Radiation hormesis?

I don't know the facts about routine general medical and dental X-ray's.  Yet I still can't help but cringe when I see children in particular with a cellular phone smashed against their skull. Or using notebooks with high power wireless in their laps.   Why not just have them stand in front of a worn out microwave oven? 

I wouldn't worry about it.  Mobile phones emit radio waves, not gamma or x-rays, and large epidemiological studies have shown no link whatsoever between use and cancer statistics.

Come to that, isn't microwave RF, too? Worst case scenario is you overheat a little bit. Kid's to dumb to step away from the thing making his balls warm up, it's probably better he not have use of 'em anyway.

I wouldn't worry about it.  Mobile phones emit radio waves, not gamma or x-rays, and large epidemiological studies have shown no link whatsoever between use and cancer statistics.

Oh, reeeeally.

OK, I'm torqued.  Bernie Cohen did a study comparing home radon levels to lung cancer mortality, a very large study of 1600+ counties across the US.  This is a large epidemiological study.  His data showed a negative correlation -- higher home radon levels corrrelated with lower lung cancer mortality.  To be thorough, he took some 54 different possible confounding factors into consideration, and still found a negative correlation.

His critics objected -- correctly, mind you -- that correlation is not causation, and cited still more factors that might be skewing his data.

Cohen reworked his stuff, this time with maybe 102 factors accounted for, and still found a negative correlation.

His critics objected again and cited still more possibly-skewing factors.

He gave a talk here where I work a few years ago; by that time he & his team had looked at something north of 130 possible confounding factors, and were still finding the correlation negative.

Mind you, it's still correlation and doesn't prove anything.  And yet, isn't it by now damned suggestive correlation?  Like, suggestive that the idea is worth serious research, instead of being (as it is in my field) blithely dismissed because it belies the prevailing theory (linear/no threshold)?

Now, what torques me here is that Cohen's epidemiological study gets wrung through the wringer -- an entirely appropriate wringer -- and yet is still written off, but elsewhere -- like here with your cell phone studies -- epidemiological studies get cited as if they were actual evidence of something, something more than correlation (or the apparent lack of it).  Have your studies been subjected to anywhere near this level of criticism?  Or are they being taken at first blush, as Cohen's would have been had it supported the prevailing theory instead of challenging it?  If the studies show what we want to hear, we are inclined to assume they're good, yes?

"Stoning nonconformists is part of science.  Stoning conformists is also part of science.  Only those theories which can stand up to a merciless barrage of stones deserve consideration."   ---Tero Sand

Until your mobile phone studies get seriously critiqued, I have to say that we don't know jack about cell phone effects, and those who rely on those studies are, in effect, operating on purblind faith.

Source, please.  What you're saying would seem to contradict the paper I linked to before on this data.

Don't know what sam is referencing, but here's what Cohen has to say (emphasis added):

"The principal data that have been cited ... (Ref 11 and Ref 12) to support the LNT are solid tumours (all cancers except leukaemia) among the Japanese A-bomb survivors, and an International Agency for Research on Cancer (IARC) study of occupational doses to radiation workers. The former data (Ref 13) are shown in Figure 1, where the error bars represent 95% confidence limits (two standard deviations). If error bars are ignored, the points do indeed suggest a linear relationship with intercept near zero dose. However, when error bars are considered, they give little statistically significant indication of excess cancers for doses below 25 cSv.

"In fact, it has been shown (Ref 14) that considering the three lowest dose points alone, the slope of the dose-response curve has a 20% probability of being negative (risk decreasing with increasing dose). The analysis utilised by Reference 13 [Pierce D.A. et al, Studies of the mortality of atomic bomb survivors, Report 12, Part 1, Cancer 1950—90, Radiation Research, vol. 146, p1—27, 1996] is very elaborate, involving fitting models with many variables — city (Hiroshima, Nagasaki), sex, age at risk, year, age at exposure, time since exposure, and dose — to determine both background expected deaths and excess deaths for each of the thousands of combinations of these variables. It is very difficult to understand the uncertainties in this analysis, but intuitively it is hard to accept that strong conclusions can be drawn from this data below 20 cSv, involving a total of 6312 observed deaths compared with the 6251 expected background deaths derived from the model."

He also notes (emphasis added again),

"While the solid tumour data on A-bomb survivors and the leukaemia data on monitored radiation workers are said to support the LNT (although the leukaemia data on the former group and the solid tumour data on the latter group do not), there are several studies that seem to contradict the LNT. The data on leukaemia among A-bomb survivors (Ref 13) are shown in Figure 3, with error bars indicating 95% confidence limits. These data strongly suggest a threshold above 20 cSv."

Finally, this is what I'm personally operating from:

1) Earth holds both "primordial" radionuclides (ones formed in stellar fusion & novas &c) and "cosmogenic" ones (being constantly created by incoming protons), and we aren't getting any more primordial ones.  Therefore yesterday was slightly more radioactive than today, and 3.5 billion years ago was exponentially more radioactive than today.  Thus, life began in a significantly higher radiation field than it now finds, and any given species today is adapted to a slightly higher ("yesterday's") background radiation.  As far as I can tell, it's just not possible that living systems can't handle the special stresses of low-level radiation.  I mean, dude!

1a) Oxygen is terribly reactive & corrosive; when it first appeared in quantity on the planet, it was a serious toxin.  Of course the same thing made it quite the energy source.  And what did life do about that?  <inhale>  So, how could it not do something of the sort with something even more ubiquitous?

2) The Navy Nuclear Shipyard Workers' study of ?1960s? 70s? looked at something like 70,000 workers, all doing much the same job; the ones with nuclear clearance had lower cancer rates & lower cancer mortality than the non-nuclear workers.  I kind of think this was never published, but it gets discussed in Health Physics Society ezines and things.

3) Personal conversation:  at a conference, a guy told me of work he & his group did with I think chlorella cells, where they reduced the background radiation to an amazing degree -- separating the K-40 (which is probably the biggest contributor to your personal daily dose) out of the K in the nutrient brine, as well as heavily shielding the lab and so on.  He said the cells they grew were just beautiful -- textbook-perfect, every little organelle just so.  And they didn't do sh*t -- very sluggish to move, very slow to divide and often badly dividing when they did.  And zero resistance to the bacteria they introduced.

4) Same guy as in 3 also described the "mega mouse" experiment, done in Oak Ridge, so named for the million mice involved.  One group was raised in, again, a heavily shielded area with food from which the K-40 had been removed; then a control group, of course; and then 3-4 more groups in successively-higher radiation fields.  According to my informant, the group in 3x the control group's rad field was the healthiest in terms of individual longevity of the mice plus production of offspring; iirc, the longevity of this group was quite striking, like twice the lifespan of the control group but don't quote me on that.

5) Cohen's well-wrung study is not alone.  Elsewhere in the world, two large (35,000+ pop.) regions of China, demographically similar, and the high-radon province has the lower lung cancer mortality; Finland has higher radon than the rest of Europe and also lower lung cancer mortality (either that or t'other way around:  lower-than-Europe radon + higher-than-Europe lung cancer mortality; I forget).

6) Rad workers in the US are limited to 5R (0.05 Sv) annually; "public dose" workers are not to get more than "background" (150-300 milliR).  But there are areas on Earth where the natural background radiation is about 5R (0.05Sv) per year, and these areas are not associated with unusually high cancer incidence or mortality.  It's true that one of these areas has significantly elevated mental retardation, but it's also a desperately poor, famine-ridden area (Kerala in India).

7) 25R (0.25Sv) in one acute dose is the lowest at which nonstochastic effects can be medically detected; 100R (1Sv) in one acute dose will lead to nausea in a few hours.  In the range from 10R to 70R, there is an increased risk of leukemia -- and a decreased risk of breast cancer.  This I got from reading one BEIR report or another, so, seriously mainstream.

Don't have cites for you.  This is what I've learned from 20 years as a health physics (radiation-safety) technician, and I wasn't accumulating it with an eye towards ever convincing someone else one way or another.  Do as you like with it.

However, in the Japanese survivor survey, radiation shows no linear effect on total deaths, with an increase of total deaths setting in at about 1000 mSv, consistent with the conjecture that doses of radiation above 100 mSv result in a small increased risk of being diagnosed with cancer and having one's death blamed on cancer, and a decreased risk of having one's death blamed on something else.

Source, please.  What you're saying would seem to contradict the paper I linked to before on this data.

You are reading the official spin, and ignoring the actual data.  Whenever a scientific paper says something politically incorrect, it begins and ends with a pious affirmation of the official truth in simplistic and colorful language, while explaining the actual truth somewhere in the middle using arcane scientific  language, which progressives are seldom capable of following, and are therefore unlikely to punish the scientist for.  The simplistic language is intended as a signal "I am just  saying this to avoid punishment for heresy and sinful thoughts", while the obscure scientific language is intended as a signal "this is the actual science"

http://www.nap.edu/openbook.php?record_id=11340&page=143

TABLE 6-1 Number of Subjects, Solid Cancer Deaths, and Noncancer Disease Deaths by Radiation Dose

dose	number exposed	solid cancer deaths	non cancer deaths
0 – 5 mSv	37,458	3,833	13,832
5 – 100 mSv	31,650	3,277	11,633
100 – 200 mSv	5,732	668	2,163
200 – 500 mSv	6,332	763	2,423
500 – 1000 mSv	3,299	438	1,161
1000 – 2000 mSv	1,613	274	506
2000 mSv up	488	82	163

Now let us calculate from that total deaths and percentages
               

dose	Number exposed	solid cancer deaths	non cancer deaths	total deaths	percent solid cancer deaths	percent non cancer deaths	percent total deaths
0 – 5 mSv	37,458	3,833	13,832	17,665	10.2	36.9	47.2
5 – 100 mSv	31,650	3,277	11,633	14,910	10.4	36.8	47.1
100 – 200 mSv	5,732	668	2,163	2,831	11.7	37.7	49.4
200 – 500 mSv	6,332	763	2,423	3,186	12.0	38.3	50.3
500 – 1000 mSv	3,299	438	1,161	1,599	13.3	35.2	48.5
1000 – 2000 mSv	1,613	274	506	780	17.0	31.4	48.4
1000 mSv up	2,101	356	669	1,025	16.9	31.8	48.8
2000 mSv up	488	82	163	245	16.8	33.4	50.2

As you can see, there is a threshold effect in the Japanese survivor data for total solid cancer deaths at 100 mSv, and though above 100 mSv, cancers go up, non  cancer deaths go down by about the same amount, indicating that exposure to radioactivity improves ones health in other ways despite increasing one's risk of cancer, or that Japanese people exposed to high levels of radiation are over diagnosed for cancer, or that those not so exposed are under diagnosed for cancer.

The fact that the increase in the one is nearly equal to the decrease in the other suggests that diagnosis is the problem.

We are the survival engine. Of course we can take radiation in some amount, we're constantly bombarded with it, anyway.
Barring radioactive isotopes of key elements, or their ringers, I do think it stands to reason that the body's autorepair can cope with basic radiation quite well.
UV is the bigger cancer killer, right? And we can't live without UV either.

I don't know the facts about routine general medical and dental X-ray's.  Yet I still can't help but cringe when I see children in particular with a cellular phone smashed against their skull. Or using notebooks with high power wireless in their laps.   Why not just have them stand in front of a worn out microwave oven? 

My father was a dentist and when he took an x-ray he used a machine that looked liked something from Dr. Frankenstein' s lab. He manually held the film caddy in place every time he took an x-ray; year after year after year...

I also have fond memories of going to the shoe store and putting my shod feet into the "fluoroscope" and looking at the bones in my feet as I wiggled my toes, again and again and again...

For a fascinating article about fluoroscopes (with pictures!) see:

http://www.orau.org/ptp/collection/shoefittingfluor/shoe.htm

Here's a quick dosage comparison chart I like to spread around, so more than 4 or 5 people have an idea of the amounts were talking about.
http://xkcd.com/radiation/

The cellphone thing really pisses me off.  It fails a basic logic test.  Cells are most vulnerable to radiation when they are dividing.  Brains are the slowest dividing organ in the body, in fact most believed it didn't divide at all until recently.  The organs that divide the most are eyes and gonads (roughly tied).  If cellphones cased cancer it would most likely be eye cancer, or skin cancer in the ear right next to the transmitter (radiation follows an inverse square law).  All the studies are about brain cancer, and all but the latest have been disproven (give it time).

If radio waves cause cancer, I'd be totally screwed, what with standing next to transmitting aircraft antennas to check them on a daily basis.  But since the guy I replaced retired after 30+ years with no issues, I'm not worried.  Aircraft radios put out 5 to 20 watts, most cellphones put out less than .1 watts, avonics guys have been doing this for decades with no noticed trends.

edit: for spell check fail

@mellyrn,

"Torqued" is a new expression to me.  I quite like it.

As for mobile phones: without wishing to get sidetracked too much, it's not just that large epidemiological studies have been conducted.  It's that every single bit of actual evidence I've seen agrees that there's no risk.   All I've seen for the other side is media scaremongering, and some really tiny and terribly designed studies. If you can actually pull up some evidence of risk, that'd be interesting.  But until then... chalk me down as unconvinced.

Now, to the wealth of other stuff you've presented on the main topic of discussion:

OK, fine, it's a bit of a hotch-potch, but enough to pique my interest.  However... let me make my point clear.  I'm not saying OMG low level radiation.  I'm saying levels above 100 mSv are not low-level, and actually dangerous, based on a trawl of Pubmed etc.  Lower than that, and yes, there isn't much evidence for harm.  Although there's also precious little evidence for benefit - I'm far from convinced by such animal studies etc. as have been presented.

And the thing about Cohen's findings on radon and lung cancer is that they haven't been reproducible.  The vast majority of studies on this seem to find the exact opposite - an increase in lung cancer risk, particularly among non-smokers.

@Sam,

Thank you for linking to that data.  I particularly like that you showed your working.  However... I have been looking at the data.  I just disagree with the selectivity you've shown in which data to consider.

For instance, one thing which your table omits is the number of people dying from leukaemia (leukaemia deaths fit into neither the "solid cancer" nor the "non-cancer" categories, and show a much stronger association with increasing dose of radiation than either http://www.nap.edu/openbook.php?record_id=11340&page=144, although in terms of looking at the total deaths, if we actually had a figure for that, it's also worth noting that these would form a pretty small piece of the whole).

We are the survival engine. Of course we can take radiation in some amount, we're constantly bombarded with it, anyway.
Barring radioactive isotopes of key elements, or their ringers, I do think it stands to reason that the body's autorepair can cope with basic radiation quite well.
UV is the bigger cancer killer, right? And we can't live without UV either.

In some amount, of course.  On average, we each get about 3 mSv each year, and some people live fine in areas of the world where the background is rather higher.  But there's no such known mechanism for benefit as there is for UV (which is used for Vitamin D generation, although not essential for it; with the right diet, it seems you can indeed manage fine without sunlight, although personally I wouldn't choose to).

The cellphone thing really pisses me off.  It fails a basic logic test.  Cells are most vulnerable to radiation when they are dividing.  Brains are the slowest dividing organ in the body, in fact most believed it didn't at all until recently.  The organs that divide the most are eyes and gonads (roughly tied).  If cellphones cased cancer it would most likely be eye cancer, or skin cancer in the ear right next to the transmitter (radiation follows an inverse square law).  All the studies are about brain cancer, and all but the latest have been disproven (give it time).

If radio waves case cancer, I'd be totally screwed, what with standing next to transmitting aircraft antennas to check them on a daily basis.  But since the guy I replaced retired after 30+ years with no issues, I'm not worried.  Aircraft radios put out 5 to 20 watts, most cellphones put out less than .1 watts, avonics guys have been doing this for decades with no noticed trends.

In populist ignorance, the word "cancer" stands for any scary illness that could be caused by the environment.   Just understand the word in the sense in which it is used - it is not always necessary to correct peoples scientific understanding.   Cell phone radio does have measurable effects on the brain - just no measurable (real medical) cancer.  The most obvious effect is brain heating.  Every device is required in the US to carry a label detailing exactly how much it heats your brain when held directly to your ear (I always use speaker phone or earphones to avoid worrying about it).  While no ill effects have been definitively found, such heating on a daily basis is worrisome. 

Every more interesting, the 900Hz band has demonstrated non-thermal effects.  Volunteers played a video game while being exposed, or not exposed to 900Hz cell phone radiation at normal cell to ear levels.  Neither participants, nor researchers, knew when the radiation was on (controlled and recorded by computer).  With radiation on, reflexes used in the video game became up to several times *faster* than when it was off.  I.e., the radiation *improved* hand eye coordination.  That's something I would consider using in a soldier situation - but again, I'd be worried about long term effects with constant daily use.

For instance, one thing which your table omits is the number of people dying from leukaemia (leukaemia deaths fit into neither the "solid cancer" nor the "non-cancer" categories, and show a much stronger association with increasing dose of radiation than either http://www.nap.edu/openbook.php?record_id=11340&page=144,

Your source draws a PC conclusion from alleged data, without, however, showing us the data which supposedly supports the PC conclusion, which is pretty much a guarantee that they are stretching the truth or lying barefaced.  If the data actually supported the PC, they would show it to you, indeed it would be up on posters.

Your link tells us:

Leukemia risks increased with dose up to about 3 Sv, with evidence of upward curvature; that is, a linear-quadratic function fitted the data significantly better than a linear function. With this linear-quadratic function, the excess risk per unit of dose at 1 Sv was about three times that at 0.1 Sv.

In other words the slope of the alleged curve was three times steeper at 1000 mSv than at 100 mSv.  This sounds very much like saying leukemia rates were dominated by a threshold at 1000 mSv.

If there is a threshold at 1000 mSv, then, given diagnostic and dosage uncertainties, hard to detect the supposed effect at 100 mSv.  It is always hard to detect an alleged weak effect in the presence of a strong effect.  We are told that the pattern was "complex" but are not shown the complexity.  "Complex" sounds like torturing the data for politically correct results.  

Take a look at the next page, where it shows solid cancer risk.  Looks linear, does it not?  Looks like solid cancer risk is linear in radiation dosage.  That is because they have played clever with the scaling and the size of the dots, so as to obscure the fact that their line is indistinguishable from flat below 100 mSv.

And if they had graphed total deaths, it would be apparent that there was something odd about their seemingly linear hazard from 100 mSv to 1000 mSv

I don't know the facts about routine general medical and dental X-ray's.  Yet I still can't help but cringe when I see children in particular with a cellular phone smashed against their skull. Or using notebooks with high power wireless in their laps.   Why not just have them stand in front of a worn out microwave oven?  

I wouldn't worry about it.  Mobile phones emit radio waves, not gamma or x-rays, and large epidemiological studies have shown no link whatsoever between use and cancer statistics.

Come to that, isn't microwave RF, too? Worst case scenario is you overheat a little bit. Kid's to dumb to step away from the thing making his balls warm up, it's probably better he not have use of 'em anyway.

Don't discount RF in that end of the scale so fast, especially when your looking at 802.11a/h/j/n all using 5ghz.   First Harmonic is at 10ghz at 1/2 original radiated power, and another second is 15ghz at 1/2 the power of 1st.  

My point, powerful micro wavelength energetic pulses jarring chromosomes back and forth really fast.   Personally I want my great-great-grand children to not consider extra appendages as "Features."

And not so much fear of Cancer as explained above, changes to prodigy or lack their of.    Microwaves have verified effect as far as I have seen.  Former roommate and friend considered suing a branch of the military for 'Low Sperm Motility' from working in the radar spaces.  He gathered signatures from other retirees with similar issue to support his case.  His wife was not exactly pleased with it's effect.

BTW, sorry to sidetrack this from pure radiation..  
Back to what happens when we have a bad day like Harry K. Daghlian, Jr.

it's not just that large epidemiological studies have been conducted.  It's that every single bit of actual evidence I've seen agrees that there's no risk.   All I've seen for the other side is media scaremongering, and some really tiny and terribly designed studies. If you can actually pull up some evidence of risk, that'd be interesting.  But until then... chalk me down as unconvinced.

Chalk in hand -- now, unconvinced of what?  Did you read what I said as claiming there is risk from cell phone usage?  Could you please quote me where I said or implied that, so I can fix it?

You write that "every bit of actual evidence" says there's "no" risk.  Great!  But you'd only mentioned "large epidemiological studies" before.  They, like Cohen's review, are subject to what I understand is called the "ecological fallacy" (at least, that's the term my old boss used) -- meaning, if you have not taken all possible confounding factors into account, your study shows nothing meaningful, like ye olde murder rate rising together with banana imports.  A mere lack of correlation on the first pass does not count as "actual evidence" of no risk; it's only evidence of "this is interesting; now, could [this result] be being caused by something else (as the murder rate and the banana import rate were both driven by season)?"  I'm asking if those studies have been subjected to serious rigor.  I don't give a flaming donkey about evidence "for the other side" at this point; I'm wanting to know if the "no damage" studies have been well conducted, and properly scrutinized, or have they been accepted at first blush just because it's what we'd like to see?

A guy who can't kick his own theory in the teeth to see if it can take it is not a scientist worthy of his lab coat.

And the thing about Cohen's findings on radon and lung cancer is that they haven't been reproducible.

What, you mean that counties with higher average radon levels do have higher lung cancer mortality rates?  How on Earth did the man ever find that they were lower, then?  The only complaint I've heard here in the industry about his work was the bit noted before, as to whether he considered all possible factors that might be causing the inverse relationship; they've said, "Maybe the low-lung-cancer-mortality counties have fewer smokers, or better health care, or more young people, and it's not radon driving it."  I've never before heard anyone dispute the observed relationship -- only whether radon was the driving factor or not.

So, what, specifically what, was "not reproducible"?

On average, we each get about 3 mSv each year, and some people live fine in areas of the world where the background is rather higher.

I guess "higher by a factor of 17"  (3mSv vs 50 mSv) counts as "rather" higher.

But there's no such known mechanism for benefit as there is for UV

Well, gosh golly gee, how about we go look for one?  There is logical reason to assume life can deal with "rather" more than "background", as well as tangible evidence.  There is precedent -- 3.5 billion-year-old precedent -- for life taking a serious toxin and turning it into a vital necessity, a source of energy -- and the ambient radiation field was around looong before life discovered oxygen.  "We don't (currently) know how it could be helpful, so it isn't" isn't much of an argument, is it?

There are areas where the natural radiation background is quite high

http://www.probeinternational.org/Ramsar.pdf

People have been living for generations in areas where they get about 200 mSv per year.

200 mSv per year is about what you would get if you lived right next door to the Japanese Fukushima nuclear "disaster", about what you currently get at the gate to the Fukushima nuclear reactor.

This is about one third of the level you get standing right beside the Chernobyl reactor today and about equal to the level you get pretty close to the Chernobyl reactor.  This is about what you would get if we stored high level nuclear waste for sixty years or so to cool down, and then just dumped it on a tip like any other trash, and you hung around on the tip.

This means that over 20 years, they get 4000 mSv, which dose will pretty reliably kill you if you get it all at once, and in five years they get 1000 mSv, a dose that no one doubts will likely cause serious risk of cancer and various ailments if you get it all at once.

There are no obvious differences between their health, and the health of people in nearby low radiation areas.  That they are not dying like flies indicates a substantial threshold effect.

For instance, one thing which your table omits is the number of people dying from leukaemia (leukaemia deaths fit into neither the "solid cancer" nor the "non-cancer" categories, and show a much stronger association with increasing dose of radiation than either http://www.nap.edu/openbook.php?record_id=11340&page=144,

Your source draws a PC conclusion from alleged data, without, however, showing us the data which supposedly supports the PC conclusion, which is pretty much a guarantee that they are stretching the truth or lying barefaced.  If the data actually supported the PC, they would show it to you, indeed it would be up on posters.

My source is actually your source, two pages further on.  (I'd gone looking for their leukaemia data when it wasn't in the table, while also wanting to check where the data in that table came from and what other evidence they were looking at and analyses they had run).

Your link tells us:

Leukemia risks increased with dose up to about 3 Sv, with evidence of upward curvature; that is, a linear-quadratic function fitted the data significantly better than a linear function. With this linear-quadratic function, the excess risk per unit of dose at 1 Sv was about three times that at 0.1 Sv.

In other words the slope of the alleged curve was three times steeper at 1000 mSv than at 100 mSv.  This sounds very much like saying leukemia rates were dominated by a threshold at 1000 mSv.

If there is a threshold at 1000 mSv, then, given diagnostic and dosage uncertainties, hard to detect the supposed effect at 100 mSv.  It is always hard to detect an alleged weak effect in the presence of a strong effect.  We are told that the pattern was "complex" but are not shown the complexity.  "Complex" sounds like torturing the data for politically correct results.  

Take a look at the next page, where it shows solid cancer risk.  Looks linear, does it not?  Looks like solid cancer risk is linear in radiation dosage.  That is because they have played clever with the scaling and the size of the dots, so as to obscure the fact that their line is indistinguishable from flat below 100 mSv.

And if they had graphed total deaths, it would be apparent that there was something odd about their seemingly linear hazard from 100 mSv to 1000 mSv

You seem very confident about what data you haven't seen would show.  And I'm really not seeing the problem with the graph.  What alternative scaling are you saying they should have used?

There is, however, one key problem on the leukaemia issue which I've uncovered.  I just went looking for the mechanism behind the high degree of vulnerability to leukaemia observed... and it seems that what happens is that various alpha-emitting radioactive elements go straight to the bone (particularly polonium-210, or rather its precursor, lead-210, and radon-222).  (A similar issue to the increased thyroid cancer, which stems from radioactive iodine being taken up along with all the other iodine by the thyroid gland).  So the increased vulnerability of people exposed to nuclear fall-out to leukaemia may not actually be of massive relevance to, say, gamma-exposure.  http://books.google.co.uk/books?id=DO4_ur2tOiAC&pg=PA2&lpg=PA2&dq=table+sieverts+leukaemia&source=bl&ots=DiHbSnWoye&sig=TUJ-lSrS0JK7pqedmR4Lah24HAg&hl=en&sa=X&ei=8Oz3T_KbJ-Ox0QXE0cihBw&ved=0CFoQ6AEwBA#v=onepage&q=table%20sieverts%20leukaemia&f=false

Chalk in hand -- now, unconvinced of what?  Did you read what I said as claiming there is risk from cell phone usage?  Could you please quote me where I said or implied that, so I can fix it?

You write that "every bit of actual evidence" says there's "no" risk.  Great!  But you'd only mentioned "large epidemiological studies" before.  They, like Cohen's review, are subject to what I understand is called the "ecological fallacy" (at least, that's the term my old boss used) -- meaning, if you have not taken all possible confounding factors into account, your study shows nothing meaningful, like ye olde murder rate rising together with banana imports.  A mere lack of correlation on the first pass does not count as "actual evidence" of no risk; it's only evidence of "this is interesting; now, could [this result] be being caused by something else (as the murder rate and the banana import rate were both driven by season)?"  I'm asking if those studies have been subjected to serious rigor.  I don't give a flaming donkey about evidence "for the other side" at this point; I'm wanting to know if the "no damage" studies have been well conducted, and properly scrutinized, or have they been accepted at first blush just because it's what we'd like to see?

To be fair, I haven't been looking into this in as much detail as the main topic, and it's sufficiently far outside my main area of research (when I worked in research) that I can't really claim to be an expert anyway.  So I can't really answer your question on the extent of rigour to which each and every study has been put through; sorry.  All I've really done is sum up what I remembered from looking into this once before, a few years ago.

A guy who can't kick his own theory in the teeth to see if it can take it is not a scientist worthy of his lab coat.

I couldn't agree more (I'd be tempted to accuse you of teaching your grandmother to suck eggs here, but that would risk turning this into a contest of academic dick-waving).

And the thing about Cohen's findings on radon and lung cancer is that they haven't been reproducible.

What, you mean that counties with higher average radon levels do have higher lung cancer mortality rates?  How on Earth did the man ever find that they were lower, then?  The only complaint I've heard here in the industry about his work was the bit noted before, as to whether he considered all possible factors that might be causing the inverse relationship; they've said, "Maybe the low-lung-cancer-mortality counties have fewer smokers, or better health care, or more young people, and it's not radon driving it."  I've never before heard anyone dispute the observed relationship -- only whether radon was the driving factor or not.

So, what, specifically what, was "not reproducible"?

There have been plenty of studies on radon showing a link with lung cancer, both occupational (e.g. in miners) and residential (living in high-radon areas).  But really, what you should do is look at those studies by Cohen in their historical context, as summarised in this article: http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3073196/?tool=pubmed

On average, we each get about 3 mSv each year, and some people live fine in areas of the world where the background is rather higher.

I guess "higher by a factor of 17"  (3mSv vs 50 mSv) counts as "rather" higher.

Indeed, but it's still below the range we're arguing over (100 to 1000 mSv).  Also, 50 mSv or above is, so far as I'm aware, confined to a few very small areas within the coastline of Kerala, some areas within a single county in Iran (Ramsar) and a very few other places.

But there's no such known mechanism for benefit as there is for UV

Well, gosh golly gee, how about we go look for one?  There is logical reason to assume life can deal with "rather" more than "background", as well as tangible evidence.  There is precedent -- 3.5 billion-year-old precedent -- for life taking a serious toxin and turning it into a vital necessity, a source of energy -- and the ambient radiation field was around looong before life discovered oxygen.  "We don't (currently) know how it could be helpful, so it isn't" isn't much of an argument, is it?

Be my guest.  But the burden of proof lies on those claiming this effect; no benefit is the null hypothesis.  And biological plausibility in the form of a mechanistic explanation is among the Bradford-Hill criteria for a reason.