How a number travels

How a figure ends up in a law

A ministry asks whether the bow is an acceptable hunting method. An institute writes a report. The report cites studies, the studies cite other studies, and at the end of the chain sit a few dozen figures. They then travel on into opinions, press releases, newspaper articles and conversations between hunters.

Reading the chain end to end shows something no single document shows: the figures change on the way. Not through bad faith — through copying. Below are five cases found in the documents that decisions in northern Europe rest on. All of them can be checked: the primary source and the citation of it can be laid side by side.

1. An estimate that becomes a measurement

We told this one at length in the article about the documents read before the Danish decision. In short: the Swedish federation’s summary of the enclosure trial of 1998–2002 has no time column at all. The federation writes that the distances covered “should correspond” to flight times under 20 seconds — a deduction, in the conditional, from distances.

In the Hannover study, twenty years later, the same sentence appears in the indicative: flight times were frequently under 20 seconds. “Should correspond” has become “were”.

2. Blades that shift by one category

Jørgensen’s thesis, from 2013, compares flight distance by the number of blades on the head. Its results, in the table: 66.7 m for two blades, 44.5 m for three, 33.6 m for four.

The Hannover study cites it like this: animals hit with a two-blade head ran 44 m, those hit with three blades 34 m. Those are Jørgensen’s figures for three and for four blades, with the label moved by one category. And the moved figures are then used as evidence that the Madrid data “agree with the literature”.

A reader who only sees the German study therefore learns that the literature confirms a result the cited work does not contain in that form. What the three papers actually say is in our article on broadheads.

3. A study that changes state

The Danish report of 2021, a state document, compares its wounding rate with those of other studies and writes that Pedersen’s 2008 paper found 18% wounding of white-tailed deer “in Minnesota”.

The study is from Maryland. It ran from 1989 to 2006 on a naval facility south of Washington, with hunters who had to pass the IBEP course and an annual shooting test. Minnesota does appear in it, but in the discussion, where the authors cite another study, from Camp Ripley: there the hunters failed to recover 28% of the deer they had wounded, and the rate fell to 13% once the animals found by other hunters were counted.

The 18% is carried over correctly. Only the place has changed — and with it the reader’s chance of checking what kind of programme produced it.

4. A year that cannot exist

The Hannover study also cites, for mechanical heads, an article by the same Maryland team, with the year 2012. The reference gives the right journal, volume and pages, but the mechanical-broadhead article appeared in 2014 and covers hunting seasons through 2012. An article cannot be published two years before its last data.

The slip is small and changes nothing in the content. The practical effect is different: anyone searching for “Pedersen 2012” finds nothing, because in bibliographies of this kind the year is the key people search on.

5. A federation report that becomes an authority’s position

The Dresden summary, published in 2021 on the German federation’s website, describes the Madrid pilot project: 45 wild boar taken between 15 December 2011 and 31 January 2012, three of which needed a second arrow. The author writes that the information comes from statements by the Spanish environmental authority and from a veterinary report by the Madrid veterinary office.

In the bibliography of the same text, the first source appears under a different name: a report of June 2015 to the European Bowhunting Federation about the Madrid programme. A footnote adds that the total figure, 517 wild boar by 2021, was supplied by the author of that report.

This does not mean the figures are wrong. It means that, in the text, a report written for a federation appears as the position of a public authority, and a reader who does not go down to the bibliography has no way of knowing.

What to do with this

Nothing above shows any intent to mislead. It shows how information travels: each step copies from the step before, not from the source. An estimate loses its conditional, a study loses its state, a category shifts, a report loses its author.

Two rules follow, and we keep them in writing this site. First: a figure is taken only from the document that produced it, never from the one citing it. Second: when that document cannot be read, we say so in the text instead of citing at second hand.

It is an inconvenient rule — it slows everything down — but the five cases above were found exactly that way, by reading the originals for an article that only meant to say what is known about broadheads.

Citește articolul în română