The language of research

 I came across an item today.

There was nothing spectacular about it – it was just one of those endless mindless "research" projects that is supposed to shock the world but in fact is just bolstering a few egos, justifying a few salaries and with a bit of luck a drop of future employment [“further research is needed”].

Three drinks a day linked to cancer


What does intrigue me about these "studies" is the language used.  They are an absolute goldmine of weasel words.

"Drinking three alcoholic drinks a day can cause liver cancer, research has suggested.

Being overweight or obese and consuming foods contaminated by aflatoxins (toxins produced by certain fungi) also showed “strong evidence” of causing the disease, the World Cancer Research Fund said.

Its research, which analysed 34 studies involving 8.2 million people, more than 24,500 of whom had liver cancer, also suggested higher consumption of coffee “probably” protects against liver cancer, while physical activity and fish consumption may also decrease the risk although further research is needed."

In other words, their research showed little or nothing.  They might as well have declared that riding bicycles may cause earthquakes and that there is strong evidence that pruning an apple tree on a Monday may contribute to climate change.

Notice however the first sentence.  They state the startling result for the sake of the headline, and then qualify it knowing that the headline will omit that bit.  It has worked perfectly in this case, and the newspaper has obediently and slavishly printed the opinion as fact.

So here we have [and I quote] "the most comprehensive review to date of global research into the relationship between diet, weight, physical activity and liver cancer" and they come up with little or nothing.  A study of that magnitude should throw up definite results if there are results to be found, and not something that is full of ifs, buts and maybes.

The fact that they have discovered sweet fuck all doesn't deter them from making recommendations however. 

"[..] women should try to limit their alcohol intake to one drink per day and men to two drinks a day.

Other recommendations included being as lean as possible without becoming underweight and carrying out physical activity for at least 30 minutes every day."

And they then come up with their dramatic conclusion –

"Around three or more drinks per day can be enough to cause liver cancer."

Unfortunately for the "researchers" the paper has also [by mistake?] printed a lovely fisking job by a Cambridge Professor who points out that the whole project is effectively meaningless.

The damage has been done however.  The "researchers" have got their headline and the Proles will lap it up.

How long before we see the Puritans quoting this "study" as gospel in their quest to abolish alcohol?

I can see the headlines now…

It's only fair to share...Share on FacebookShare on Google+Tweet about this on TwitterShare on LinkedInPin on PinterestShare on RedditShare on StumbleUponShare on Tumblr


The language of research — 6 Comments

  1. I read this earlier. 3 drinks MAY cause liver cancer. If they may then they may not too. 1% of men and 0.5% of women get liver cancer. BUT…….most liver cancer is triggered by one or more hepatitis viruses. This information was omitted. Of the 8.2 million patients, only 0.3 had liver cancer. The vast majority did not. There was no timescale quoted re the duration of having 3 drinks a day and then 10g of alcohol (just over 1 'unit') shaves more life away. Well, the stats are saying we cannot tell if drinking causes liver cancer. If 99% of 8.2 million had it then OK, but 0.3% is noise. This study is junk science littered with opinions and those are of well known temperance and prohibition campaigners. It was junk science and there were no concrete analyses of the results. It was rubbish.

    • Of course it is rubbish but will that stop the fanatics quoting it?  The problem is tat people see "World Cancer Research Fund" an assume that it is a respectable high-level institution and wouldn't dream of questioning it. It's like having "Doctor" or "Professor" in front of your name – every utterance is gospel and always must be fact.

      Science has gone so far down the toilet that soon people are just going to have to smell the stench.

  2. And don't drink tea.

    "Most people love a good cup of hot tea, so it may be surprising to learn that your morning cuppa could be bad for you.

    Recent research at University College London suggests that drinking tea could be detrimental to health, causing frequent nosebleeds and brittle bones.

    It is thought that the steam from tea can lead to the weakening or rupturing of the vessels in the nose, causing nose bleeds.

    Anyone who has had a nosebleed is advised to avoid drinking any hot drinks for 24 hours.


    Researchers found that drinking the liquid while it is extremely hot is especially bad for you, and that allowing the tea to steep for at least five minutes will not ruin the flavour."

    • For fuck's sake!!!!!  I drink about a gallon of tea a day [actually not an exaggeration – I must cut back to around half a gallon] and have had, as far as I remember, one nosebleed in the last fifty years.  My bones also still seem to be supporting me well enough.  The last broken bone breakage was around twenty years ago [three ribs!].  That was not down to my beverage consumption either.

      In other words – a load of steaming bollox.  [let the bollox stand for five minutes though – that steam can be deadly]

  3. I sometimes drink a cup of coffee with brown sugar and a double tot of Jameson whiskey. If I live to be 90 I'll let youse all know. 

Leave a Reply

Your email address will not be published. Required fields are marked *