Skip to content
Thoughtful, detailed coverage of everything Apple for 29 years
and the TidBITS Content Network for Apple professionals

Photo by John Price on Unsplash


Worried about 5G and Cancer? Here’s Why Wireless Networks Pose No Known Health Risk

Can cell phones or Wi-Fi give you cancer? The answer is reasonably definitive: No. That’s equally true for new 5G cellular networks currently being rolled out worldwide, all previous cellular networks, and all versions of Wi-Fi.

Decades of research make it clear that the use of wireless networking, whether through mobile phones or Wi-Fi devices, carries with it no elevated risk for cancers or other diseases in humans. Some studies with rodents have garnered attention, but in a recent much-cited one that delivered a mix of results, some rats lived longer under high exposure to cellular signals than their control group counterparts.

It can be hard to accept that there’s no conspiracy here. Corporations and governments routinely engage in misdeeds, and that has instilled mistrust in many of us. It’s easy to become cynical and worn down, and to lump all concerns into the same basket when giant pharmaceutical companies sell drugs like Vioxx and OxyContin that they know have significant side effects (including premature death) as prescribed; when firms dump poison into communities, deny it, and then walk away paying nothing; and when city governments knowingly allow unsafe levels of lead in municipal drinking water.

People also tend to try to find a pattern of conspiracy where none may exist. It makes for good headlines. The Bloomberg News story in October 2018 about Chinese intelligence agencies installing spy chips on motherboards used in data centers by Apple, Amazon, and others produced a lot of noise but was full of inconsistencies and was categorically denied by all parties. Over a year later, nothing in the story has been expanded on by Bloomberg or confirmed by security experts, other reporters, or, well, anyone.

More recently, the Chicago Tribune published the results of testing from a firm it had hired to check if emissions from modern smartphones truly fell below FCC safety limits. In those tests, many appeared to exceed regulatory limits. The Tribune didn’t overstate its results, but the bottom line more or less suggested that smartphone makers were deceiving the FCC and the general public. This plays into our fears, even though the work was presented rigorously. (Smartphone makers dispute the methodology of the testing; the Tribune stands by its research. Regardless, there’s a big difference between detecting higher-than-approved emission levels and proving a link between those levels and cancer.)

But the potential of an increased risk of cancer and other diseases from routine exposure to electromagnetic fields (EMFs) used for cellular and Wi-Fi networks falls into a different category than other risks. The principles behind EMFs are founded in well-understood physics. Safety issues surrounding EMFs, particularly the ranges used in cellular networks and Wi-Fi, have been studied since the invention of radar in the 1940s—and science learned powerful lessons from early overexposures that led to injury and sometimes permanent disability or death among soldiers and technicians.

Fast forward to the present, and many people have expressed fear that the deployment of new 5G networking hardware introduces a new kind of risk. To achieve the promised high rates of speed and serve new categories of devices, 5G networks will draw from a much broader range of frequencies, some far higher (or shorter) wavelengths than current technologies. And many more base stations will need to be deployed.

But the newness and differentness of 5G don’t matter. Whether we’re talking about 5G, 4G, 3G, Wi-Fi, or other consumer-level wireless technologies, the sum total of results from many studies and many years of research paints a straightforward picture—there’s nothing to worry about.

Let me explain why wireless technology and associated industries aren’t like drug and tobacco companies, break down why you have nothing to fear from EMFs, why the Tribune may have gotten it right without their results being anything to fear, and how to evaluate the evidence.

A Global Industry Has Many Competitors, and Many Researchers

Many companies and industries that have perpetuated fraud and other crimes against the populace can typically do so because they have amassed enough power to control information about what they produce now or once made, all while keeping government oversight and checks on that power at bay.

The cellular industry certainly has influence around the globe, especially in the United States. And companies that sell Wi-Fi and Bluetooth gear ship billions of dollars of products worldwide.

Yet both groups—which have some overlap in chipmaking and mobile phone handsets—are diffuse. No single company controls cellular or Wi-Fi technology. It’s a broad mix of patent holders, chipmakers, hardware manufacturers, and—with cellular—network carriers. Thousands of companies compete for market share across all the different technologies. Plus, the network protocols in play are designed and promulgated by non-profit groups made up of representatives from numerous companies.

Further, the actual hardware in use is readily available, and the specifications governing that hardware and associated protocols are all public. The industry relies on widespread and well-understood technology.

Because of that, there’s another facet to consider: the networking world has no central point of control, whether a company or cabal, and no restriction on testing and publishing results, as the Chicago Tribune report made clear. Thousands of research studies of all sorts have taken place worldwide. Academic institutions, private organizations, government agencies, and media outlets have all carried out tests that range from simple hardware measurements to decade-long epidemiological projects that involve hundreds of thousands of people.

Could research be tainted? Certainly—in the US, there’s often a fear that universities don’t wish to risk corporate funding by publishing negative research, and many people have entirely valid concerns that state and federal agencies are captive to industries they regulate. However, in other countries, particularly in the European Union, governments and nonprofits fund research and either limit corporate involvement or ban it altogether, and are willing to act forcefully against firms, even if there’s an economic cost. In reality, industry groups have sponsored only a relatively small fraction of wireless research, and even then, some of it generated inconclusive outcomes or provided unsatisfying or unfavorable results to their business interests.

I would be overstating the case to say these facts about wireless research guarantee that no coordinated conspiracy exists to suppress information about widespread health risks. But it’s highly unlikely—there’s too much transparency.

Also, wireless communications technology is now so ubiquitous that there’s a vast amount of exposure to examine and plot data against previous generations when evaluating cancer and other kinds of ailments. The past is a control group.

Even if there were a conspiracy, if the health risks over EMF exposure were elevated, researchers would see the results across huge swaths of the world, which hasn’t happened.

If that’s so, why do some people point to EMFs as the problem? It has to do with history.

EMFs Are an Easy Target for Ill-Defined Conditions

Every few years, as people attempt to find a cause for random disease clusters and inexplicable health problems, EMFs are held up as a possible culprit. It’s part of being human that we look for patterns, and often create them when none exist. (Look at both extremes of contemporary politics everywhere in the world!)

In 1979, a study connected increased incidence of childhood leukemia to living near high-voltage power transmission lines. This single study—call it Patient Zero—is likely the origin of the fear of EMFs causing cancer. However, dozens of studies since have been unable to find strong correlations, although some show that kids living extremely close to high-voltage power lines might have slightly increased odds of developing leukemia.

But there are three substantial missing pieces in the discussion:

  • First, very few children—defined for this disease as newborns to 20-year-olds—contract leukemia. The rate in the US is 4.7 per 100,000 per year, or fewer than 4000 new cases each year.
  • Second, of those kids, only a very small percentage live close to high-voltage power lines. Thus the number of possible additional kids contracting leukemia is no more than a handful each year, making it extremely difficult to associate the power lines with the disease directly.
  • Third, scientists have discovered no biological mechanism that might explain an increased risk.

As a result, even those researchers who have found a correlation between high-voltage power lines and childhood leukemia are dubious that what they’re measuring has to do with power lines at all. In 2005, one researcher suggested that living near power lines is also correlated to something else that really does increase a child’s leukemia risk.

A 2018 paper in Nature reviewed dozens of studies over decades and determined that kids who lived within 50 meters (165 feet) of a 200 kilovolt or higher power line had a slightly elevated risk of contracting leukemia. But the researchers concluded that the intensity of magnetic fields couldn’t be the culprit—the intensity of the magnetic fields wasn’t high enough to explain the findings. (That’s partly because power line magnetic fields max out at about 2.5 microteslas when you’re directly underneath, whereas the earth’s magnetic field, to which we’re all exposed all the time, varies from 25 to 65 microteslas, 10 to 26 times higher.) “Reasons for the increased risk, found in this and many other studies, remain to be elucidated,” wrote the researchers.

Some researchers have speculated that “something else” is the kind of hygiene, nutrition, general quality of life, and chemical exposure conditions that exist in communities through which high-voltage power lines are allowed to cross. Wealthier people either don’t live under them or prevent them from being built nearby, long before any health risk was a concern.

If you have a population of poorer, sicker people with worse access to healthcare and food, and who may have a higher concentration of industrial contaminants in their neighborhoods, that might be a trigger for the conditions in the body that result in childhood leukemia. It remains speculation. (Frankly, this reminds me of the origin of the field of epidemiology: Londoners had many explanations for outbreaks of cholera, and it took the rigor of one person to convince others that well water was the cause.)

It may also be that the data about children’s ages, where they lived, and intensity of fields near the powerlines isn’t sufficiently accurate, and the elevated risk is entirely statistical noise. The field data used for the studies isn’t generally directly measured from any of the locations analyzed. Rather, it’s calculated from maps of powerlines, information provided by utilities or from public sources, and geographical information systems that map properties for municipalities. Given the distances and power levels involved, some of these numbers aren’t much better than guesses.

Similarly, some people have self-diagnosed a condition labeled electrosensitivity. Those who suffer from this putative disease say they experience fatigue, mental fog, and other debilitating symptoms when in the presence of EMFs, even at a great distance. They sometimes label the miasma of EMFs that they believe afflicts them as electrosmog.

Teasing out the cause of electrosensitivity is tricky because the environment around us is suffused with EMFs and signals. However, because of the “inverse-square law,” most of that is at levels of energy that can barely be distinguished from noise by the miracle of modern data-receiving radio equipment in Wi-Fi adapters and phones. The inverse-square law states that a signal decreases by the square of the distance—stand 1 foot away from a transmitter and get a power measurement of 100, whereas 2 feet away it will be 25, and so on. Even with a cell phone next to your head, the design points the signal away from you, and your skin and bone absorb quite a bit of it before it reaches the sensitive soft tissues in which cancers are most likely to form.

The New York Times noted in July 2019 that a widely circulated chart that purported to show the absorption of signals by brain tissue was never evaluated by anyone but its creator, a physicist without biological expertise. One prominent expert witness who relied on the chart in testimony for years admitted that he had no idea that skin blocked the passage of high-frequency signals in particular. “…maybe it’s not that big a deal,” he said.

Scientists have been unable to find any electrosensitive condition, which makes sense given it has no basis in which it could exist. In study after study—many double-blind and peer-reviewed—in shielded rooms to block outside EMFs and using EMF generators with known properties, people who believe they have this ailment are able to detect the presence or absence of signals at rates no better than chance and no better than control subjects.

But here’s the thing, and it’s crucial: Researchers who hooked people up to measure their physical response during testing found that they did experience a variety of ailments. It’s just that those ailments weren’t correlated with whether signals were present or not. These people have real problems, but these tests prove that EMFs are not the cause.

Oddly, as with powerlines and childhood leukemia, those who claim to suffer from electrosensitivity seem less concerned about other, more likely culprits for unusual health problems: massive pesticide exposures in food and the environment, prescription drug outflow into water supplies that wind up back in our bodies, and the dangers of microplastic exposure, the impact of which we’re learning more every day, to name just a handful of global health concerns.

Nonetheless, the obsession with EMFs has persisted.

Wi-Fi and Cell Signals: Same Song, New Refrain

In the 1990s, when I helped out on a book about ways to improve your health in an office, including better monitor position, ergonomics, and exercise, the author and I looked into research that alleged increased health risks from the extremely low frequency (ELF) and very low frequency (VLF) signals emitted by cathode-ray tube monitors.

Not surprisingly, the studies were poorly done and not reproducible, and as the population of users increased, the predicted increase in health problems didn’t happen. (One of the people pushing ELF/VLF health fears had spent many years doing the same thing with power lines.)

A decade later, while I was writing a daily blog, Wi-Fi Networking News, concerns about the effects of cellular networks and Wi-Fi base stations began to rise. These debates felt like the same lyrics with a new melody, but the number of people who would be exposed meant it was essential to study the issue with care instead of dismissing it. I took that tack, even though the physics of EMFs put out by Wi-Fi and cellular signals and known biological mechanisms meant a connection was implausible.

Once again, early studies on cellular use showed the potential of causation. Many used lab animals exposed to levels of EMF far above what people would ever experience and for long periods of time every day. Others used retrospective analysis, asking people to fill out surveys about past use, and then associated their answers with a corresponding health survey to look for an increase in rare cancers.

Just as with the power line concerns, as the years progressed and more studies were conducted with greater rigor, the potential effects disappeared. At the same time, with a large population of people using cellular phones and in regular proximity to Wi-Fi networks, specific health effects should have bubbled up in the population, showing increases especially with cancers and diseases that were otherwise uncommon. No such widespread health concerns appeared.

The fact is that Wi-Fi and cellular networks use exceedingly small amounts of power, far below the thresholds known to cause risk by exposure. Back in the 1940s and 1950s, when powerful microwave use in military radar and other purposes was just getting started, research helped understand the effects microwaves have on health, and limits were set with significant margins.

In particular, Wi-Fi and cellular networks, including 5G networks, use relatively high frequencies, which have short wavelengths. They don’t travel far and, the higher the frequency, the shorter the distance they can travel using the same power as lower frequencies. By deploying 5G densely, less power is needed, and by using high frequencies, it can’t penetrate far—whether through walls or into our bodies. Even though many more base stations will be deployed, they’ll be sending out far less power than today’s networking systems. (In comparison, the microwaves in early radar used longer wavelengths and lots of power, rendering them much more dangerous.)

Lies, Damned Lies, and Statistics

There’s no question that you can find seemingly compelling research results online that link aspects of some EMF studies with a potential increase in risk for certain cancers. Many of these results fall afoul of a modern scientific research problem that’s currently causing a massive upheaval across all fields in which data are gathered.

The problem is generally called “p-hacking,” where p refers to the standard measurement of probability. To say that research results are “statistically significant,” researchers need to find a probability of p < 0.05, which means that there is a less than 5% probability that the results happened by chance. No matter how slight the result, if p < 0.05, it can be cited as statistically significant.

Because of the pressure academics feel to publish and obtain grants, some scientists have unfortunately developed a habit of p-hacking for results by analyzing research data in ways other than those stated in the study’s hypothesis. This practice, though common in the business world, is increasingly considered illegitimate in science. A celebrity Cornell University professor’s lab engaged in this practice, according to Buzzfeed investigations, and he recently resigned after a number of his group’s seminal papers were retracted or corrected.

The emerging gold standard for research will likely involve stating the hypothesis to be tested at the outset and placing it in escrow—it would be like a fortune teller placing their predictions in an envelope held by a member of the audience—plus retaining raw data in a standardized form that others can analyze later. (These changes are all aimed at addressing the “reproducibility crisis,” the inability of researchers to replicate the results of earlier, sometimes seminal experiments.)

I read many studies in the 2000s in which the research appeared to look for one result, but failing to obtain it, sliced and diced the data to find something—anything!—that showed p < 0.05. These conclusions were often baffling, as they would demonstrate a noticeable risk for a sliver of, say, rats or people by age, gender, and exposure, but sometimes decreased risk for other, very similar groups. That’s the problem with p-hacking—you might stumble across something that’s p < 0.05, but simultaneously makes no sense, fits no models, and can’t be reproduced independently.

A good example is the 2010 Interphone study, which was initially heralded as a landmark set of studies conducted across countries. But the p-hacking of its results made it ripe for citing by skeptics and of dubious quality for those who work with statistics. One of the conclusions drawn from the study was that using a cell phone a lot apparently reduced the cancer rate in an odd demographic group of people, a result for which there’s no sensible biological explanation. The most likely explanation stems from design flaws in the study.

More recently, large-scale, long-term rat studies found nothing conclusive. The New York Times wrote a few times in 2018 about the rodent study I mentioned earlier. The study clearly demonstrated the risk of high levels of exposure that far exceed US safety limits. The catch is that the risk was evident only when the EMFs were applied to the full bodies of rats, not just the areas of our bodies that are exposed to EMFs. Oh, and male rats had a reduced risk of cancer. Moreover, the kinds of cancers these poor rats experienced haven’t increased in frequency in the human population, nor have they been found in studies that characterize cell phone users by heaviness of use.

Otis Brawley, chief medical officer of the American Cancer Society, told USA Today in regard to the new research, “The incidence of brain tumors in human beings has been flat for the last 40 years.” His organization has written up in a highly readable and sensible way the current understanding of the risks, complete with short analyses of key studies.

With no evidence of sensitivity, no increase in cancers, irreproducible studies that have led to dead ends, and no epidemic of conditions among the large, long-term population of cell-phone and Wi-Fi users, we’re led to one conclusion: There is no health risk associated with everyday exposure to common EMFs. That’s the case even if, intuitively, it feels like there must be a link. We may never get over that feeling, but we should base our behavior and policy on solid science, not feelings.

The good news is that, even if you’re still concerned, the inverse-square rule is your friend, as long as you accept physics. For those who worry about EMF exposure and can’t shake it, anxiety is obviously a legitimate condition, and one that you can mitigate. The American Cancer Society recommends if you remain concerned about EMFs and spend a lot of time talking on a cell phone, for your own peace of mind, use earbuds or keep the phone further away from your head.

And maybe don’t put your Wi-Fi router under your pillow.

Subscribe today so you don’t miss any TidBITS articles!

Every week you’ll get tech tips, in-depth reviews, and insightful news analysis for discerning Apple users. For 29 years, we’ve published professional, member-supported tech journalism that makes you smarter.

Registration confirmation will be emailed to you.

Comments About Worried about 5G and Cancer? Here’s Why Wireless Networks Pose No Known Health Risk

Notable Replies

  1. I remember when the first UFO Airport base station was just out I got a chance to talk to the lead engineer who worked on it. This was at Apple Netherlands HQ in the town of Bunnik. HQ has since moved to Amsterdam BTW. He told me that he experienced a very noticeable effect of getting dry eyes and having to blink more than usual. This was in very close proximity to the active device. It seemed the signal helped ablate the tear film on his eye balls. There was no other biological effect that he could find.

    (That’s partly because power line magnetic fields max out at about 2.5 microteslas when you’re directly underneath, whereas the earth’s magnetic field, to which we’re all exposed all the time, varies from 25 to 65 microteslas, 10 to 26 times higher.)

    I don’t think you can compare those two. The Earth’s magnetic field is stationary and is just that, a plain magnetic field. The fields around HT power lines are alternating at 50 or 60 Hz and therefore radiate an EM signal.

    Come to think of it, people having trouble with power lines may very well be on to something. Langoliers tend to have a penchant for power lines as is well documented in this report. Then again, they are known to eventually eat everything, so their preference for power lines may not rise above the signal noise in the end.

  2. He was enamored with it, he forgot to blink.

    Nope. He observed the effect clearly during the entire project, comparing it with his time away from his workbench.
    He did not see cause to initiate follow-up though, he also didn’t compare with colleagues as far as he recounted. To him it was just an interesting side-effect, so it remained just that. That is not reason for dismissing it out of hand of course. That would require looking into it by means of a separate research project.

  3. 5G “information” litters the Internet: propaganda from telecoms, concerns of scientists, physicians, and parents, crazy stuff from paranoid people, and fake and distorted news from RU to stir up the controversy. So thanks Glenn, for addressing the topic.

    You refer a couple times to the New York Times articles about 5G - my beef with the Times - they have a multi-million dollar 5G joint venture with Verizon and print pro-5G “science” articles, rarely (if ever) mentioning their conflict of interest.

    So forget the studies for a moment. I’m worried about 5G in my neighborhood. The FCC safety standards have not been updated in decades, meaning they are based on old studies of old technology. Still, the FCC dictates that communities accept 5G installations under existing “right of ways” with little or no say in where new towers may be placed and with no liability on the part of the telecoms for even potential harm. And proposed deployment density of 5G towers (new, or added to existing infrastructure) in a so-called “smart city”, would mean a 5G box on a pole for every 5 or 10 homes on a block.

    5G is a boondoggle for telecoms. (I don’t believe it is needed, but nevermind that.) It’s an opportunity to sell lots of new hardware, cut the cable companies out of the business, and obtain access to massive amounts of new user data (think “smart homes”). Oh, and remember, 5G isn’t replacing existing infrastructure any time soon, it is going to supplement existing cellular and WiFi architectures. Old towers, new towers, you get both.

    Back to science and safety. There’s a lot of resistance and caution about possible harm (cancer or other) from 5G deployment, especially in Europe and California. Why, when there really aren’t very many studies? Because there aren’t enough studies! I agree with a more conservative analysis, such as the one expressed in this article from Vox (below). Which essentially says, there is not enough evidence one way or the other, so go slow.

    “A comprehensive guide to the messy, frustrating science of cellphones and health”

  4. I certainly hope everyone would agree that more research is always welcome, but I’m curious where in the Vox article you’re seeing the underlying details about how 5G is sufficiently different from other EMFs to cause particular concern? The frequencies are higher (with shorter wavelengths) and the power levels are lower, both of which should reduce penetration compared to current networking gear.

  5. Increased density is the “solution” for reduced penetration. The saturation of persistent signal (plus different frequencies, etc.) is different - there is no history, and hence there are no studies, of “prolonged or long term exposure” to these EMFs. Dr. Moskowitz, referenced in the Vox article, writes about that here:

    Definitive proof of cancer seems to be unlikely, but what about the possible “other biological effects?” Acoustic neuromas, reduced sperm motility? Keep the “5G small cells” away from my house please.

  6. Electromagnetic radiation is radiation. The amount broadcast by humans for TV, radio, cell phones, etc pales in comparison to the amount one gets from a tooth X-ray but they are all the same thing. Certainly there is going to be less trouble with less radiation. We also know there is no minimum dose that will not cause trouble.

    The other thing about 5G which is mostly about switching from copper to fiber is copper wire coming into the home carries the voice or data as well as 48VDC of electrical power. The Boston Globe: ‘Aug 31, 2014 - When there’s a power outage, a phone connected to a copper wire line will still work, while the technically superior fiber optic lines will work only as long as a battery back-up does …’ another way to put this is ‘Fiber to the home service works differently than traditional copper pair telephone service. With traditional telephone service very small amounts of electrical current are carried over a copper circuit to power the telephone in your home. Since the new signal is entirely optical and can carry no appreciable power, the electronics attached to your home must be locally powered.’ I have an old rotary dial phone which is my ultimate backup phone. It gets it power from the copper wire coming in from the pole on the street.

    Just because it needs to be pointed out first there is no question cable companies have a very bad service record and an aggressive approach. This is also becoming more evident among telephone companies. ‘The Verizon documents obtained by the “Inquirer” apparently confirm what customers have been saying for years about Verizon technicians’ reluctance to fix copper lines. “Once at the customer’s home, the Verizon technician tells the customer that the only solution is to switch to fiber, which includes the installation of a FiOS box,” the Inquirer reported. “If a flagged copper customer needing repairs ultimately declines fiber upgrade, the Verizon document commands: ‘Do not fix trouble’ with the copper line.” (The newspaper quoted from the documents but did not publish them in full.)’

    All that being said fiber is a better medium for the transmission of data and it emits less radiation than copper. Supporting existing copper networks becomes very very difficult as speeds increase. Some smart young engineer at Digital Equipment Company (DEC waspart if the triumvirate that introduced ethernet along with Intel and Xerox) thought they could use existing telephone wiring in buildings. The program had two flaws. First the telephone companies didn’t use existing wiring although they knew there was unused wiring that had been installed and abandoned. They didn’t know where it went because they didn’t document it they just pulled new wiring as needed. In addition the quality of existing wiring was problematic. As wires were made they often spliced them together by wrapping two wires together to make the pieces longer or repair breaks then added insulation put it on a reel and sold it. These splices had no impact on voice but they could give ethernet fits. Also there was often little bits of corrosion in connections, remember some existing wiring was decades old in the 1970s, and these also gave digital signals fits.

  7. That’s a rather broad description. And I also don’t think it’s correct to claim “there is no minimum dose that will not cause trouble” because I have yet to see literature that would attempt to prove that statement (there have been experiments that try to dis-prove it though). While the mentioned sources of radiation are indeed all part of the electromagnetic spectrum and they are all governed by the same laws of physics (Maxwell), the way these types of radiation act on the human body can be most different. Some of it is ionizing, some of it isn’t. Non-ionizing radiation cannot be lumped into one basket, just as we distinguish between the effects of a hard gamma emitter and a dentist x-ray even though both are considered ionizing. While I personally am not worried about 5G per se, as a physicist I don’t see how anybody benefits from broad generalizations. When you tell people not to worry, because “it’s all the same and harmless”, that usually gets everybody worried.

    So the questions might be: What are the differences between 5G and LTE in terms of spectrum, power, etc.? Have those differences been scientifically studied in terms of their effect on the human body? Are those studies extensive and the results conclusive? Have they been reproduced (to quote one of my faculty colleagues, “one study is no study”). Those I believe are the kinds of questions and the cautious but diligent attitude that advances the public debate over these issues.

  8. There’s a lot of broad information that’s known and applicable. It’s still all non-ionizing radiation. Thermal effects are the main predicted concern, and my article is about how those concerns haven’t played out. (Because the radiation isn’t ionizing, it can’t cause the sort of direct cellular damage that X-rays, gamma rays, etc., can cause. The only way for the EMF to produce an effect on a subject is through heating.)

    Computerized models that use real-world data to predict effects are developed. Signal strength can be calculated, as can be absorption, and this can also be tested with material and rodents—rodent tests haven’t yet been performed, but we know the limits of those in any case. (One thing I didn’t discuss is that rodents have been a particularly bad way to predict risk except in dosing chemicals across most fields of study, but they’re still used because nothing is considered better…)

    The key fact is that because the wavelength is shorter and because base stations will be more densely deployed by necessity, the overall signal strength at any given point will be less, potentially far less, than comparable 2G, 3G, and 4G networks—possibly even less than the average Wi-Fi signal in a home.

  9. dwstclair: But they are all the same thing.

    Simon: That’s a rather broad description.

    But it is true. If not explain the difference.

    Simon: And I also don’t think it’s correct to claim “there is no minimum dose that will not cause trouble” because I have yet to see literature that would attempt to prove that statement (there have been experiments that try to disprove it though). While the mentioned sources of radiation are indeed all part of the electromagnetic spectrum and they are all governed by the same laws of physics (Maxwell), the way these types of radiation act on the human body can be most different. Some of it is ionizing, some of it isn’t. Non-ionizing radiation cannot be lumped into one basket,

    Whoa big fella. The fact you have no data to disprove my contention is not a rational argument disproving my contention.

    Part of your argument is based on the false assumption there is no ionizing radiation emitted by microwave sources.

    The likelihood that any particular particle or wave with break a bond is based on the strength of the particle or wave and the strength of the bond. It would seem that those strengths are not really that predictable or that the strength of all the particles or waves from some source are all the same. There is no reason to assume the iPhone signal to a cell tower does not contain a wide range of particle and waves. Clearly most will be around the intended frequency but the possibility of an X-ray now and then still remains. If you study Wien’s displacement law which states that the blackbody radiation curve for different temperatures peaks at a wavelength inversely proportional to the temperature. Everything which is not at absolute zero emits radiation of all strengths. It is a cold day in hell when things a couple degrees above absolute aero put out a cosmic ray but it does happen. Which means your stove and refrigerator will occasionally produce ionizing radiation.

    I am guessing your area of physics does not deal with radiation. I worked for Baird-Atomic in the 1960s. We were building stuff for nuclear medicine. Our top of the line produced an image of a roughly 10" square area and contained 128 pixels. Another item was a neutron activator that was installed at Boston City Hospital. An engineer from the company that made the activator walked past me in the hall talking to an engineer from our company (we made the lead room, power supplies, and controls) and he said, “We have a lot of dying people in the hall and I think we are getting the technician too.” All the physicists working on those projects not only wore dosimeters at work they logged the radiation doses received by their families for X-rays as well as reviewing and sometimes resetting the doses set by the dentist. One said, “All the dentist knows is don’t point it directly at the patient’s brain and use about x ms.” They went on to say that dentists didn’t seem to know that the enamel on people’s teeth would bounce the beam into the brain if the wrong up angle were set fo the beam." I wore a dosimeter for four years while working on RADAR equipment that produced a 32 Megawatt pulse. I taught a course in Systems Safety Analysis at Picatinny Arsenal to the folks that designed the Army’s nuclear weapons. Their approach to radiation like mine is much more conservative than yours. I think it boils down to a difference in risk management approaches. Most people, like you, decided to take a risk or not based on the ‘likelihood of the event’. If it is more than a level they are comfortable with they take the risk. There are a few others like me who decide to take a risk-based on their ‘magnitude of regret’. I know the risk is tiny but I don’t rest my arm on the window sill of my car because I would have a high magnitude of regret should I get sideswiped and have it ripped off.

    Simon: just as we distinguish between the effects of a hard gamma emitter and a dentist x-ray even though both are considered ionizing. While I personally am not worried about 5G per se, as a physicist I don’t see how anybody benefits from broad generalizations. When you tell people not to worry, because “it’s all the same and harmless”, that usually gets everybody worried.

    Simon: So the questions might be: What are the differences between 5G and LTE in terms of spectrum, power, etc.? Have those differences been scientifically studied in terms of their effect on the human body? Are those studies extensive and the results conclusive? Have they been reproduced (to quote one of my faculty colleagues, “one study is no study”). Those I believe are the kinds of questions and the cautious but diligent attitude that advances the public debate over these issues.

    I think I have answered that question.

  10. Moderator’s note here: I have edited the last few posts to remove or change the bits that could be construed as attacks—I will not allow this to devolve. I don’t believe anyone meant ill, but I am utterly sick of people arguing with each other, rather than discussing a topic. If you disagree with something, try asking a question to learn why the information disagrees with what you believe to be true.

    I also edited the post above to use Discourse’s quoting and make it easier to read. If you’re writing in the Web interface, just select the bit you want to reply to and click the Quote button, or just put an angle bracket > in front of quoted bits. If you’re writing in email, don’t quote because it won’t work well.

  11. It’s a blog entry that the publication doesn’t vouch for (see disclaimer) by a long-time EMF truther with invalid scientific citations.

  12. This is not accurate. The NYT joint venture is a journalism project, not an investment in the development, marketing or sales of 5G products or services. It’s all about content. 5G will have about as big an impact on the gathering and delivery of news as moveable type, printing presses, mail delivery services, radio, television, and the internet etc. has had.

    The first iPhones were 2G, and the impact they had on they had on podcasts and news delivery were groundbreaking. 3G started to get video clips just about off the ground. 4G made streaming content viable. One of the reasons the NYT has been able to survive when so many other newspapers have folded or consolidated is because they always jumped on new ways of digital media:

    To quote the article:

    “We believe 5G’s speed and lack of latency could spark a revolution in digital journalism in two key areas: how we gather the news and how we deliver it. In the short term, having access to 5G will help The Times enhance our ability to capture and produce rich media in breaking news situations. Over time, as our readers start to use 5G devices, we will be able to further optimize the way our journalism is delivered and experienced.”

    Having worked in magazine and news publishing for more decades than I care to admit I’m old, I have also seen the evolution of internet speed continues to revolutionize the way journalists gather and organize information. In addition to sending info to the newsrooms, live streaming and off loading video will become easier. It will even make traffic, weather and transit reporting better.

  13. But, but, it is so comfy with an Apple Express under my pillow!

  14. I essentially agree with the main conclusion of the article: There is no empirical evidence that cell phone radiation causes cancer. However, the part that talks about “p-hacking” had a slew of egregious errors. There are plenty of problematic research designs and the media feasts of the occasional sensational exposure of research mal-practice. But the critique as stated in this piece is superficial and in places not well informed: there is nothing “sacred” about p. > .05 neither will pre-stating hypotheses (a.k.a. prospective design) save us from the occasional erroneous conclusions. Every (!!) honest research design is a compromise between risk of overlooking a real effect and mistakingly making an assertion of an effect where in fact there is none. In the “trade” known as type I vs Type II error. I appreciate that the details of research design alternatives and the risks of different statistical analyses would make a boring article. However, if the matter is too complex for the kind of article --or if you do not fully understand what you talking about-- it is wiser to omit than to add to the confusion the public already been subjected to a great deal of misinformation on how research works. Love this publication, but please be a little more careful in what you write. Adam (with 35+ years of experience in the kinds of studies you are talking about)

  15. The “concerns” over WiFi (and now 5G) are just a less dangerous version of antivaxers or a slightly more annoying version of audiophiles, built on a lot of misunderstanding and a total lack of any actual evidence. The amount of electromagnetic radiation produced by these systems is tiny, and has been studied and there;’s never been the slightest evidence of anything at all. Confirmation bias, even unconscious, is compelling, but it is not science. You can still spend $1000 on speaker cord that performs identically to w couple of coat hangers for your home audiophile setup though.

    Let’s change the terms here, ok. Water is dangerous. Water causes flooding, erosion, and all sorts of problems, just like a strong microwave oven leaking microwaves could cause serious health and safety issues. Now, a gallon of water is a lot less dangerous than a 1,000 year flood, sort of like the electricity in your house which is pretty safe, but yes, you could kill someone in a gallon of water. Now, with 5G we’re talking about something on the scale of between a single drop of water and maybe a teaspoon.

    Yes, Analogiens aren’t the best, but still WiFi and WiFi type radiation is a lot less dangerous than a light rain storm.

    And the article on Scientific American is not a Scientific American article, as Glenn rightly pointed out.

  16. While I agree that the CURRENT microwave cellular system poses no health risks, it is by no means clear that just because the microwave frequencies are fine that 5G will be as well. The whole point is that 5G uses frequencies that are absorbed by the body (unlike microwave frequencies that pass right through). So to say that just be microwave frequencies are fine, 5G will be too is akin to saying that if the cellular industry wanted to use x-ray frequencies everything would be fine as well. You need to look at the frequencies used and the POWER needed to permeate walls and human bodies and then only after studying these new frequencies and power levels could one (maybe) say they are safe. I’m not yet convinced that 5G frequencies at the power levels needed to permeate walls and bodies are safe. I would VERY happy if they are. But I don’t think that’s conclusive yet (and certainly not just because the very different microwave are safe)

  17. Thank you. This physicist loves to see the popular press (yes, you) speak in sensible language about an issue resolved long ago. And, your discussion and explanation of conspiracy hoaxes deserves an award.

    Well done!

  18. Excellent article Glenn, thank you.

  19. I accept and recognize your skepticism, and it’s entirely healthy so long as you don’t go on–as you are not!–to assert that it’s inherently dangerous. Skepticism is very powerful when it is paired with an open-eyed realism about accepting empirical data as it expands.

    I don’t think we have a particular need for 5G, honestly! I think that’s a market-driven interest–a hammer in search of nails, so the industry is making a lot of nails, too. The Internet of Things (IoT) has already shown that most non-computer/non-smartphone devices attached to the Internet have insecure operating systems and are easily suborned. That’s true of a lot of (but not most?) computers and smartphones, too. The 5G promise is that everything will be connected, a billion billion devices. Great! All sending us spam.

    But on the issue of 5G safety: it’s still non-ionizing radiation; it is, absolutely, still microwave radiation, even if the wavelengths are shorter; signal strength will be strictly below several factors below expected potential harm; and biological testing and models reveal nothing unique.

    I don’t think based on all that, there’s any chance given what we know already from the last 20 years of research that a magic new biologic effect will emerge just because the wavelengths are smaller. As you recollect from the article, microwaves lack the energy that ionizing radiation has, so the entire effect is heating or “thermal.” Thus, any higher-frequency 5G signal effects would have to demonstrate they use a pathway to produce unusual heating that no models can show they do.

  20. I don’t think that’s the argument. Note also microwaves do not “pass right through” tissue (that’s why microwave ovens work).

    LTE uses very roughly 600 MHz to 2.mumble GHz (~ 10-50 cm). 5G uses those similar bands plus “mm-waves”, i.e. IIRC 25-70 GHz (4-12 mm).

    My microwave oven at home uses 2.45 GHz which is why it has to be shielded if it’s not to interfere with my wifi (which it by the way still does when it’s running). The reason it works as an oven is because there is are rotational/vibrational modes in water at those frequencies. The human body is to large extent water so microwaves are very well absorbed by our bodies. We are familiar with that effect when we get a warm ear after using a cell phone for prolonged periods of time without a headset. However, a warm ear doesn’t mean danger and there has been a lot of research that showed that despite local heating, use is safe. An obvious crucial difference is that while my microwave puts 900 W into my popcorn, my iPhone emits on the order of ~2 W when setting up a call IIRC.

    The 5G absorption issue has I believe to do with the fact that mm-waves are much more easily absorbed by everyday structures around us and hence range is reduced compared to LTE. That is why we hear that 5G will require many more antennas to be set up and/or signal strengths could be increased. But ultimately, both LTE and 5G employ wavelengths (“microwaves” which covers everything from 1 mm to 1 m) that will lead to heating of human tissue. And neither is ionizing. While it is reasonable to assume that a mm-wave will not have the exact same effect on human tissue as a 0.5-m wave, it is just as correct to point out that both will interact with tissue using the same fundamental mechanism (i.e. heating vs. for example generation of free radicals). I believe this is what Glenn points out.

  21. The signal from a phone is not directed like that of a magnetron in a microwave oven, and cell phones rarely use wattage that high, only when extremely far from a transmitter, and only for extremely brief periods.

    There is no evidence that a warm ear from a cell phone has anything to do with wireless signals, holy cow. Find me a peer-reviewed study, because I’ve been reading research for decades, and I’ve only seen this claim on tin-foil hat sites.

    It requires many more transmitters because the signal strength will be as low as other forms of wireless communication, meaning that its propagation will be relatively short. I would love to know where you’re reading this stuff, so I can debunk it better!

  22. I can’t help but remember how internet memes were flying around about Airport and Airport Express radiation when they were first released, like this discussion here:

    And more recently, AirPod radiation causing cancer:

    My thanks again to Glenn for debunking memes and clearly articulating facts in an excellent article.

  23. Super High Frequency is a long way from ionizing radiation, but it still carries energy. It would be helpful to know how much a 5 GHz wireless transceiver heats a quantity of water sitting right next to it. Now, if the water in your brain were heated the same amount, does this result in any biological damage? I kind of doubt that it would, as you need frequencies about a million times higher, in the petahertz range, before the radiation is ionizing.

    And don’t think that using earbuds is going to help here. If you’re using wireless earbuds, the signal strength doesn’t degrade that much, and your sending that right into your head. If you’re using wireless earbuds, you’ve moved the transceivers that much closer to your brain.

  24. It’s negligible and covered by FCC regulations and studies. Consider how long it takes a 1,000-watt microwave oven with an enclosed surface reflective to bounce the EMF repeatedly instead of letting it dissipate over distance. A 5G network transmitter (or a 5 GHz Wi-Fi network base station) produces vastly less energy and disperses it over a broad directional swath or omnidirectionally (not entirely, but mostly like a sphere of signal emitting outwards).

    Wireless earbuds use Bluetooth, which transmits at extremely low signal strengths, orders of magnitude lower than Wi-Fi, often, which is how they have long battery life with tiny batteries, and use frequency hopping (FH), which dwells on a particular frequency swath very briefly, instead of continuously outputting across a common range.

  25. Except microwaves operate at 2.4 GHz, so a 5 GHz signal is carrying more than twice the energy. I’m only concerned about the possibility of heating causing damage to living tissue.

  26. That is not how that works. I would highly suggest to learn more about this, you consult the FCC and CDC sites for background reading.

    And as noted above 5 GHz ≠ 5G networks.

  27. Actually, a household microwave operates at up to 2.45 GHz, but I used 2.4 GHz for simplicity. And while I appreciate your advice, my first amateur radio license issued more than thirty-five years ago by the FCC required a pretty good understanding of frequencies and radio waves.

    This is a decent explanation that energy is directly proportional to frequency:

    Moreover, I never conflated 5 GHz with 5G; a wireless administrator would not make such a mistake.

  28. The FCC regulates output signal strength, so the issue isn’t potential power with the same inputs, but how the output is controlled. However, as it stands, the FCC (and with varying rules, other regulators around the world) allow far higher maximum EIRP from 5 GHz networks than comparable 2.4 GHz, partly because the signal propagates less far.

    Regardless, the highest output levels permitted for consumer microwave equipment–stuff that’s not mounted away from people, locked up, and covered with warning signs–is designed as I note in the article to be factors below to orders of magnitude below the lowest level at which research shows any thermal effect that could affect health.

    The article focuses on 5G networks, and 5 GHz is not substantially higher than 2.4 GHz compared to far higher non-ionizing frequencies that 5G encompasses, so I am sorry to be confused by your focus on a widely deployed technology.

  29. That explanation pertains to the photon model of light, which is useful for understanding the difference between ionizing and non-ionizing radiation. It’s not really useful for understanding how a wifi or 5G signal interacts with the human body.

  30. We are SOOOO overdue for seeing a summary like this one. Thank you, Glenn!

    But first, how can you talk about p-hacking, without citing a world-recognized analysis of it, at ?

    Second, a feature that is absent from all the reports I’ve seen, is that there are so many published studies that look at impact from a chemical or EMF or whatever, that do not find a strong association. We often see the suggestion that because the study examined “only” 300 rats (or 5000 people, or whatever), there may be an effect that could be found with MORE subjects. The flip side of this coin is the statistical power of the study: if the effect were strong enough to show a correlation in 300 rats, then it would be at a level of X cases per 100,000 humans; the absence of a correlation suggests that fewer than X cases are expected. Of course, “fewer than X” can mean “zero,” but there is NO NUMBER OF SUBJECTS large enough to absolutely guarantee X is actually zero, that there is NO RISK WHATSOEVER from the tested agent.

    My impression is that a meta-analysis of EMF or other studies could set this upper limit of risk, and that we would see X is so small that surely we should spend our collective energies on other things with larger X’s, e.g., the #GlobalCO2Crisis.

    PS: I’m exactly NOT anti-research, and I’m generally happy with the focus of research dollars. What makes it into the press is a completely different story. Keep up the research about as before—it probably represents smart priorities—but let’s have some better reporting on what we’re working on, why we’re working on it, and what lines of research aren’t likely enough to be valuable to spend our energy on them

  31. What about research showing EMF impact on voltage-gated calcium channels (VGCCs)?

    “This article reviews, then, a substantially supported set of targets, VGCCs, whose stimulation produces non-thermal EMF responses by humans/higher animals with downstream effects…”

    It’s noteworthy that the paper mentions there are both therapeutic and pathological effects.

    What if there is a biological impact but it falls short of a “provably-causes-cancer” threshold? What if it just makes one weaker or less resilient? (Yes, that’s more in favor for EMF hygiene - use a headset, don’t put your wifi under your pillow while you sleep, etc.)

  32. This isn’t a study, but a literature examination by someone who has consistently and repeatedly, without substantive evidence, pushed a message of EMFs causing harm. He lectures around the world about it.

    There’s nothing wrong with being a contrarian with evidence, or being a gadfly. And it’s always possible that someone would find a pattern of evidence other people miss, and which is valid. A friend’s wife received a MacArthur Fellowship for her work, where she spent well over a decade ignoring conventional wisdom in her biological research and her work changed her entire field.

    However, there are a handful of scientists around the world who have spent from several years to more than two decades acting as anti-EMF enthusiasts without engaging scientific rigor. To avoid the potential of libel suits—which can be filed even when one is telling the truth and would win such a suit—I won’t list the several people who are well known for this here.

    I will note that in my article, I point out two of the most prominent people who have pushed a particular, inaccurate view of EMF absorption were quoted in the New York Times, and one essentially recanted his position to the reporter and the other was unable to defend it. So.

  33. Thanks for your reply. You’ve obviously researched and thought about this much more than I have; I wasn’t trying to suggest you’re wrong or 5G is more dangerous than other EMF types, I’m genuinely curious about these areas that don’t seem to get much discussion. I didn’t mean to cite an article from a questionable author, his history obviously weakens the case.

    In general, it seems like most of the EMF discussion and research is on the thermal effects - rightly so, when the thermal load is high the impact can be significant. But there are also non-thermal biological effects of EMF exposure. Are those effects significant? I honestly don’t know. (Here’s one example - it might affect quality of sleep in rats - obviously not relevant to the discussion about 5G - they used different frequencies, but a good example that there are non-thermal biological effects.)

    Regardless, those are more concerns about the biological impact of EMF in general, not about 5G specifically, so I’m bringing my questions to the wrong article.

    I get and support the point of your article - there is unfounded hysteria about 5G with no supporting evidence. While skepticism might be reasonable, hysteria is not necessary. You do a good job of addressing that. Thanks for the good work you do.

  34. Sorry to push back so hard, but that fellow in particular is one of the people who makes it difficult to sort out potential risk from a miasma of thinly applicable aspersions.

    The thermal issue keeps coming up, because with non-ionizing radiation, there’s no known theoretical mechanism–much less one that can be quantified–that would produce other effects, particularly at the extremely low energy levels that 5G high-frequency transmitter will use.

    The goal of most cellular networks is to use just enough power to be distinguishable from background noise, which is mostly the rest of natural emissions. So I believe that folks hearing about 5G requiring millions of new base stations and (in the future) using new frequencies as something huge and scary, while we already have Wi-Fi deployed with a billion base stations at higher power than 5G will use with no measurable ill health effects.

    I do see the concern about millimeter wave radiation’s potential to penetrate living things in different ways, but the baseline amount of energy that would transmit and its nature remains at the heart. If the frequencies aren’t such that they can knock around stuff at an atomic scale, but rather just at worst agitate molecules a bit, then thermal is all there is to consider.

  35. Glenn, did you include the survey paper “Thermal and non-thermal health effects of low intensity non-ionizing radiation: An international perspective” (2018; doi:10.1016/j.envpol.2018.07.019 ) in your research? Notes in the summary of that paper seem to contradict what you are saying here:

    There is strong evidence that excessive exposure to mobile phone-frequencies over long periods of time increases the risk of brain cancer both in humans and animals. The mechanism(s) responsible include induction of reactive oxygen species, gene expression alteration and DNA damage through both epigenetic and genetic processes. In vivo and in vitro studies demonstrate adverse effects on male and female reproduction, almost certainly due to generation of reactive oxygen species.

    Earlier in the discussion, you said, “The thermal issue keeps coming up, because with non-ionizing radiation, there’s no known theoretical mechanism–much less one that can be quantified–that would produce other effects.” The quantification of reactive oxygen species (ROS) in biological tissue is largely problematic because the charged particles are by their nature highly reactive. Historically, ROS concentrations have been measured indirectly by the consumption/depletion of antioxidant molecules in tissues. Quantification of ROS and its effects is a human shortcoming, but the health impacts of ROS should not be ignored.

    Broad issues about ROS have only been studied and found in only the last few years. For instance, we know that burning β-Hydroxybutyrate (the smallest ketone body) in our mitochondria generates far less ROS than burning glucose. This small change is a principal factor in the positive neurological performance on a ketogenic diet (as noted in papers like this (2007) and this (2012). We’ve known for over 75 years that LCHF diets are highly effective at treating diseases like childhood epilepsy, but the mechanisms have been elusive until now.

    I’m hardly an expert in this field; I’ve been studying bits of the literature for about 5 years. The positive results (neuro and otherwise) that I’ve seen on a LCHF diet made me highly curious about the science; it’s given me some literacy in the questions about non-ionizing radiation. I’ve just started to read the 2018 survey paper I started with; reading and chasing links will take some time. I welcome your commentary on that survey article.

    Is the energy threshold of ionizing radiation a sufficient threshold for our safety? I do not know. It’s certainly one of the more interesting science questions out there today.

  36. I have doubts that 5G will do much for mobile. Latency doesn’t seem to be a big issue and the poor penetration through walls and short distances it can travel from base stations makes it impractical. It could improve GPS, but not enough for us civilians to care. However, what if you’re not mobile?

    Home use seems to be where 5G will really shine. If your house can receive the high frequency 5G signal, it’s going to keep receiving it, and high frequency 5G speeds are compatible to home internet. For almost a decade, the two major telephone providers, Google, and other companies have tried to compete with cable providers. FIOS and U-Verse have been a bust. Even Google didn’t have deep enough pockets to implement their own fiber network.

    It’s just too damn expensive. You have to put fiber on every street and every block. It costs hundreds of millions to wire up even a small neighborhood and even then, you’re not guaranteed to have any customers.

    However, high frequency 5G doesn’t depend upon tearing up streets and cellphone providers have customers. 5G towers are also smaller and could even be placed unobtrusively on a rooftop. Unlike fiber, 5G is cheap to implement and companies already have the customers. Imagine a company implementing 5G home internet service neighborhood by neighborhood. In the end, maybe only 30% of the US would have 5G coverage, but that would include 95% of the US population.

    I see 5G as a way to break the cable monopoly’s stranglehold on home internet. By the way, I just got notified that my home internet 100mbs service (which gives me around 40mbs service) is going up from $73 per month to $90 per month. Oh, they’re also charging me a $2.50 future internet service fee.

  37. One of the coauthors of that paper, David O. Carpenter, is mentioned in Glenn’s article:

    “One prominent expert witness who relied on the chart in testimony for years admitted that he had no idea that skin blocked the passage of high-frequency signals in particular. “…maybe it’s not that big a deal,” he said.”

    Doesn’t fill me with confidence about the paper.

  38. This is a pseudoskeptical dismissal.

    The beauty of science: you can feel any you want when evaluating the science. You can be happy, sad, full of confidence, or even entirely lacking in confidence. It doesn’t matter at all; your personal feelings are irrelevant. All you have to do is download the paper and evaluate the science on its facts and reasoning.

    Glenn is right: if there is a health impact from cellular phones, there must be something materially altered in our tissues attributed to the cellular phone radios. Reactive oxygen species is a good candidate. We already know about the risks of excessive ROS concentrations for other health issues. And note: Glenn didn’t mention ROS anywhere in his discussion to this point. In order to categorically state that these cell phones are safe, a discussion of this particular mechanism is mandatory.

    Please read the paper. If you think the ROS risks outlined can be dismissed, please give us the facts–from the paper–that led you to that conclusion. That’s how scientific discussion works.

  39. No, it’s a “we have evidence an author of this paper really has no idea of what they’re talking about” dismissal.

    That sounds reasonable but it’s deeply problematic in two ways: 1. there are thousands if not tens of thousands of papers written on scientific issues and no one has the time to read them all, which means that it’s impossible to read one paper and understand how it fits into the larger scientific consensus and 2) as Glenn has pointed out, it’s reasonably easy to put together a credible scientific paper that is nonetheless absolutely misleading by making a range of different choices in organization and analysis.

    In all the “bits of the literature” that you’ve read, did you download the data sets behind the articles to see if the authors were fiddling with things? Did you with this one?

    As for me, I’m going to stick with my “we have evidence an author of this paper really has no idea of what they’re talking about” dismissal.

  40. Oh Glenn. I hope this article doesn’t come back to bite you.

    The question of wireless radiation causing cancer in humans remains wide open. The rodent test you (IMHO) cherry picked is at odds with two other rodent tests that verify one another and indicate wireless radiation can indeed cause cancer.

    Here are relevant article links, all well worth reading:

    Cell Phone Radio Frequency Radiation @NTP

    Report of final results regarding brain and heart tumors in Sprague-Dawley rats exposed from prenatal life until natural death to mobile phone radiofrequency field representative of a 1.8 GHz GSM base station environmental emission

    $25 Million NIH Study Proves Wireless Technology Causes Cancer and DNA Damage - US Brain Tumor

    5G And The IOT: Scientific Overview Of Human Health Risks

    IARC, WHO: Move Radio Frequency Radiation from Class 2b to Class 1

    As usual, the effects of exposure to EM radiation are a matter of:

    1. Wavelength of EM rays,
    2. Amplitude of EM rays,
    3. Length of time of exposure.

    We wait for further studies specifically relevant to human exposure.

    There is no reason for paranoia or panic at this point. But wisdom dictates minimizing one’s exposure to wireless network EM radiation. One helpful rule is the Inverse-square Law. The intensity of EM radiation is inversely proportional to the square of the distance from the source. IOW: Exposure drops off exponentially the further you are from your cell phone or WiFi device.

    Inverse-square Law @Wikipedia

    A possible exception would be when one is exposed while in the path of EM beamforming.

    Beamforming @Wikipedia

  41. They are worth reading – for example, the first study shows that exposure to RFR actually lengthened the lives of the tested rats. Excellent! Time to get on my cell phone even more.

  42. I would not try to prevent you from expressing your opinion, but the entire article is devoted to the extensive and still-growing set of evidence that finds no linkage between EMFs from regulated wireless devices and health effects. I call out the rodent study because it’s both recent and is an example of overreach, as it demonstrates what happens when you dig through data to find any correlation instead of looking at the effects as the whole.

    I’ve read some of those links in the past and read others of them now. It’s critical to distinguish between high-exposure tests, which subject rodents to levels of radiation far beyond the outer safety limits of wireless devices, and those that look at regulated device exposure. This is discussed in my article. Groups like the US Brain Tumor Association lack scientific rigor and credibility.

  43. We’re starting to devolve into a not particularly useful back and forth here, so I’m going to close comments on this article. Suffice to say that more solid, peer-reviewed science is always good, and given the attention this topic gets, I’m sure more studies will be done.

Join the discussion in the TidBITS Discourse forum