Back in 2013 when we naively thought social media could be a force for political good, Andy Carvin wrote a book called Distant Witness, which was about his experiences covering and understanding political upheaval from the other side of the world, and using social media (especially Twitter) to do so.

The key point of Carvin’s book has nevertheless been on my mind these past few weeks – how do we use social media to understand complex problems, even if we are not there? And even if we are not specialists?

In mid-March there was a problem I wanted to understand: why were vaccine supply headaches in the European Union still so severe? At the start of my investigations I had no more insight into this than any other lay person. But as is the case so often when I have a question stuck in my head and there’s no stopping me until I have answered it.

The public debate back in mid-March was an emotive and uninformed one, based on allegations as to whether AstraZeneca was supplying the UK in preference to the EU, whether the EU had got its joint procurement wrong, and whether worries about the safety of AstraZeneca’s vaccine would damage take up, or if the more mundane issue of getting people to vaccination centres was really the problem.

I did what I always do when trying to grasp a problem like this. Work out what sources are available, cross-reference them, and build up a kind of framework to understand the issue. And then, if something does not make sense or does not add up, openly and publicly pose the questions as to why that is the case, and then rely on the wisdom of others who have either knowledge of a sector or are immersed in a different political, linguistic or media culture to my own to help me find answers.

Yesterday proved the value of this approach.

I chanced upon this piece in Spiegel by Claus Hecking about Johnson & Johnson supply shortages in Germany. Is J&J supplying the United States to the disadvantage of the EU was the question the article posed, and then sought to make a case as to why that was so.

An alarm bell was ringing in my head. That is the opposite conclusion to the one I had myself come to three weeks ago. Dave Keating had even managed to get some stats out of the European Commission on 12 April – but only concerning exports EU to US since 31 January, putting total exports for that period at just under 1m. And on 28 February the New York Times had published a number of doses of J&J vaccine that had been exported from EU to US – 3.9 million.

And yet here was Spiegel alleging something like 9 million doses had been exported. “Something really does not add up” I wrote in a Twitter thread – but I did not know where the problem was exactly. Having looked at the numbers the discrepancy could be even higher than Spiegel had implied – the US had at the time of writing used 8m doses of J&J and had 9.6m further doses ready to administer. With 3.9m doses sent from the Leiden plant, and no more than 1m of those sent since 31 January, where did the 4.1m to 13.6m other doses come from?

Option 1 – that Hecking sought to defend in tweets to me – was that all the J&J doses ready to use in the USA had come from Leiden. That would have rendered the New York Times 3.9m number wrong, and would be based on the assumption that all the extra doses were shipped to the USA before the beginning of the EU transparency mechanism on 31 January or that J&J had been exporting since 31 January without being in compliance with the mechanism.

Option 2 was that we were missing something, and that there was supply from somewhere else.

Twitter user @el_snoop provided the first clue as to the answer. A further plant – Catalent in Indiana – was producing J&J vaccine. A further Twitter user @arekmakarenko located the information about what the FDA had approved the Indiana facility, and that the Indiana plant was both a manufacturing and fill and finish factory.

Do we know all the extra doses over and above the 3.9m shipped from Leiden are from Catalent in Indiana? No we do not. But we do have a plausible explanation for those doses that better fits the facts than Hecking’s story does. On balance those doses are more likely to have come from Indiana than from Leiden.

Did I know that from the start? No. But I posed the right questions to the right people, and came up with a plausible answer.

But of course for a European publication – US stockpile of J&J doses comes from a US plant – is much less of a story than making an allegation that the doses were supplied from Europe.

Underneath all of this there is also the matter of attitude. Had Hecking and Spiegel wanted to check this, they could have asked. They could have had that “hang on, something doesn’t add up” kind of question on their minds, and posed it openly online and invited contributions. But that would have required the sort of selflessness that too few journalists seem to possess.