it was at the time. But a half-century later his ideas form the foundation of modern cryptography on which all of e-commerce depends. He never lived to see the practical application of his work, unfortunately. We don’t necessarily know whether something will have a purpose or not until we study and try to understand it.


          16 November 2013 at 06:31

      • Martin, with regard to riot-quake distances, my analysis is based on the findings of contemporary research, known as the ‘New Geophysics’ which has found retrospective evidence that earthquakes can be forecast on the basis of stress monitoring at locations remote from the ensuing quake. This relatively recent development indicates that changes in tectonic stress and therefore, also, seismic effects caused by changes in tectonic stress, can occur over large regional / continental areas. On this basis, I stand by my decision to compare all the riots in England and Wales with all the Earthquakes of 2.5ML & greater within the same geographical area which has relatively consistent tectonic conditions.

        The unusual relative incidence of riots and earthquakes must have some rational explanation. I just want Geoscientist readers and the wider scientific community to see the facts and I will welcome comments / suggestions about the meaning of my findings.

        Alan Watson

        15 November 2013 at 20:21

        • Thanks Alan. I think Rachel’s comment about people being too busy to riot in the aftermath of NZ quakes is interesting but, I admit, this does not help explain your UK data (because none of the UK quakes was that serious). However, in addition to the possibility that there is a rational explanation for the non-uniform distribution, is it not also the case that there might not be a rational explanation? I am drawn to this conclusion myself because you admit you could not do your analysis in NZ because there are too many quakes and not enough riots. Surely, if the hypothesis is that humans can subconsciously respond to increasing tectonic stress, this begs the question,’Why are there so few riots in a tectonically active country like NZ? This would appear to be in danger of reducing your UK correlation to the equivalent of seeing the face of Jesus on a burnt piece of toast.

          Martin Lack

          16 November 2013 at 10:45

        • Martin, how could you! My research is certainly not toast!

          ‘Our environment is everything that is not us’

          A variation between responses in different parts of the world merely demonstrates different forms of expression. Who would deny that the human species is capable of this?

          Alan Watson

          16 November 2013 at 22:12

        • Hi Alan. I hope you appreciate that I was being deliberately provocative (as opposed to gratuitously dismissive). Nevertheless, how significant an effect can this be if the nice people in NZ can overcome it despite much more serious tectonic provocation?

          Martin Lack

          17 November 2013 at 13:40

        • The tectonic stress variations and consequent seismic effects in England and Wales may be in particular resonance with the human population, whilst potentially more pronounced effects in New Zealand do not resonate with the population in the same way. Rachel has mentioned New Zealand crime statistics – and my e-book includes a sub-section on this aspect – so the different level of resonance simply induces a different form of expression.

          Alan Watson

          17 November 2013 at 18:49

        • I can see I am going to have to read the book… Nice work, Alan. 😉

          Martin Lack

          18 November 2013 at 11:11

  2. More evidence of how human society affects Nature. If riots can cause earthquakes, then it is reasonable to consider that human activity can influence the climate.

    Bill Everett

    8 November 2013 at 04:35

    • Bill, I am conscious that this is what my silly title may be misleading. With the benefit of hindsight, it would have been far better for me to have called it ‘Ground-breaking data leads to a riotous conclusion’. The statistical analysis of all the data implies that tectonic stress may somehow cause civil unrest (i.e. 3 times more riots in 14 days before an earthquake than in the 14 days after one).

      Of course, your supposition that riots cause earthquakes would be one hypothesis to explain the non-random distribution. However, with respect, I think it is one of the least plausible. Surely the simplest hypothesis (that tectonic stress causes psychological stress) is indeed most likely to be correct?

      Martin Lack

      8 November 2013 at 10:53

  3. I haven’t read the full paper either Martin but if I get the chance later in the day I will. Having lived through three rather large >Mag 6 earthquakes I have to say my first thought is that this is complete nonsense.


    8 November 2013 at 06:51

    • Your initial reaction appears similar to that of the editor of Geoscientist magazine. However, I applaud him and the Geological Society for publishing the article.

      Martin Lack

      8 November 2013 at 10:56

  4. I read the article in Geoscientist magazine and it does read as though Alan Watson knows what he’s talking about and is onto something. But the problem is that unless you’re a statistician or have a good grasp on statistical analyses, which I don’t, then it’s very hard to judge.

    Something that happened after the Christchurch earthquakes (and there were thousands of them but four main big ones), is that crime fell immediately afterwards. So it does seem a strange thing to compare violence and rioting in the 14 days prior and the 14 days afterwards when crime falls as a consequence of earthquakes.

    I read an article recently about crime in Christchurch and it said something about the offender population only just returning to pre-quake levels but this is a couple of years later now.


    8 November 2013 at 10:00

    • Thanks for taking the time to read the whole article, Rachel. I think it is you that may be on to something here. Given your personal experience, you have identified a reason for at least one of the ‘unexpected’ results (i.e. 3 times more riots in 14 days prior to quakes compared with 14 days after quakes). As I have not found a way to contact Alan Watson direct, I will feed this back to the Geoscientist/Website editor.

      Martin Lack

      8 November 2013 at 11:01

      • Do keep us posted on this topic. I’m a bit obsessed with all things earthquake-related.


        8 November 2013 at 11:15

        • Interesting read, Martin. But must direct a question to Rachel, namely what causes said obsession with earthquakes? Just curious!

          Paul Handover

          8 November 2013 at 14:20

        • In case you struggling to piece the biography together, Paul… Rachel is an NZ ex-pat now living in UK (having left Christchurch after the record-breaking 7.1ML earthquake).

          Martin Lack

          8 November 2013 at 15:26

        • Paul,

          There’s nothing quite like being woken in the middle of the night by a magnitude 7 earthquake to ignite an interest in geology. I’ve experienced hundreds of earthquakes since then and it’s fair to say that this is what brought on the obsession. I didn’t think much about them before that.


          8 November 2013 at 14:26

        • Thanks, Martin. I probably should have been clearer. I used to live in Christchurch, NZ and was there for all of the earthquakes. It was awful and that is why we left. The people in Christchurch have become armchair geologists as a result and we are remarkably good at guessing magnitude and location of an earthquake when we feel one.


          8 November 2013 at 16:00

        • All may enjoy reading this article on the UK’s Carbon Brief website: ‘Can climate change cause earthquakes? We look at the science and the spin’ (11 June 2012).

          Martin Lack

          8 November 2013 at 16:21

        • Rachel, thank you for your interest in this new area of research. To clarify one aspect of my study I should say that the human response to tectonic stress varies between regions of differing tectonic conditions. The conclusions presented in my Geoscientist article are based on an assessment of the land area of England and Wales between 1980 and 2012. England and Wales as a tectonic region is ideally suited to a comparison of riot and earthquake distribution as I found out when I saw the outcome of my analysis for the first time! New Zealand is a very different environment in this regard as not only are tectonic conditions completely different to England and Wales, but there are also far fewer riots. It would not be possible to undertake the same type of statistical appraisal for New Zealand as the high number of earthquakes precludes such an approach. The subtle influence of tectonic stress (about 10% of all riots are affected in England and Wales) would be very difficult to pick out of the riot-quake data in areas of high seismic activity such as New Zealand. Nevertheless, there are many worldwide examples of riots shortly before earthquakes so the human response, which may find different forms of expression in different tectonic regions, is certainly likely to be proven as a global phenomenon. I hope this helps.

          Alan Watson

          15 November 2013 at 19:49

        • Thanks very much Alan for taking the time to respond. Do you have any explanation for how a relationship between tectonic stress and riots might exist? How can humans respond to tectonic stress prior to an earthquake? How could they possibly know? Seismologists have instruments for measuring even the slightest of movements in rocks and they have no way of predicting earthquakes. How could it be possible for humans to do so subconsciously? It all sounds a bit ‘magic’ but that’s not to say that there isn’t a reasonable explanation, just that I don’t see it.


          16 November 2013 at 06:23

        • Rachel, the title ‘Human Response to Tectonic Stress’ could be expanded more accurately to ‘Human Response to the Effects of Tectonic Stress’. There are seismic effects prior to earthquakes which include, for instance, the emission of positively charged ions from rocks under stress. There is a reference to this type of effect by Richard Batchelor on the Letters page of Geoscientist Online. This is by no means the only type of seismic effect which would be capable of explaining my observations.

          Your question about how it can be ‘sub-conscious’ is a slightly different issue because I get the impression you mean that there is no currently known way that humans can detect remote changes in physical stress itself. Well, you may be correct about that, but I would not rule it out as a possible explanation. At this early stage I am not ruling anything out. A combination of multiple seismic effects as well as stress itself may actually be contributing to the phenomenon.

          I am in full agreement with your other post targeted at uknowispeaksense that you don’t have to know the purpose of research in order to carry it out. I make this point in the Afterword to my e-book.

          Alan Watson

          16 November 2013 at 21:41

        • Sorry to intrude but, over what sort of distances are you suggesting these ‘positively charged ions’ can have an effect? I am willing to accept that humans can be locally affected by electro-magnetic radiation – from mobile phones in particular (see ‘Beings of Frequency’ documentary) – but I think it highly implausible that we can be affected over large distances by ‘positively charged ions’. Are you saying there is evidence for such low energy radiation being generated over wide areas in the days before an earthquake?

          Martin Lack

          17 November 2013 at 14:03

        • There is certainly evidence that seismic electrical signals have been recorded at distances of the order of 100km (and possibly greater distances) from ensuing quakes.

          Ref :

          Alan Watson

          17 November 2013 at 19:11

        • Thanks for that clarification, Alan. If you have not seen it, I would recommend the ‘Beings of Frequency’ video (link now included above), as it is very convincing.

          Martin Lack

          18 November 2013 at 11:08

      • Thank you Martin and to all of you who have expressed an interest in my article. I will try to respond to all the comments and questions on this blog, which may take a while, so I hope you can all be patient! One thing I will say right now though, is that the Geoscientist article, yes, is about a Human Response to Seismic Effects prior to Earthquakes. On one level it is about riots and earthquakes, but on another level, and in answer to uknowispeaksense, who queried the motives behind the article, this is about the relationship between the Human Species and Planet Earth.
        Martin, it is highly appropriate that this new area of research should be aired on your blog devoted to environmental concerns.
        I will be in touch again soon…..

        Alan Watson

        15 November 2013 at 00:35

        • Dear Alan, Many thanks for your visit; and for offering to respond to comments submitted. I very much look forward to your return.

          Martin Lack

          15 November 2013 at 16:28

        • “and in answer to uknowispeaksense, who queried the motives behind the article, this is about the relationship between the Human Species and Planet Earth.” – Alan Watson

          Hi Alan, I do hope you are going to be more specific than this. I’ve written a number of papers that could broadly be defined as “examining the relationship between fungal pathogens and the biosphere”. That said, if you’re prepared to send me a copy of the paper I’ll be prepared to have a more indepth look rather than relying on the comments here. That way I can be truly informed and the answers to the questions I have raised may become apparent. My email address is and so you’re not interacting with someone completely anonymous, my name is Mike.


          15 November 2013 at 20:30

    • Hello again Rachel, since our last exchange I have been looking into the background to your questions about crime statistics before and after earthquakes. There are a number of competing influences in this regard, with one significant factor in the case of the Christchurch quakes being that many people left the area afterwards resulting in a noticeably reduced population. This would inevitably impact on the subsequent total crime statistics (as opposed to criminality per head of population).

      There are other potential factors in other geographical locations, for instance in Haiti there were reports that convicts had escaped from prisons damaged by the quake, with an inevitable increase in crime afterwards.

      This means that my study of England and Wales has significant advantages over other geographies because the low magnitude quakes here are not followed by reduced population densities nor substantial increases in escaped convicts! I think you will find that England and Wales is an ideal site for a proof of statistical significance in the unusual distribution of riots when compared with the distribution of earthquakes.

      Alan Watson

      31 December 2013 at 15:15

      • “This means that my study of England and Wales has significant advantages over other geographies because the low magnitude quakes here are not followed by reduced population densities nor substantial increases in escaped convicts!”
        I have to say, Alan, that this is a good point well made. Happy New Year!

        Martin Lack

        31 December 2013 at 18:26

Please join the discussion

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


Get every new post delivered to your Inbox.

Join 351 other followers

%d bloggers like this: