Jack the Ripper and the Case for Scotland Yard's Prime Suspect - Rob House

[fido]

My thanks to Wolf Vanderlin. I will append an account of my work showing how it was certainly not a matter of looking for Kosminski, finding him, and deciding I needed an alternative.

Without prejudice to Rob's book or thinking (which I haven't read) may I say that John Douglas formed his conclusions about David Cohen from my writing and before he or Mark Olshaker had met or communicated with me. When Mark came to London on other business and traced me, he asked me to read their ms, and I made a few corrections - mainly in the direction of disclaiming good discoveries or arguments I hadn't actually made.

And of course Rob is entirely right to say Kosminski wasn't an imbecile. I don't know why in 1890 Leavesden was described as the Asylum for Imbeciles.

Now to my work in 1986-7.
I did not find Kosminski, see that he didn't work, and put together a theory based on other names I had found or went looking for.

I found Kaminsky in Black lion yard - one of the variants of the name given by Farson, who by that time could no longer remember why he was uncertain whether the handwriting said Kosminski, Komansky or Kaminsky. (Today we know it's because Lady Aberconway's page is damaged at that point).

Richard Whittington-Egan was very enthusiastically encouraging because

(a) he had always thought Andeson's suggestion could have merit, only it
seemed absolutely impossible to put any flesh on it, and

(b) being of the generation which remembered Darky the Coon and Barnet Whatsisname and the other Jewish gangsters of that time, he was not impressed by the idea of a universally law-abiding ethnic group in Whitechapel.

More important still, he saw the beauty of Black Lion Yard as an address - (much more truly central to the murder sites than Plummers Row or Sion Square, though those addresses were unknown at the time). So I started by trying to find out more about Kaminsky with Kosminskys, Komanskys and similar names as possible alternatives. (In scanning my old notes for How the other day I noticed, what I'd forgotten, that I was very attentive to Golinskys at that time).

Bear in mind that I started my work with three "clues" (over and above my perception of the identity of Macnaghten's Kosminsky with Anderson's Polish Jew, and the surprising discovery that Anderson was not the empty liar previous writers had represented him as being). Based on existing Ripper writing, these were:

1. There had to be a reason why the murders stopped when they did

2. The murderer left Mitre Square in the direction indicated by the Goulston Street apron

3. There was something fishy about the complete dropping of all Leather Apron based enquiries after the clearance of Pizer, who had been falsely identified by Violenia, and of whom one writer said the accusing streetwalkers refused to make any identification. All this looked so peculiar that you may remember Don, at that time, proposed Pizer as Anderson's suspect and Violenia as his witness.

Point 3 explained itself in my research, which indicated that the only person known to have said he knew Pizer as Leather Apron was Thick, and before the inquest he did not declare this as certain knowledge. I'm so bucked by Rob McLaughlin's work because he has established what I had always suggested, that in some respects the overall dodgy Dorsenne was offering genuine inside information, and this was in the area where he ascribes the Leather Apron identification to Thick.

Point 1 is explained by the incarceration of Cohen, and no other Jewish asylum inmate. Even with today's knowledge that serial killers will stop if they know they'll really be caught if they continue, Kosminsky was clearly too far detached from reality to have made such a rational choice.

Point 2 strongly supports Kaminsky in Black Lion Yard. And so, draft one, in the press in my original ms, said Kaminsky was David Cohen, though I was making fairly desperate guesses as to how the name came out that way .

Before this reaches print, I've found Kosminsky. It's obvious to me he wasn't the Ripper and there's absolutely nothing about him to shake my prior conviction (or antecedents, as Macnaghten would say).

You see the very important difference between finding Kosminsky, saying "Tcha!" and looking for an alternative?

After I've published, the supportive evidence comes in. Cohens from America and Italy contact me to say their family name was bestowed by Gentile immigration officers who couldn't cope with names like Keithowi or Martinowi.

And then the Swanson marginalia appeared. Now we have a police document saying that Kosminsky lived at his brother's house in Whitechapel (true) but was under restraint when taken to the infirmary (not true of Kosminsky, but true of Cohen's transfer to Colney Hatch). It also says he dies soon after going to Colney Hatch (amazingly untrue of Kosminsky: uniquely true of Cohen among young or relatively young Jewish patients). There is clear support for my argument that the two men, so similar in simple paper descriptions, had been confused.

Of course I can't prove that I only saw the probability that the two forces followed the two different Whitechapel Jews as I was trying to puzzle out why on earth the marginalia contained false information, and not out of any attempt to use them to prove my own previous thinking. But I hope my willingness to acknowledge mistakes when I've made them will make you trust me. American police officers (like John Douglas) who are very familiar with overlapping jurisdictions, find it very convincing.

I would only add that neither I nor anyone else, as far as I know, regards Anderson as an impeccable witness. I said of him from the outset he "may have been wrong. He was always opinionated." But he is the best historical witness because unlike almost all others who identified suspects, we cannot find factual error in his statements. (This is also true of Littlechild, but we have no idea why he thought Tumblety a good suspect. Whereas in Anderson's case, as Philip Sugden pointed out, we have the unique suspect with a piece of definite - if puzzling - evidence given to establish the reason for suspicion: the mysterious identification at the Seaside Home.

Martin F

[mariab]

French profiling of Ripper-like killers

I apologize most profoundly for highjacking this thread and I have no idea if anyone's interested in this at all, but I'm presently going through the Alexandre Lacassagne collection in Lyon, which is abundant, as Lacassagne was extremely prolific.
Pertaining to early 20th century profiling of serial killers, Lacassagne has kept super abundant notes about both Vacher and Henri Vidal. (The latter is another serial killer/murder of prostitutes, caught in 1902.) There is enough material here in Lyon for someone (fluent in French) to write a very interesting book on early profiling of serial killers, and in my opinion I'd even dare go as far as to say that anyone wishing to write a “suspect book“ on the Ripper, especially one using “profiling“, perhaps ought to consult these French materials; since both Vacher and Vidal committed similar crimes as the Ripper and were caught and studied in great detail, both psychologically and medicinally, and both alive and postmortem. What's interesting is that Vacher featured a psychological profile quite a bit different than Vidal.
I thought that this might perhaps interest Mr. House or anyone involved with early profiling?
By the by, Vacher's brain (examined by Italian luminary Lombroso and by other doctors) contained some “white particles“ (not lesions) typical for mentally retarded people, even if his brain was unusually developed (developed meant as in active), plus he was hormonically defficient (as in not able to have kids), and Vidal was impotent.
Pertaining to Vacher there are 3 references by Lacassagne to JTR, one of them discussing Neill Cream as a suspect. Plus I've got a picture of the knife Vidal used, and it's not big at all. If anyone's interested, I can post these.
With many apologies for highjacking the thread with this.

[jason_c]

Just what sort of police department would actually condescend to having a moron, such as yourself, on its payroll?

This seriously made me laugh.


Im thinking Keystone Cops, C.I.D.

[Colin Roberts]

I mentioned the names of the two top experts in the field – Rossmo and Canter – for a reason.

Experts within the realms of their own 'creations', I suppose.

Should we describe Amy Winehouse as having been an 'expert' within the realm of her particular musical creations?

Both these men have done geographic profiles based on the Ripper murders, and both have published their findings.

Where have they been "published"?

As for Rossmo; I don't believe that a couple of sentences regarding the "peak of the profile", as stated in a television documentary, and as written in an article, amounts to the publication of his "findings".

As for Canter; ...?

Neither of these experts came to the conclusion that the Ripper lived anywhere near Greenfield or Yalford Streets.

No geographic profiler would ever come to the "conclusion" that 'Jack the Ripper' lived anywhere, in particular! Period!

Where, exactly, within the probability distributions that were generated by Rossmo's and Canter's respective profiles, do Kosminski's possible Greenfield Street and Yalford Street addresses happen to fall?

Within the ninetieth percentile? The eightieth? The seventieth? ...?

I don't know!

Neither do you!

But, in any case, Aaron Kosminski, in all likelihood, resided within the observed parameters of the 'killing field' of 'Jack the Ripper', between 7 August and 9 November, 1888; and that is plainly and simply all that needs to said.

It's really all that Rob should have said.

I believe that Rob allowed himself to drift into depths, which precluded his being able to 'touch the bottom', so to speak.

But, you have done precisely the same thing, in your criticisms of Rob's analysis.

~~~

Some observations that I made recently, at JTR Forums.com:

... we should perceive a probability of just 22 percent that 'Jack the Ripper' resided in closer proximity to the murder-site mean-center, than did Aaron Kozminski. This, in turn, would suggest that Kozminski's most likely 1888 residence, should be perceived as having fallen within the 78th percentile of the distribution of probability, pertaining to the location of the residence (or, base of operations), of 'Jack the Ripper'.

78th percentile? Is that all?

Actually, that's pretty damn impressive!

Kozminski plainly and simply has geography 'on his side'!

"78th percentile? Is that all?"

~~~

"Kozminski plainly and simply has geography 'on his side'!"

Indeed, he does!

But, I am afraid that I must take issue with Rob's use of geographic profiling, to bolster "... the Case for Scotland Yard's Prime Suspect".

I do not believe that any form of criminal profiling, should be used to either bolster, or weaken the case, for a particular suspect.

In fact, I do not believe that a suspect's adherence (or, lack of adherence), to any sort of 'profile', should be perceived as being 'circumstantial evidence'.

Criminal profiling is intended to facilitate a prioritization of investigative focus.

In other words, it is intended to provide the investigator a 'standard', by which potential persons of interest can be categorized and/or arranged in order of ascendency.

In the absence of any tangible incriminating evidence, potential persons of interest can be brought to the attention of the investigator, by way of various leads, or through 'canvass'/'dragnet' efforts, - such as the Metropolitan Police Force house-to-house search, of October, 1888 - which should, themselves, be conducted, in accordance with the prioritizations that are facilitated, by applicable geographic and/or socio-economic profiles.

I have argued repeatedly that the case for Montague Druitt should not be called into question, on the basis of geography; even in light of the fact that my Geographic Profile Model would suggest a perceptual probability of more than 99 percent that the 1888 residence of 'Jack the Ripper' was in closer proximity to the murder-site mean-center, than was Druitt's home, in Blackheath.

All forms of criminal profiling notwithstanding, there is a case to be made for Montague Druitt having been 'Jack the Ripper'. Period!

Questioning this case, on the basis of geography, would be tantamount to questioning the qualifications of a Doctor of Philosophy (Ph.D.), on the basis that he managed to bypass high school, and go straight to Oxford.

Similarly, ...

All forms of criminal profiling notwithstanding, there is a case to be made for Aaron Kozminski having been 'Jack the Ripper'. Period!

Attempting to bolster this case, on the basis of geography, would be tantamount to attempting to bolster the qualifications of a Doctor of Philosophy (Ph.D.), on the basis of his high school transcripts.

All that Rob needed to say, was that Aaron Kozminski, in all likelihood, lived within the 'killing field' of 'Jack the Ripper', during the latter months of 1888. That's it!

And my own comment on those complex computer programs you refer to is this. They rely on essentially heuristic mathematical models constructed by analysing the behaviour of modern serial killers, very few of whom (if any) operated on foot over distances of only hundreds of yards. So because the models are fundamentally nonlinear, they are completely inapplicable to the Whitechapel murders. A simple method like Canter's may give some kind of rough indication, but these complex models - applied way outside their range of validity - really can't tell us anything useful at all.

Canter's so-called 'Circle Hypothesis' / 'Circle Theory' is plainly and simply too fundamental; whereas complex models of geographic-profile analysis, such as Canter's proprietary Dragnet and Rossmo's proprietary Criminal Geographic Target, in my opinion, involve excessive manifestations of TMI (i.e. too much information) that tend to 'milk' the spatial relationships that exist within a set of murder-sites, for infinitely more than they are worth.

But, wouldn't you agree, Chris, ...

"very few of whom (if any) operated on foot over distances of only hundreds of yards"

... that dispersion is relative; whether measured in yards, ... or in miles?

I spoke with D. Kim Rossmo on the phone and he basically agreed that Kozminski fit the geographic profile.

As it plainly and simply goes without saying that he would!

This is not an exact science obviously. But I don't think any geographic profiler would tell you that the Ripper lived exactly on the corner of Flower and Dean Street and Thrawl Street. The profile generates a heat map (http://www.wesleyenglish.com/wp-cont...the-Ripper.png) which shows areas in which the killer might have lived, statistically speaking. I mean, how much accuracy are you expecting? 50 feet? 100 feet? It is not that precise.

Yes, this is another relevant point, which I should also have mentioned above. People often look at these coloured spatial probability distributions and assume that the model is predicting that the killer lived at the point where they can see a red "hot spot."

But all a hot spot means - in the model's own terms - is that the probability density is higher than elsewhere. For example, it might mean that the probability that the murderer lived in a particular street is 6%, where it would be only 3% if the probability density were more evenly distributed.

Because the area where Aaron Kozminski's family lived is central to the murders, it really will be the case that any of these models will predict a reasonably high probability density in that area (even if there isn't a "hot spot" nearby). In that sense, the models will all be consistent with the killer having lived in, say, Yalford Street.

The problem is that they'll also be consistent with many other possibilities. But that's "the nature of the beast."

If only there were a way of getting the field of 'Ripperology' to understand the most fundamental principles of a probability distribution.




Hypothetical Probability Distribution (Elliptical) (Click Image, to Enlarge in flickr)
Underlying Aerial Imagery: Copyright Google Earth, 2010
Overlying Plots, Labels and Color-Shadings: Copyright Colin C. Roberts, 2011

Let's pretend that the above imagery is a Doppler Radar depiction of the most concentrated portion of a cloud of precipitation that happens to be hovering over London's inner 'East End'.

Let's also pretend that each of the color-shaded isopleths represents exactly ten percent of the rainfall that is occurring, and that the density of each set of ten percentage points is greatest at the inner elliptical contour of each of the respective isopleths, and least at the outer elliptical contour of the same.

In other words, let's also pretend that the density of the depicted rainfall is greatest at the center of the overall elliptical distribution (i.e. the intersection of Thrawl Street and George Street, in the Parish of Christ Church Spitalfields), and least at its periphery.

Now, let's pretend that somewhere within this distribution of raindrops, there is a single Golden Raindrop that is proving to be quite elusive.

I would contend that its single most probable location is the center of the overall elliptical distribution of rainfall (again, the intersection of Thrawl Street and George Street, in the Parish of Christ Church Spitalfields).

But, does that imply that its probable location is the center of the overall elliptical distribution of rainfall?

In other words, does that imply that its location is 'probably' the aforementioned intersection of Thrawl Street and George Street?

Absolutely, positively not!

Under no circumstances, whatsoever, does that imply anything of the sort!

In fact, the probability that the elusive Golden Raindrop is to be found, specifically, at that intersection, is so low that it is effectively zero percent.

So, what then, must I say, in order to define the area, in which the elusive Golden Raindrop is probably to be found?

Any takers (other than Chris Phillips)?

As my writing prowess has been acknowledged I am now contemplating a new book titled "A Pin stickers guide to the world of criminal profilers"

Do tell:

Just what sort of police department would actually condescend to having a moron, such as yourself, on its payroll?

[Roy Corduroy]

Martin decided to slide over from Kosminski and choose another Jewish candidate...

No, Martin found Cohen first, because he searched the records around the time of the murders. A logical place to start. Only by searching records from later did he find Kosminski. He didn't slide at all.

Rob I like your book. It is a suspect book. A good suspect book stays on track, as you did. Thank you for explaining and also including a visual chart of the family members. Who immigrated and their particulars. This cast Aaron in relief, as the others married, worked, changed their names, etc. That is much clearer to me now.

Also you raised a key point, that Aaron was at no time classified an imbecile. That Leavensden had two distinct classes of patients.

All modern JtR suspect books utilize some modern methods, such as criminal and geographic profiling, reverse diagnosis based on current knowledge, and comparison with those on the infinite list of serial killers since. And so forth. And a lot of it is a minefield because it leads to various interpetations and forks in the road.

Sometimes simpler is better. For instance, your area map, Rob, was easy to understand. It showed the murder sites, and there was Kosminki's abode, "bang in the heart of the district" to borrow a phrase. It says more than explaining geo-profiling.

A satisfied customer,

Roy

[Robert]

I know nothing of profiling, geographical or otherwise. But I would just like to raise the point that in a population like that of the east end, surely the chances of the murderer moving from house to house, or lodging house to lodging house, could be quite high, even within a period of a few short weeks. So IF the killer was on the move, what price the profiling?

[Chris]

This is not an exact science obviously. But I don't think any geographic profiler would tell you that the Ripper lived exactly on the corner of Flower and Dean Street and Thrawl Street. The profile generates a heat map (http://www.wesleyenglish.com/wp-cont...the-Ripper.png) which shows areas in which the killer might have lived, statistically speaking. I mean, how much accuracy are you expecting? 50 feet? 100 feet? It is not that precise.

Yes, this is another relevant point, which I should also have mentioned above. People often look at these coloured spatial probability distributions and assume that the model is predicting that the killer lived at the point where they can see a red "hot spot."

But all a hot spot means - in the model's own terms - is that the probability density is higher than elsewhere. For example, it might mean that the probability that the murderer lived in a particular street is 6%, where it would be only 3% if the probability density were more evenly distributed.

Because the area where Aaron Kozminski's family lived is central to the murders, it really will be the case that any of these models will predict a reasonably high probability density in that area (even if there isn't a "hot spot" nearby). In that sense, the models will all be consistent with the killer having lived in, say, Yalford Street.

The problem is that they'll also be consistent with many other possibilities. But that's "the nature of the beast."

[Chris]

But, if I might say so, for me personally (rather than as an editor of Ripperologist), the Anderson/Fido/Kosminski/Cohen/Kaminsky conundrum rather points out the fallacy of Anderson's theory. Martin decided to slide over from Kosminski and choose another Jewish candidate much like the FBI profilers said that the killer could be Kosminski or someone very like him, which rather indicates that Kosminski (no first name) is a straw man rather than an actual living person who might have committed the murders. In other words, Anderson was pointing to a possible candidate who could have done the murders rather than relying on any solid evidence that Kosminski was the Whitechapel murderer.

That's exactly what Anderson wasn't doing, according to his own account.

Anderson claimed that there was definite evidence that a particular suspect was the murderer. Specifically, he claimed that the suspect had been identified.

Personally I don't believe Anderson was right. But in any case he wasn't "profiling" the killer. He wasn't saying the killer was "very like" this man. He certainly wasn't pointing to a generic Jewish suspect, as his comments in response to Leopold Greenberg's criticisms made clear. He was saying that it had been proved that the killer a particular man (who happened to be Jewish).

[ChrisGeorge]

The same basic reasoning went for other things. I decided to not go into Martin Fido's theory of Cohen/ Kaminsky simply because I do not think it is a very good theory. Am I really expected to have to debate things like that? I did not see that it was my responsibility to have to debate every idea that is tangentially connected to Kozminski. I commend Martin for his work and research in finding Kozminski, but I do not agree with his conclusions. I did not see the point in getting sidetracked debating a theory that is barely even comprehensible to me.

Hello Rob

Thanks for your well written explanation of the aims in your book, in not getting into the more controversial areas in regard to Anderson's advocacy of Kosminski as the likeliest suspect to have been the Ripper. Your intentional reticence of getting into the debate about Anderson's reliability or possible delusions and/or senility is understandable, also your decision to stay away from Martin Fido's decision to plump for another suspect other than Aaron Kosminski.

But, if I might say so, for me personally (rather than as an editor of Ripperologist), the Anderson/Fido/Kosminski/Cohen/Kaminsky conundrum rather points out the fallacy of Anderson's theory. Martin decided to slide over from Kosminski and choose another Jewish candidate much like the FBI profilers said that the killer could be Kosminski or someone very like him, which rather indicates that Kosminski (no first name) is a straw man rather than an actual living person who might have committed the murders. In other words, Anderson was pointing to a possible candidate who could have done the murders rather than relying on any solid evidence that Kosminski was the Whitechapel murderer.

Best regards

Chris George

[robhouse]

Thank you for your comments Jonathan,

I have already replied to you in a private message on some of these points, but since you raised them here, I will reply again in this open forum.

In brief, let me just say... yes certain things did not make it into the book. It is indeed difficult to juggle and wrangle all the facts into a narrative form. First off, I didn't want to be boring, so there is a bit of a dramatic tension that I was very conscious of putting in. This is perhaps a bit cheesy in spots, but again, I was gearing the book at a general audience, in addition to the Ripperology community.

Second.. I tried my best to be objective, and to bring up counter points when I thought it was appropriate. I did for example leave a lot open for various interpretations on certain points. But there were certain topics that I intentionally left out.

For example, I did not delve into all the extensive debate over Anderson's character, and whether or not he would have bragged or lied, or misremembered things. I left this out because:

1. I do not agree with these ideas, nor do I agree that the premise for raising such questions is particularly valid. In short, nobody would have bothered to attack Anderson if he had not made his claim about knowing the Ripper's identity. Hence the question of whether Anderson was addle-headed was motivated by a prejudiced idea... specifically that "well, he must be wrong." So people have mainly just pondered, "why did he make such an idiotic and deluded false claim?" Well how do we know he was wrong? People seem to have assumed he was. As I have said, my book explores the possibility that he perhaps was not wrong, was not lying.

2. I also thought this stuff would bog down the narrative, and be very dull—especially to someone new to the case. If I had included all these points, then argued against them, a general reader might have been thinking, "Why is he going to all these lengths just to say that the guy was not addle-headed, or a braggart, or senile?" Do I really have to introduce a point I disagree with, and which I do not even think is a particularly fair question, just so I can dismiss it?

3. In short, these things have been debated endlessly, and have gone nowhere. My opinion would not be new to anyone on casebook, so why bother printing it in a book. In short it comes down to this: nothing new for Ripper experts, boring to general readers, and I don't even think it is a particularly fair or valid argument to bring up. So I decided to not include it.

The same basic reasoning went for other things. I decided to not go into Martin Fido's theory of Cohen/ Kaminsky simply because I do not think it is a very good theory. Am I really expected to have to debate things like that? I did not see that it was my responsibility to have to debate every idea that is tangentially connected to Kozminski. I commend Martin for his work and research in finding Kozminski, but I do not agree with his conclusions. I did not see the point in getting sidetracked debating a theory that is barely even comprehensible to me.

The same basically goes for what you refer to as the "sublime" Sailor's Home theory. Much as I respect Stewart Evans and consider him a friend and even a mentor in some sense, the Sailor's Home theory makes no sense to me at all. I can understand a bit of confusion here and there, but this simply requires too much confusion, it is too convoluted.

But that is not the point. The main point I am trying to make is that I consciously decided (and felt I had the right to do so) that I was not going to have to discuss and debate all these various counter arguments that have popped up over the years. The ones that were most relevant and most pertinent to Kozminski... and the ones that I thought were most valid... I included.

In addition, as I mentioned before, my original draft was 40% longer than the final book. So in other words the book would have been close to 500 pages in its original form. I had to go through deleting everything that was not absolutely essential to the story as I wanted to tell it. That is not an easy thing to do. The publisher actually wanted to cut out all the "bring history stuff"... ie. the stuff about Poland and Russia. And I had to fight to keep that in.

So yes there were some omissions, some errors. It was an error to not mention the Lawende thing, and as Wolf pointed out, that Douglas later changed his opinion on Kozminski fitting the FBI profile (in his opinion... not Roy Hazelwood's).

RH

[Jonathan H]

To write a book, whilst working full or part time, is a tremendously challenging undertaking.

As Rob writes, he was under pressure from publishers, and no doubt a lot of other considerations too. You cannot include everything, and bits and pieces fall by the wayside. Ienvitable mistakes creep in.

But any non-fiction book, or fiction for that matter, has to be judged on its overall merits.

I congratulate him for getting his book published and I urge anybody interested in this subject, and/or true crime, to purchae the book as they will find it a stimulating read, whether they agree with the author's [provisional] conclusions or not.

A number of people have already praised it as one of the very best books on this subject and, again, people should buy it and make up their own minds, as reading a book is just as individual an experience as writing one -- bar the assistance one receives from living sources.

On the other hand, I think that this book is for the lay person on this subject as it is so biased, to its core.

Not that a lay person would not enjoy it -- probably they would enjoy the elegant, understated prose, and the exemplary research into Jewish immigrant life very much -- but that the average reader would be unaware that this is a 'stacked deck' on behalf of one side of an argument.

That the book's analysis of certain primary and secondary sources, regarding the police suspects, is -- for me -- excruciatingly narrow to the point of being like a lawyer's brief, rather than what the best works of history do: present alternate arguments in order to show how they can be refuted, or at least to try and refute them.

Then leave it up to the reader to make an assessment of the author's assessment.

To be specific:

1. Stewart Evans is approvingly quoted, but not his and Don Rumbelow's 'Sailor's Home' sublime theory (2006) that Anderson and/or Swanson is misremebering the failed identification of sailor Sadler by a witness, almost certainly the German-Jew Lawende.

Rob House does not have to agree with this controversial theory. Of course not. Obviously he does not. But it would strengthen his argument if he outlined their theory, and then put a counter-argument to it. By leaving it out he prevented his book, for me, from being definitive about Kosminski, compared to the 'Did Anderson Know' chapter in Scotland Yard Investigates'.

2. Due to an understandable oversight caused by those pressures, mentioned above, in writing a book on a very complex subject, the 1891 claim that [almost certainly] Lawende was brought in to 'confront' Sadler, is not mentioned at all.

In effect, both the primary and secondary sources which would give the reader the opportunity to ponder such an alternative to Anderson and/Swansons's claims, and thus an alternative to Aaron Kosminski as 'Scotland Yard's Prime Suspect', do not exist in this book. They have to be found elsewhere, though the lay reader would not know this.

3. George Sims is quoted from 1907 as he mentions the Polish Jew suspect as a police suspect, and so he should be. But Sims, in the very same article, was totally an advocate -- rightly or wrongly -- for the un-named Druitt as the fiend, and had been in many pieces between 1899 and 1917. Furthermore, Sims mentions an alternate theory to the 'Drowned Doctor' but it is not the Polish Jew -- whom the same writer dismisses because it was too long a gap between the Kelly murder and his incarceration -- but rather that of a 'young, American medical student'.

Could Sims have been wrong?

We know he was because Druitt was not a medical man, for starters. What I am getting at is that Sims is a complicated source who needs to be examined from multiple angles if you are going to do it justice. This does not happen here, and if nothing else it deprives the lay reader of the intrigue and mirage-like quality of the mystery inside the mystery. Instead it becomes Sims, with top police contacts, who wrote that the un-named Kosminski was an important suspect.

But it could also be argued that Sims, as a source -- or Macnaghten pulling the strings behind him -- provides a stronger argument against Kosminski being a major suspect (in Mac's anti-Anderson memoir chapter, 'Laying the Ghost of Jack the Ripper', he judged 'Kosminski', rightly or wongly, to be not worth even mentioning).

Rob House's thesis, seems to me, to be that if you look at the sources objectively, strip away all the encrustations of myth and bias, then Aaron Kosminski is obviously viable and the strongest suspect. Well, that's one way of looking at it but I would have preferred a richer argument/counter-argument, for example, as I perceive it in the Kosminski chapter of Paul Begg's '--The Facts' (2006).

Yes, it's a 'case for' alright; but for me it was too heavily weighted on the side of the 'prosecution' and not enough airtme was given to the 'defense'.

Look, all books are biased. It's not a 'con trick'.

It is a question of how much, and whether this limits a book's scope and depth? Such a judgment must be made by the individual reader.

'Autumn of Terror' (1965) by Tom Cullen is brazenly biased and a leftist polemic (eg. who cares if Druitt wasn't a doctor, as he was still a privileged parasite), and yet it is a brilliant work, and it's prose is marvellous.

Other people regard it as totally outdated in its access to sources, ludicrous in its conclusions, and embarassingly novelistic in style.

Again, it's in the eye of the beholder.

'Case Closed' (1993) by Gerald Posner -- a lawyer -- is a bracing and entertaining demolition job against the Kennedy Assassination conspiracy theorists. It seems to be fair in putting all the arguments, and counter-arguments.

Actually Posner never even mentions the Coleman/Slawson initially classified section of the Warren Report (declassified in 1975) which postulated that 'the Odio Incident' could lead to a conclusion that Oswald was the assassin-dupe of Anti-Castroites.

That sub-report, according to historical methodology, is a very strong and reliable source as it goes against its expected bias; the agenda of the Warren Commission being to back the FBI conclusion that Oswald acted totally alone (eg. no Communist and/or foreign, or domestic conspiracy).

That's a big, big flaw for me.

It does not mean that the book is not still dazzling, and a great read, and many people might think exactly the same way about Rob's booK: biased but brilliant.

Or, they might just think its brilliant, and later feel glad that Rob left out what they also consider to be redunant and silly alternate theories. After all, hen somebody writes about Darwninan evoltuon, do they really have to drag in 'Intelligent Design' dross -- to be 'fair'?

I am very glad that I bought Rob's book and was very absorbed reading it, and coming to my own conclusions about it. I learned a lot about the travails of a minority group of that era, all quite fascinating. I urge other readers to do the same, and to not rely on what myself, or anybody else says the book is supposedly like, but to find out and engage with it for themselves.

[The Grave Maurice]

I can think of absolutely no suspect book published that lays out all the evidence, pro and con, and then asks the reader to make up their own mind.

Of course you can't, Wolf. If you could, they wouldn't be suspect books, would they? Suspect books are, by their very nature, more or less biased. The point is that Rob's book shows much less bias than most and arrives at a reasonable, if not universally acceptable, conclusion.

And, by focussing on a couple of, in your view, shortcomings, you seem to ignore the book's strengths. I can't remember the last time I read a JtR book (suspect oriented or not) that was written so elegantly and free (as far as I can tell) of historical and grammatical errors.

[robhouse]

Hello Wolf,

I admit that it was an error on my part not to include Douglas's later revised opinion on Kozminski. I realize this is no excuse, but an earlier version of the book did include this in an endnote and I removed it... in fact I thought it was still in there. I think I removed it simply for space reasons, as the publisher asked that I cut out 40% of my book's original length.

Fido clearly convinced Douglas that his suspect Cohen/Kaminsky was a better fit for the Ripper. However, I have some doubts whether Fido gave Douglas an impartial and complete account of the situation.

For one thing, Fido was clearly wrong in his assumption that Kozminski was an imbecile. Also, a full 16 years of Kozminski's asylum file is missing, so I don't see how anyone can conclude that "during that time he was (...) not violent and never gave any indication of being the Ripper." In other words, we have an asylum record that covers only 42% of the entire time he was in the two asylums... and most of the surviving records are from the last 10 years of his life which commence in 1910, over 20 years after the murders were committed.

In the more relevant part of Kozminski's surviving record (the 3 years at Colney Hatch) he is described one one occasion as being "at times excited & violent" and on another occasion "at times noisy, excited & incoherent." (Noting the suggestion that he was violent on more than one occasion.) So in my opinion, Mr. Fido's conclusion that Kozminski was "harmless," is not necessarily accurate.

In any case, I focused more on Roy Hazelwood's opinion, especially since he told me one thing that was not revealed in that TV show... that the FBI concluded that the Ripper was likely schizophrenic. I am not clear honestly if Douglas also held this opinion, or if just Mr. Hazelwood did. I did try to get in touch with Douglas to get a comment but he never replied to my emails.

About geographic profiling.

I spoke with D. Kim Rossmo on the phone and he basically agreed that Kozminski fit the geographic profile. In the following article (http://www.txstate.edu/gii/jacktheripper.html) Rossmo states that "The peak area of the geoprofile focused on the locale around Flower and Dean Street and Thrawl Street." He also adds that "While the geographic profile for the Whitechapel murders is interesting and has some supporting evidence, we cannot assess its accuracy."

During the time of the murders Kozminski was probably living either at 34 Yalford Street or 74 Greenfield Street. This is approximately 500 yards from Canter's "peak area." You can walk this distance in about 3 minutes. How is it that you can say: "Neither of these experts came to the conclusion that the Ripper lived anywhere near Greenfield or Yalford Streets." Are you claiming that Greenfield Street is not "anywhere near" Thrawl Street?

This is not an exact science obviously. But I don't think any geographic profiler would tell you that the Ripper lived exactly on the corner of Flower and Dean Street and Thrawl Street. The profile generates a heat map (http://www.wesleyenglish.com/wp-cont...the-Ripper.png) which shows areas in which the killer might have lived, statistically speaking. I mean, how much accuracy are you expecting? 50 feet? 100 feet? It is not that precise.

When I said that "every geographic analysis of the Ripper murders fits with a model that has the Ripper living at or near Kozminski’s presumed residence on either Greenfield Street or Yalford Street," I meant that Kozminski's residence fit with a) geographic profiles of the Ripper b) the idea that the Ripper was returning home after the murder of Eddowes and passed the spot of the graffito, and c) other factors which indicate the general area around St. Mary Matfelon was a sort of jumping off point for the murders, especially the early ones.

Yes, I admit this is only a layman's approach. I am no expert in geographic profiling. But I think that the basic conclusions are quite simple.

I thank you for your input and feedback in any case.

Rob House

[Chris]

Wolf

Perhaps Rob will respond to your comments himself, but I have to say I think your description of what the book says is itself pretty misleading.

For example, you claim that it's not explained that "geographic profilers run complex mathematical software programs in order to make their predictions," but in fact it's made perfectly clear that Canter's circle theory is "one of the simplest applications of geographical profiling" (p. 285) and immediately after it's discussed Rob explicitly says "In reality geographical profiling is more complex than this, and crimes scenes are often distributed in complex spatial patterns" (p. 286).

You imply that Rob's conclusion is based solely on Canter's circle theory, but that's not true. There are three pages of discussion of the weaknesses of the theory and the other factors that may be significant before this conclusion is stated.

And my own comment on those complex computer programs you refer to is this. They rely on essentially heuristic mathematical models constructed by analysing the behaviour of modern serial killers, very few of whom (if any) operated on foot over distances of only hundreds of yards. So because the models are fundamentally nonlinear, they are completely inapplicable to the Whitechapel murders. A simple method like Canter's may give some kind of rough indication, but these complex models - applied way outside their range of validity - really can't tell us anything useful at all.

[Wolf Vanderlinden]

“it reminded me exactly why suspect driven books are mostly a con job.”

I say this because, by their very nature, suspect driven books are a con job. I can think of absolutely no suspect book published that lays out all the evidence, pro and con, and then asks the reader to make up their own mind. Instead, authors of suspect driven books skew the evidence towards the theory that they are flogging while pretending, if we are lucky, to be objective about the subject. The object is never to come to a truth supported by all the evidence but to come to a truth that the author is trying to establish in order to support their theory. That’s not harsh or unthoughtful just the nature of the beast. Rob’s book is no different.

For example:

The chapter on profiling. Rob goes into some detail on exactly what profiling is and what profilers use in order to draw their conclusions. Luckily the FBI actually did a profile on the Ripper, written by Special Agent John Douglas in 1988, for the television show The Secret Identity of Jack the Ripper. Rob reprints the profile then points out how both special agents Douglas and Hazelwood chose Kozminski from the five Ripper candidates offered by the program and shows how the profile fits his suspect. Great support for his suspect Kozminski….except that John Douglas, who wrote the profile, later changed his mind about Kozminski.

In his 2000 book, The Cases That Still Haunt Us, Douglas writes that “there are a couple of problems with Kosminski, information I had not been given at the time.” He points out that information found in Kozminski’s asylum records do not conform to what he would expect would be the behaviour of Jack the Ripper. He states: “During that time he was often dissociative but not violent and never gave any indication of being the Ripper. I would expect a paranoid individual of this nature to talk frequently of this. Kosminski seems too docile and passive to have been a predatory animal nightly on the hunt for victims of opportunity.” Instead Douglas comes to the conclusion that David Cohen’s violent antisocialism and early death are a much better fit for the Whitechapel murderer and concludes his chapter on the Ripper: “Therefore, I am now prepared to say that Jack the Ripper was either the man known to the police as David Cohen…or someone very much like him.”

Nowhere was Douglas’ 2000 published opinion on his profile and Jack the Ripper mentioned in Rob’s book and instead only his opinion from 1988 is used. It seems that Douglas is good enough when he supports Rob’s theory but not good enough when he doesn’t. Why?

The chapter on geographic profiling. We get a brief outline of what, exactly geographic profiling entails: “a technique in which a mathematical model, based on spatial analysis of the location of crime scenes, is used to predict the approximate location of criminal residence.” We are told that this was first proposed by Kim Rossmo. Although not explained in the book, geographic profilers run complex mathematical software programs in order to make their predictions. Rob then uses the most basic tool of the Geo Profiler, “circle theory,” developed by David Canter, in order to state that, surprise, surprise, geographic profiling supports his theory that Kozminski was the Ripper. He writes: “So it seems that every geographic analysis of the Ripper murders fits with a model that has the Ripper living at or near Kozminski’s presumed residence on either Greenfield Street or Yalford Street.” (my emphasis) Really? Every geographic analysis of the Ripper murders?

I mentioned the names of the two top experts in the field – Rossmo and Canter – for a reason. Both these men have done geographic profiles based on the Ripper murders, and both have published their findings. Neither of these experts came to the conclusion that the Ripper lived anywhere near Greenfield or Yalford Streets. This fact was not even hinted at in the book and, instead, we have only Rob’s layman’s conclusion, one which supports his own theory, rather than the work of experts, which don’t. Why?

Of course the simple answer to “why?” in both examples is that evidence or opinions that don’t support the author’s theory are not welcome in the suspect driven book because they tend to muddy the waters too much. Can’t have that.

Wolf.