Second-author traps: the position rule

Author–date styles cite by first author. Not by the author you know, not by the senior author, not by the one whose work the paper is really about — by whoever appears first in the published author list.

That sounds obvious and is broken constantly, for reasons that are entirely understandable. It also produces a defect that is unusually hard to find, because the citation and the entry both look completely correct on their own.

What the rule actually is

The entry
  Okafor, T., Lindqvist, M., & Duarte, P. (2021).
  Workflow variance in regional centres.

Correct
  (Okafor et al., 2021)
  Okafor et al. (2021) found…

Not correct, however reasonable it feels
  (Lindqvist et al., 2021)
  (Duarte, 2021)
  (Okafor & Duarte, 2021)   ← skips the second
                                author

The citation must begin where the entry begins.

The reason is mechanical rather than ceremonial. Your reference list is sorted by first author, so the first author is how a reader locates the entry. A citation naming someone else sends them to the wrong part of the alphabet, where they find nothing.

Why people break it

Three causes, and none of them is carelessness.

You know one of the authors. If Lindqvist supervises your field or gave the keynote you attended, Lindqvist is who the paper is in your memory. Citing by the name you think of is a small, natural slip.

The senior author is last. In many disciplines the principal investigator appears at the end of the author list, and is the person the work is associated with. Citing them feels like giving credit correctly and produces a citation nothing can match.

A reference manager showed you the wrong field. Some export formats and some display settings surface a different author than the leading one, particularly for edited volumes, and a citation typed from what you were looking at inherits the error.

The citation is correct-looking and the entry is correct. Only the pairing between them is wrong, which is why nothing you can read reveals it.

Why a matcher cannot simply look anywhere in the line

The obvious fix is for a checking tool to be generous — if the surname appears anywhere in the reference, count it as a match. That is worth explaining, because it fails spectacularly at any real scale.

Your reference list contains
  31 papers with Lindqvist somewhere in the
  author list — a normal number for an active
  researcher in one subfield

Your text contains
  (Lindqvist, 2021)

Lead-position matching
  → looks for entries BEGINNING with Lindqvist
  → finds one, or none, and says so

Anywhere-in-the-line matching
  → finds every one of the 31
  → picks whichever the sort happens to surface
  → reports a confident match to an arbitrary
    paper

The second behaviour is worse than reporting nothing, because it produces an answer that looks settled. A citation left unmatched sends you to look at it; a citation matched to the wrong paper does not, and it is the same class of failure as pairing on surname and year alone — a confident attribution built on a coincidence.

So the position is a guard rather than a limitation. The surname must open the entry; the year must match exactly, because an adjacent year is a different paper; and where two candidate entries fit, the pairing is handed to a person rather than resolved by picking the likelier one.

The version that is hardest to see

There is a particular case worth its own attention: the same work cited under two different authors in one document.

You cite a paper as (Okafor et al., 2021) in chapter two and, three years later, as (Lindqvist et al., 2021) in chapter six — because by then you had met Lindqvist. Both citations refer to the same work; one matches the entry and one does not; and the two citations share no surname at all, so nothing comparing citations to each other can see they are related.

The only place the connection is visible is the entry itself, which lists both names. It is one of the shapes that a duplicate check surfaces and is invisible to everything else.

Three neighbouring mistakes

Position is one of four things a citation can get wrong about authorship, and they are worth distinguishing because only one of them is about position at all.

WRONG PERSON, RIGHT WORK
  (Lindqvist et al., 2021) for an Okafor entry
  → the subject of this post
  → fix the citation

RIGHT PERSON, WRONG SPELLING
  (Al Mansour, 2019) vs "Al-Mansour" in the list
  → the citation and entry disagree on the same
    name
  → make them agree, in the published direction

RIGHT PERSON, WRONG WORK
  (Okafor, 2021) where two Okafor 2021 papers
  exist
  → not an author error at all; an identity one
  → year suffixes fix it

SKIPPED AUTHOR
  (Okafor & Duarte, 2021) for a three-author
  work
  → naming the first and third, omitting the
    second
  → rare, and produces a citation that matches
    nothing under any rule

The fourth is unusual and worth mentioning because it is the one people invent under time pressure — abbreviating a three-author citation by dropping the middle name rather than using "et al.". It reads plausibly and matches nothing, since the entry's second author is not the one the citation names second.

If you are unsure which of the four you have, the entry tells you. Look at how it begins and how many authors it carries, and the answer is usually immediate — which is the general principle in this whole area: the reference list is where the truth is, and the citation is what has to be brought into line with it.

Finding them yourself

Two passes, both quick, and the second is the one that works.

The forward pass tells you something is wrong. Any citation reported as having no matching entry is a candidate — if the work is genuinely in your list, the likeliest explanations are a spelling variant, a year mismatch, or a second-author citation.

The backward pass tells you what. Take the unmatched citation's surname and search your reference list for it anywhere, not just at the start of an entry. If you find it as a co-author, you have your answer in one lookup: the work is listed, and your citation named the wrong person.

That asymmetry is worth noticing. Looking anywhere in the line is a bad rule for a matcher, which must decide, and an excellent technique for a human, who is investigating one specific citation and can look at what comes back.

Edited volumes, where the question is genuinely confusing

One case where citing the "wrong" author is an easy mistake to make honestly, because the entry names two sets of people and only one of them is the author.

The entry
  Duarte, P. (2019). Workflow and staffing. In
  T. Okafor & M. Lindqvist (Eds.), Health systems
  in transition (pp. 88–110). University Press.

Cite as        (Duarte, 2019)
  → Duarte wrote the chapter you read

Not            (Okafor & Lindqvist, 2019)
  → they edited the book, and did not write
    this chapter

The editors open nothing. The chapter author does.

It goes wrong because the editors are frequently the better-known names, and because a reference manager importing the book rather than the chapter will hand you an entry with the editors in the author field. If the work you actually read is a chapter, the entry should begin with whoever wrote it.

What to fix, and what to leave

Fix the citation, not the entry. The entry reflects the published author order and is not yours to rearrange. Change (Lindqvist et al., 2021) to (Okafor et al., 2021), and check whether you did the same thing elsewhere.

Credit the person in prose if you want to. If Lindqvist's contribution is what matters to your argument, say so in words — "Lindqvist's analysis in Okafor et al. (2021)" — which is accurate, readable, and leaves the citation matchable.

And be careful about narrative attribution. Writing "Lindqvist showed that…" when Lindqvist is the third of five authors tells a reader something about who did the work, and it may not be true. Most of the time this is a harmless slip of memory. It is worth being deliberate about, though, because attribution is the one thing a citation exists to get right — and describing someone as the author of a finding they contributed to, rather than led, misdescribes the record in a way that would matter to the people involved.

None of this is a misconduct question and it would be silly to frame it as one: citing the wrong co-author is a memory error with a one-line fix. The reason to take it seriously is smaller and more practical — a citation that names the wrong person is a citation your reader cannot follow, and the whole apparatus exists so that they can.