This week, Heather Heying has written about the curious fact that not only the Covid “vaccines” but also many traditional vaccines have never been tested against placebo, but only against prior vaccines. Such a “turtles all the way down” approach (the current vaccine having been tested against its “father”, the “father” against the “grandfather”, and so on) may work in some situations, but will fail badly in others. I would like to provide a few analogies.
Mathematicians often employ the proof technique of induction. You observe, for example, that
1 + 2 = 2 * 3 / 2,
1 + 2 + 3 = 3 * 4 / 2,
1 + 2 + 3 + 4 = 4 * 5 / 2,
and since you are not afraid of math with letters (i.e., algebra), you conjecture that
1 + … + n = n * (n+1) / 2
for all n = 1, 2, 3, …, i.e., for all natural numbers. Proof by induction then consists of two parts:
The induction step assumes that the formula has already been proven for a specific, yet unknown n, and demonstrates that the formula also holds for the successor of n, which is n+1. If we know that
1 + … + n = n * (n+1) / 2
for some (specific, yet unknown) n, we can compute
1 + … + (n+1) = (1 + … + n) + (n+1) = n * (n+1) / 2 + (n+1) = (n+1) * (n/2 + 1) = (n+1) * (n+2) / 2,
so the formula also holds for n+1.
However, all of this amounts to nothing if we forget the base case (each falling domino will topple the next, but someone has to push the first domino). We have to show that the formula holds for n = 1, which it does:
1 = 1 * 2 / 2
By induction, we now know that the formula holds for 1+1 = 2, then for 2+1 = 3, then for 3+1 = 4, and so on, ad infinitum.
We might be tempted to view the testing of a succession of vaccines against a specific disease as some kind of induction: the ancestor is better than placebo (base case), and each descendant is better than its father (induction step). So where’s the problem? I can see at least four:
There might be no base case, i.e., no ancestor had ever been tested against placebo. Maybe the ancestor was a real killer, and its descendants just kill fewer and fewer people, but they still kill.
The definition of “better” changes with time.
The definition of “better” does not change with time, but is useless.
The definition of “better” does not change with time, and is not useless, but the actual checking of “better” is prone to error.
To illustrate problems 2, 3 and 4, let me present some analogies.
Intransitive dice
Consider Efron’s dice, a set of four six-sided dice A, B, C, D, with curious numbers on their faces:
A: 4, 4, 4, 4, 0, 0
B: 3, 3, 3, 3, 3, 3
C: 6, 6, 2, 2, 2, 2
D: 5, 5, 5, 1, 1, 1
Each die in this set has a 2/3 chance of winning against its successor – but D also has a 2/3 chance of winning against A. So, in a way, A is better than B, B is better than C, C is better than D, and D is better than A.
Analogously, the grandfather vaccine might protect against infection (protect better than the placebo, that is), the father vaccine might produce fewer side effects than the grandfather, and the child vaccine might produce more antibodies in eight mice than the father. And yet, all-cause mortality after the child vaccine might be higher than after a placebo.
Weather forecasts
On weather forecast web pages, you can often start little animations showing weather patterns (clouds, rain, snow) during, say, the past 90 minutes followed by a forecast for the coming 90 minutes. Usually these two halves of the animation look very different: in the past, stuff happened (clouds or rain suddenly appear or disappear); in the future, there will allegedly just be boring linear movement.
The point of these animations is not actual forecast but entertainment. In particular, it would not make sense to try to predict tomorrow’s weather using these incremental steps (meteorologists, who are among the best-calibrated people, know this).
Analogously, what is the point of, say, counting antibodies if all-cause mortality goes up after vaccine uptake?
Walking in circles
Why aren’t we able to walk a straight course when dropped in the wilderness without navigational help (a compass, the sun)? A while ago, some scientists studied this phenomenon by marooning people in the Pfälzer Wald (one of the largest forests in Germany) and the Sahara desert. It turns out that we walk in circles not due to biomechanical asymmetries (one leg being slightly longer than the other) but simply due to accumulating noise. That is, small errors add up, and you end up crossing your previous path, sometimes by repeatedly turning left and sometimes by repeatedly turning right.
Analogously, even vaccine trials free of shenanigans (if there are such) will suffer from small sample size and involuntary error. After some generations, we might end up in a place that is worse than placebo.
"Analogously, what is the point of, say, counting antibodies if all-cause mortality goes up after vaccine uptake?"
Surrogate end points are becoming more and more popular. https://www.fda.gov/drugs/development-resources/surrogate-endpoint-resources-drug-and-biologic-development#:~:text=A%20surrogate%20endpoint%20is%20a,to%20predict%20that%20clinical%20benefit.
I wouldn't even call it bad science. It's a scam, plain and simple.