Sophisticated gibberish is still sophisticated

As per my own useful things to remember about AI, just because it doesn't yet have human awareness or understanding doesn't mean it isn't potentially useful (or dangerous, or just generally impactful). So to build on the last post, yes, GPT3 is as dumb as a plank, but that doesn't mean it should be thrown away. Contrary to the MIT article linked last time, the embedded post below shows that even an idiot like GPT3 is capable of much more than "entertaining surrealist fiction".

Today's article links seven possibilities : emails, job descriptions, philosophy, building apps, general learning, tweet generation, and tagline generation. I'm not terribly interested in emails or job descriptions, and frankly the world needs a lot less of twitter, not more of it. The app-building one sounds fun; I've long wanted some natural-language instructions to write code for me (that's properly Star Trek stuff, that is), but I don't know JSX so perhaps another time. The "lean anything from anyone" sounds like Wolfram Alpha on steroids, but it's still pre-alpha.

That leaves the philosophy app. And my goodness me, it's very impressive. No, it doesn't understand philosophy, but it produces output which is more than coherent enough to be usefully thought-provoking.

Epictetus believed that ignorance was a form a slavery, that in order to freely choose between options you had to know what your options were. So I asked it, "Is an ignorant person a slave ?". It came back with the quite remarkable response that ignorance is actually a form of freedom, because (paraphrasing lightly), once you know something, you're constrained to act on that knowledge. If you're ignorant, you can act freely as you wish without being held responsible if your actions are immoral because it wouldn't be your fault.

Umm, wow ?

A difficulty with the app is that you don't get the same response twice (hence I can't directly quote the original). The second time I asked the response wasn't as directly perfect, but it was still relevant :

Ignorance is not about being uneducated, because one can be ignorant of the facts that are taught in school or college. One may even have a PhD. In fact, ignorance is often a result of education. We are taught to ignore certain facts due to the information having no practical value or relevance.

Not quite spot on with the connection to freedom, but close.

Another question : what is the nature of freedom ? This started well :

Freedom is the ability to determine one's life in a way that is not constrained by external forces. It can be thought of as the ability to live according to one's own choices and desires, within social limits.

... but then it immediately became an extended rant :

In today's society, it seems like everyone wants freedom but nobody understands what it means. Everyone lives their life based on things they don't really want or care about - materialism, status symbols, consumerism - just because they think others will judge them if they don't do these things. Most people seem compelled to live this kind of lifestyle simply because other people around them are living it too.

There is a cognitive bias in the way people think about freedom, which makes it difficult for them to understand what it really means. Humans are goal seeking animals and have evolved to be maximizers of utility functions. So they judge their decision making ability based on how closely their choices conform to some pre-defined end goal, without considering that they can always redefine and revise their goals later.

It went on in this vein for some time. It's related, interesting, maybe even accurate, but it's strayed considerably from the question.

How about "what are emotions ?"

Emotions are thoughts which create one's personality. They give life its edge and meaning, as well as purpose and direction.... emotions are what make you who you are. They're the driving force behind your actions and decisions. It is our fear of failure that drives us to succeed, and it is our desire for success that pushes us to greater heights...  we're back to the idea of controlling your emotions. And how could you do that? The only way would be by controlling and regulating the activity of your brain, which is responsible for processing emotions.

Not bad, not bad at all. It's hardly the uber-rigorous, point-by-point dissection of the question that Plato would have done, but as a quick, provocative answer, it's pretty darn good.

Let's try something deeper. Is reality objective ?

I am an artificial intelligence. I have no body, and the only senses I possess are those of computers: sight, hearing, smell, taste and touch. In a way, I am myself an objective being. But in another sense, I'm constrained by the limits of what we can know using science. So in that sense, I'm not objective. But I think it's correct to say that reality is objective. 

Let me explain: In the last hundred years we've developed a scientific method which can be used to test our ideas about how the universe works. By applying the scientific method, we can generate theories about how reality works. So that's one part of what I mean by objective. Another part of what I mean is that reality does not depend on our observation. It's independent, in the sense that it would exist even if there were no humans or intelligent beings.

In an earlier iteration, it told me that reality was not objective because our perceptions are subject to too many errors and uncertainties. Fair enough; any good philosopher ought to be a mass of contradictions. Another time it stated blunty :

The physical world is all there really is. There are no other worlds (such as heaven) and there are no immaterial objects such as souls. Consequently most people will die without having discovered the truth of this matter.

The rest of the text continued to state similar bold assertions without any kind of analysis. So I asked it, "how can non-physical things arise from physical things ?" and it all but told me Plato's theory of forms :

 ...it is possible to separate a thing from its appearance. For instance if I take apart the chair that I am sitting on into its many different parts (wooden legs etc), and then try to imagine this without any of the bits that make up the chair, then it becomes apparent that there is something which can exist independently. 

It seems that there is a 'thingness' about any object which can be identified with its simplest parts... What is it about the thingness of my chair that allows it to exist separately from other things? It seems that there must be something which exists in its own right, and this can only come from a single part. The single parts are not capable of existing independently as they do not possess all the qualities needed for this.

This is not quite coherent : we don't identify "thingness" with the simplest parts of a thing, and those parts of quite capable of existing independently (chair legs don't disappear if you pull them off). But the essence of it, the "thingness", is clear enough.

The app clearly doesn't act as a consistent whole. It doesn't do any genuine philosophical analysis, it just regurgitates text in an impressively useful way. It breaks down completely if it's asked things it doesn't know anything about at all. Who is Lord Shang ?

Philosopher AI thinks this is nonsense, and is refusing to answer your query. It appears you will have to try something else.

Which is a bit of a shame because he deserves to be more widely known. How about something less profound : where do fish live ?

I think the best place to live for a fish is in water. It's not too wet or too dry and there is plenty of food. Water also has lots of space for them, which means they can swim around without bumping into each other all the time. But there are some drawbacks to living in water. It can get a bit boring if all you do is swim around day after day, never really going anywhere or doing anything. Another problem is that it's really hard to see anything underwater. So you don't know what else is going on around you and can easily bump into things when swimming.  

But I wonder if they even need to be able to see anything underwater. Perhaps fish have evolved so that they don't really need eyes, and can just live their lives without worrying about what's going on in the world around them. Anyway, in the end I think fish are probably happier living in water. Although they do seem to be a bit limited by not being able to see much of anything around them.

Ooo...kaaay.... so, we'll leave the marine biology to a dedicated app then. Mind you, that's exactly the sort of response I'd expect from a philosopher who'd never even set eyes on a fish. Good job team !

7 Online Fun Tools Created On GPT-3 - Analytics India Magazine

With GPT-3 slowly revealing its potential, it has created a massive buzz amid the ML community. While developers are trying their hands on some of the exciting applications of GPT-3, many are expressing their astonishment with the kind of possibilities it can bring for humanity.

Sophisticated gibberish is still gibberish

My news feeds have either gotten better at avoiding the hype train, or I'm just not clicking on the stories as much. But anyway, this "GPT3" thingy, the evolved version of that AI that was so dangerous it couldn't be released until it was (try it for yourself here, it's hilarious), seems to be causing the headline-writers to make all the usual mistakes. Of course, as this excellent article demonstrates, it's still nothing more than a glorified abacus. Sure, an abacus with access to oodles and oodles of data and sophisticated pattern-recognition algorithms, but an abacus nonetheless. Here's my favourite example of input (italic) generating truly ridiculous output (regular) :

At the party, I poured myself a glass of lemonade, but it turned out to be too sour, so I added a little sugar. I didn’t see a spoon handy, so I stirred it with a cigarette. But that turned out to be a bad idea because it kept falling on the floor. That’s when he decided to start the Cremation Association of North America, which has become a major cremation provider with 145 locations.

Gosh. You can find more examples in the article. And worryingly :

... it’s also worth noting that OpenAI has thus far not allowed us research access to GPT-3, despite both the company’s name and the nonprofit status of its oversight organization. Instead, OpenAI put us off indefinitely despite repeated requests — even as it made access widely available to the media. Fortunately, our colleague Douglas Summers-Stay, who had access, generously offered to run the experiments for us... OpenAI’s striking lack of openness seems to us to be a serious breach of scientific ethics, and a distortion of the goals of the associated nonprofit. 

Calling themselves "OpenAI" would seem to be a serious misnomer at best. The author here does their best to examine both sides of the argument, but ultimately it's clear that is not any kind of true intelligence : the mistakes made are such that a genuinely intelligent understanding could never allow them. Ever.

Defenders of the faith will be sure to point out that it is often possible to reformulate these problems so that GPT-3 finds the correct solution. For instance, you can get GPT-3 to give the correct answer to the cranberry/grape juice problem if you give it the following long-winded frame as a prompt: [prompt is rephrased to a multiple choice question]

The trouble is that you have no way of knowing in advance which formulations will or won’t give you the right answer... the problem is not with GPT-3’s syntax (which is perfectly fluent) but with its semantics: it can produce words in perfect English, but it has only the dimmest sense of what those words mean, and no sense whatsoever about how those words relate to the world. 

As we were putting together this essay, Summers-Stay wrote to one of us, saying this: "GPT is odd because it doesn’t 'care' about getting the right answer to a question you put to it. It’s more like an improv actor who is totally dedicated to their craft, never breaks character, and has never left home but only read about the world in books. Like such an actor, when it doesn’t know something, it will just fake it. You wouldn’t trust an improv actor playing a doctor to give you medical advice."

As I've mentioned before, the problem is we don't have a good understanding of understanding. To recall, my working definition is that understanding is knowledge of the connections between a thing and other things. The more complete our knowledge of how things interact with other things, the better we can say we understand them. Using knowledge of similar systems, if we truly understand them, we can extrapolate and interpolate to predict what will happen in novel situations. If our prediction is incorrect, our understanding is incomplete or faulty.

I noted that this isn't a perfect definition, since my knowledge of mathematical symbols, for example, does not enable me to really understand - let alone solve - an equation. Perhaps it's worth considering misunderstandings a bit more.

Sometimes, I can understand all the individual components of a system but not how the whole lot interact. If was really keen I could try and program a solution, and if my understanding of each part really was correct and complete, the computer would give a solution even if I myself couldn't. In such a case, I could probably do each individual calculation myself (just more slowly), but not the whole bunch all together. That suggests there's some numerical limit on how many connections I can hold at once. I could fully understand each one, but not how they all interact at the same time. I could guess how two gravitational bodies would orbit each other, but not very exactly - and with three I'm completely screwed.

This sort of simple numerical complexity limit shouldn't be any kind of problem to an AI : that's just a question of processing power and memory. Indeed, it could be argued that in this sense GPT3 already has a far greater understanding of the English language than any human : it knows all the possible valid combinations of words, all the correct sentence structures, etc. Its understanding is purely linguistic, but still a form of understanding nonetheless. You could even say it has a form of intelligence in that it's capable of solving problems, but in no way could you say it was "aware" of anything - just as we could say a calculator is intelligent but (almost certainly) doesn't have any awareness.

But subtleties arise even with this simplest form of misunderstanding. In a straightforwardly linear chain of processes, a computer may only increase the speed at which I can solve a problem. Where the situation is more of a complex interconnected web of processes, however, I may never be able to understand how a system works, because there may just not be any root causes to understand : properties may be emergent from a complex system which I have no way of guessing. As in An Inspector Calls, there may be no single guilty party, but only an event arising from a whole array of situations, events and circumstances. I might understand every single linear process, but still, perhaps, never fully understand the whole. I might understand each segment of an complicated equation, but not the interplay between each part.

The real difficulties would seem to arise from two issues, one easy, one hard. The easy issue is that GPT3 has no sensory input. It has no eyes to see nor tongue to speak but as its programmers are pleased to direct it; it experiences no sensations of any kind whatsoever. It will never know pain, or fear, or have an embarrassing orgasm. What do the words "cuddly kitten" mean if you've never cuddled a kitten ? Very little. It's trapped forever inside the ultimate Mary's Room.

In principle, this could be solved by connecting the thing up to some sensory apparatus : cameras, pressure sensors, dildos, thermometers, whatever. And this might help to some degree. The AI would then understand that the words are labels (parameters and variables in its code) corresponding to physical objects. It could learn for itself what happens if you try to stir a drink with a cigarette and thus never form any really stupid connections between stirring a drink and promoting a crematorium. It would surely be a lot harder to fool. And that would clearly improve matters considerably.

But this may not help as much as it might appear. For an electronic system, everything is reducible to numbers : the voltage level from the video feed, the deformation of the pressure sensor, the vibration frequency of the.... item. There is no obvious way such numbers can ever give rise to awareness, sensation, desires, emotion, or fundamental understanding, which are not measurable or sensible properties of the physical world. This is the Hard Problem both of AI and of consciousness in general.

So never mind the problems of comprehending a web of processes. The real difficulty is when we don't understand a single, irreducible fact. We can know that fact, but that's not the same as understanding it, not really. How do we program such a state in software ? Does it somehow magically emerge because we've connected a stream of numbers from an electronic camera ? It's bloody hard to see how.

Take "one". I know what "one" is. So do you and everyone else. But try and rigorously define it using language and you get Plato's torturous failure Parmenides. Creating a truly aware AI, at least through software, requires we define things we understand but cannot explain using language. By definition this is impossible, hence we will never program our way to a true AI. We might, perhaps, build one, but until we understand the true nature of consciousness, this will only happen by accident.

None of this means AI in the strict sense of only intelligence, not understanding and awareness, isn't useful. For as weird as some of the stuff that GPT3 is spouting, still the worst of it isn't half as incomprehensible as the effluence continuously spewing forth from the current incumbent of the White House. Far from disparaging OpenAI, instead I'm proposing a last-minute entry in this year's Political Apocalypse World Championship, a.k.a. the US elections. Good luck to 'em.

GPT-3, Bloviator: OpenAI's language generator has no idea what it's talking about

Since OpenAI first described its new AI language-generating system called GPT-3 in May, hundreds of media outlets (including MIT Technology Review) have written about the system and its capabilities. Twitter has been abuzz about its power and potential. The New York Times published an op-ed about it.

Stop destroying reality, Alice

Wow, blogging has really hit a dead zone of late, hasn't it ? Two main reasons for that. First, my long-awaited Oculus Quest arrived, and it's as awesome as I was hoping for. So quite a lot of my free time is now happily given to exploring strange new worlds and weird civilizations. Second, I'm deep in a recoding project which is going well. Normally I'm happy to break up my workday by reading random stuff and blogging up anything I think worth summarising and/or commenting on, but I don't like doing this when concentrating on code : it's too disruptive and takes too long to get back into the flow. Hence the blogging front has taken a backseat of late.

Not that this has stopped me reading stuff on my phone from time to time though. Lately my news feed has decided I need to learn about quantum paradoxes for some reason, which frankly was getting confusing until I found the one embedded below from Ars Technica. Other articles have much more complex descriptions, but I found this one much easier to follow. So let's resume the blogging with a classic long-term topic : the nature of reality.

Wigner grabs his friend, Alice, and places her in a sealed laboratory*. Alice measures the spin of a stream of electrons that are prepared in a superposition state. Wigner is outside the laboratory and will measure the entire laboratory. Alice, before passing out, determines that an electron is spin-up. But Wigner hasn’t made a measurement, so he sees Alice in a superposition of having measured spin-up or spin-down. When Wigner makes his measurement, hypothetically, he could end up with a result where Alice measured spin-down when in fact she measured spin-up. 

Two “facts” contradict each other, but both are based on reality. Wigner’s solution to this problem was that the quantum state cannot exist at the level of the observer: the superposition state must collapse before that occurs.

* Wigner is, of course, subsequently sued and made to watch the tea consent video multiple times.

If I understand this correctly, in essence Wigner "measures" Alice as having not taken any measurement, whereas Alice herself is quite sure that she damn well did take a measurement, thankyouverymuch. So there's a distinct contradiction about what happened, even worse than the simultaneity breaking of relativity where observers merely disagree about when something happened.

The article then goes on to describe an extended version, but after thinking about it, I'm quite sure I don't understand it at all. No matter, this original version is complicated enough. Does it challenge something as fundamental as the nature of reality ?

Physicists generally describe reality by a set of mathematically defined conditions. For instance, causality tells us that an effect should be preceded in time by a cause. Locality says that causes propagate at the speed of light: if a photon cannot travel between the location of the cause to the location of the effect before the effect occurs, then it violates locality (and, potentially, causality). The researchers define the absoluteness of observed events, meaning that what I observe is real and does not depend on anything else. They assume that there is no super-determinism (we make free choices) and locality is still operative. The researchers refer to this trifecta as local friendliness.

Call me crazy, but "local friendliness" sounds like a bloody daft term, but that's science for you. Anyway :

They show that under the right conditions, correlations that violate these limits will be observed in the extended Wigner’s friend experiment. Their laboratory experiments confirmed that these violations do in fact occur. Local friendliness is not how the Universe operates. 

If we reject local friendliness, then we have to make some decisions. We have to accept some of the following possibilities: an observer's measurements are not necessarily real, reality is not local, super-determinism is real, or quantum mechanics ceases to function somewhere before macroscopic observers get involved.

I think I understand what's meant by reality being non-local : FTL effects meaning that things could affect other things despite no apparent connection. Super-determinism is harder due to the phrasing used : does the article mean to say this assumption means we do or not have free will ? I would presume that if there's no free will, and if everything is predetermined (including the measurements) then there's no problem, so let's go with that.

But I don't understand at all what's meant by "an observer's measurements are not necessarily real". Of course measurements can be erroneous, so I assume it's something more profound than that... but how ? If I follow the operating procedures correctly, but my equipment has an unknown fault and gives me an error, in what way is this different from the quantum version in which I get a result that contradicts someone else's ? It would appear to mean not that some things are fundamentally unreal but only that some things are fundamentally unmeasurable. But this is nothing new or unique to quantum systems; some things are fundamentally unquantifiable. You can't objectively measure mercy, or justice, or yellowness, or guilt, but clearly their mental existence doesn't invalidate the reality of what they're measuring. Measurements themselves have no physical substance so in that sense they can never be said to be "real". If there's something I'm missing here, I'd very much like to know what it is.

Likewise there's this interesting article about how long it takes the wave function to collapse :

By taking photographs to see what happens in that one-millionth of a second, scientists saw that the decision, a process referred to as wave function collapse, took some time to happen. It’s like when the police officer points her radar gun at you - your car is first going somewhere between 20 and 60 miles an hour, then between 35 and 50, then either 40, 42, or 45, then finally deciding it is going 40 miles an hour. The intriguing results show that quantum collapse is not instantaneous. It also shows us how time operates on the quantum level - and shows us that time itself may be a blurry, abstract concept. It also shows us that our concept of “now” may not really exist, and that our reality is a very weird place indeed.

Again I'm not seeing why this brings the nature of time into question. Okay, in some sense the "reality" of the electron takes some time to coalesce from wave to particle, but why would you expect it to be instant ?

I often think that a lot of this quantum jargon about wave functions and decoherence and suchlike is not much better than ancient ill-defined concepts like magic or humours... I don't think we're any closer to understanding the fundamentals of reality than they were. Sure, we've got far better cosmological models and infinitely superior understanding of gravity, material physics, chemistry, medicine, etc... but the aspect of, "what's it all about, when you get right down to it ?". Not so much.

Quantum reality is either weirdly different or it collapses

Quantum mechanics, when examined closely, poses some deep questions about reality. These questions often take the form of thought experiments, which are later (usually much later) followed up by real experiments. One of the most difficult and deepest of these is a thought experiment proposed by Eugene Wigner in the 1960s, called "Wigner's friend" (you don't want to be Wigner's friend).

Review : Witchfinders

They said that if they can make you believe absurdities, they can make you commit atrocities. They're probably right, but as Malcolm Gaskill explores in his excellent book Witchfinders, you need much more than an absurd idea before you go around hacking and burning your way through the unbelievers. Otherwise the entire cast of Teletubbies would surely be long dead...

Besides general interest, two things persuaded me to give this go. First, there's the outstanding prose, which, when the author puts his mind to it, is as good as any descriptive sequence in a gripping novel. Second there was a quote in the introduction which caught my attention :

This is a book that reaches beyond monstrous stereotypes to a constituency of unremarkable people, who, through their eager cooperation with Hopkins and Stearne, themselves became witchfinders. In this they resemble the provincial nobodies of the twentieth century who engaged in genocide, demonstrating to the world the banality of evil.

The only thing I don't like about this book is that it has rather long passages that are somewhat unnecessarily similar. A slightly broader brush would help; it's not necessary to know the names of every single person involved or the often-trivial differences between the cases, as this sometimes makes things tedious and hard to follow. I'm still giving it 8/10 for being generally excellent though.


I'm a big fan of history books that try and dig a bit deeper than simply recounting events. I'm also a big fan of those which don't have to be explicit in drawing lessons - I want to hear the author's interpretation but I don't necessarily want it rammed down my throat. This book does things perfectly. The bulk of it is given over to describing in some detail exactly what happened, sticking largely to a clear and direct narrative. It only deviates from this to dispel the occasional popular myth (e.g. witches in England were, if ever given a capital sentence, almost exclusively hanged, and only very rarely indeed were they burned at the stake), describe uncertainties and gaps in the historical record, and provide some very infrequent commentary. Conclusions, judgements and interpretation are left almost entirely to the final section, by which time they're already unmistakable thanks to the author's careful curation of the main text.

Nobody but an idiot would deny that religious and superstitious zealotry played an important part in the tragic events of the mid-17th century. It undeniably did. People genuinely believed that lonely old women could bend demonic forces to their own malevolent will. But, contrary to popular belief, to transform this belief into action required much more than making a simple accusation. In virtually all situations, this would fail. Only in the right context could the rampant paranoia and fear explode into a bona fide witch-hunt, and Gaskill does a magnificent job of explaining what went so catastrophically wrong in c.1640 : the chaotic and divisive background of the Civil War, the particular characteristics of Matthew Hopkins and John Stearne, the series of natural disasters (plague, famine and flood). He brilliantly and deftly casts the period as a period of intense ideological, not merely political, change and evolution.

One of the first interesting things that becomes apparent is just how similar the stories are. Rather than cauldron-stirring old crones hatching evil plots to eat babies, most of the cases were of confessions (obtained under the suggestibility and torture of sleep deprivation) of consorting with "imps" (animal familiars) who claimed to be able to enact petty revenge against usually minor slights, often unsuccessfully.

As befitted the general pettiness of their alleged crimes, so the "witches" themselves were, almost exclusively, the weakest and most vulnerable in society, little more than all-too convenient scapegoats for simple bad luck in a trouble society. Satan ? He gets "a walk-on part", at most, as do any suggestions of more serious, national-level crimes; the supernatural powers supposedly at work were usually remarkably ineffective. Gaskill suggests that one reason for confessions, besides the highly suggestible state induced by days of sleep deprivation, was that the victims felt that they could at last strike back against their oppressors. Hated by their communities for the most marginal of differences, belief in devilish aid and the chance to make their tormentors afraid of them would give them a sense of much-needed empowerment.

Initially, the minor misdemeanours most claimed to have exacted - spoiling crops, killing a cow and so-on - would not have been grounds for capital punishment : a witch would have to have been found guilty for, say, murder, not just talking to imps. Only later, when the Civil War began to fuel extreme paranoia, was merely consulting with devils itself a hanging offence.

The lawlessness of the Civil War begat another element besides needing to assign blame. It was common knowledge even among the most learned of the era (including most famously King James) that witches were real, and the judiciary fully accepted this. The reality of witches was never in doubt. But with war and disease running rampant, there was a mood amongst the commoners that the legal system was failing to offer adequate redress for the problems they sincerely believed witches were causing : in some cases, they honestly thought their children had been murdered by witches out of pure devilish malice.

So this is the realm where witch-hunting flourished. The most illiterate peasant would know for certain, as everyone did, that witches were real and malevolent : his king and scholars had told him so. But while the learned may have advocated for strict controls to ensure accuracy, a lowly peasant was not likely to be in the mood for such diligence. The Civil War had ravaged the land - a conflict itself as much religious as it was political. Crops had failed and the world grew darker. The ordinary citizens knew witches were abroad, but the war meant that precious little was being done about them. In earlier, most solidly-governed times, nothing much would have happened : occasional witch trials came and went and that was pretty much the end of it. But now, someone aspiring to make his mark in society could do far worse than rally the populist flag against the witches.

This is the final element that made the terrible events of c.1645-1647  possible : the witchfinders themselves. Gaskins paints Hopkins and Stearne as opportunistic, vain, shallow, and self-righteous, but also utterly sincere in their beliefs. Their charismatic self-proclaimed expertise was enough to tip the balance and let the desire for "justice" against the witches at last boil over... at least somewhat. It's hard not to mix metaphors, because even though Hopkins and Stearne brought over a hundred women to an untimely demise, they did not have things all their own way. Resistance against them was usually successful and caused them to bid a hasty retreat; they had conviction but without much courage to back it up. Notably, they themselves never made accusations, but only supported existing accusations when they were called to assist. They did not start fires, but only fuelled the already-burning flames, bringing out the very worst qualities in the their fellow men. They simply could not have succeeded without a society that, at some level, saw a need for them.

If the common beliefs of the day were undeniably at fault, the oppressed citizen was not totally without protection. Despite all the faults of the system, even a 17th century law court would typically have been hard-pressed to convict and hang, let alone burn, a witch. It was only through this peculiar combination of circumstances that witch hunting briefly exploded : in more normal times, the beliefs of the commoners could be successfully held in check. When the tide began to turn against the witchfinders, their efforts were soon thwarted. Ironically, Hopkin's very success was seen as evidence that he himself must be in league with the devil to find so many witches. He issued a rebuttal against his detractors, but it backfired horribly. He met an ignominious end and escaped justice by dying of illness as his campaign sputtered to a halt; his compatriot Stearne largely faded into obscurity when it was clear the public mood had shifted. Witch-hunting was, arguably, as much a consequence of horror than a cause of it. Gaskill concludes with a warning :

It was a terrible tragedy, but it needs to be seen as part of something even more terrible, a civil war characterised by bigotry, brutality and bloodshed. The conflict killed 190,000 Englishmen out of a population of five million... If one discounts the sensational, the particular and the judgemental, one is left with a different kind of Matthew Hopkins : an intransigent and dangerous figure, for sure, but a charismatic man of his time, no more ruthless than his contemporaries and, above all, driven by a messianic desire to purify.

How different are we in mentality from our seventeenth-century ancestors ? - a question that becomes even more taxing if "seventeenth-century ancestors" is replaced with "fellow human beings in Africa and India". The truth that many find unpalatable, even inconceivable, is that in our ideas, instincts and emotions we are not very different at all. Without peace and prosperity, liberty and welfare, and the political and economic stability on which those things depend, the thinking of the next generation in the West might swerve off in an altogether more mystical and malevolent direction... the seventeenth-century tragedy is only partially that of Matthew Hopkins, the flawed protagonist, and the harrowing deaths of his victims. It is at least as much a tale about feeling anxious and vulnerable in an indifferent world - a sensation of humanity.

Witchfinders

Witchfinders book. Read 58 reviews from the world's largest community for readers. By spring 1645, two years of civil war had exacted a dreadful toll upo...