Citing two authors with the same surname – is it worth noting that they are distinct people?

Question

In the paper I am currently writing I cite works by Bill Smith and Fred Smith.

I refer to them both several times, as several of Bill Smith’s papers provide the basis for some of the techniques I am using, and Fred Smith’s work provides origin for the dataset I am using. This dataset is commonly called The Smith Corpus.

Currently my paper has a paragraph (boldface mine):

We use the Smith Corpus (Smith 2010), as prepared by Johnson et al (2015).
This corpus is partitioned into test, development and training subsets, and has minor clean-up from the original data collected by Smith (2010).
It is also used by Smith et al (2016) and Otter et al (2016).

Where the all uses of Smith, except the last refer to Fred Smith. Only Smith et al (2016) refers to Bill Smith.

Anyone who checks the references section will realize that these are different people. But the citation style I am using, is a surname–year style. So on a casual reading, one might expect the dataset in question to have been introduced in one of Bill Smith’s papers. However, this is incorrect: It is first used in this area by a third author, John Johnson (whom I also cite), then later by Bill Smith, and Sam Otter.

Bill Smith he refers to the corpus as the Smith Corpus in his paper – without qualification in any way.

Is there anything I should be doing about this possible confusion?
Or can I trust the reader to check the reference list if they want the details on who actually did what?

---

The ~5 page style guide for this conference says

Citations: Citations within the text appear in parentheses as (Gusfield, 1997) or, if the author’s name appears in the text itself, as Gus- field (1997). Using the provided LATEX style, the former is accomplished using \cite and the latter with \shortcite or \newcite. Collapse multiple citations as in (Gusfield, 1997; Aho and Ullman, 1972); this is accomplished with the provided style using commas within the \cite command, e.g., \cite{Gusfield:97,Aho:72}. Append lowercase letters to the year in cases of ambiguities. Treat double authors as in (Aho and Ullman, 1972), but write as in (Chandra et al., 1981) when more than two authors are involved.
...
We will reject without review any papers that do not follow the official style guidelines, anonymity conditions and page limits.

This level of guidance is typical for CS conferences in my field.

While I can of-course decide to make up my own rules for what to do in this case that is not covered clearly by their style guide -- and they probably won't desk-reject for it -- what additional rules should I use (if any)?

Answer 1

Check your style guide. Many surname-year styles, such as APA, have provisions for exactly this problem.

But do you know how to proceed when a reference list includes publications by two or more different primary authors with the same surname? When this occurs, include the lead author’s initials in all text citations, even if the year of publication differs (see the sixth edition of the Publication Manual, p. 176). Including the initials helps the reader avoid confusion within the text and locate the entry in the reference list.

http://blog.apastyle.org/apastyle/2014/01/when-to-use-author-initials-for-text-citations.html

In your case, this would lead to:

We use the Smith corpus (F. Smith 2010), as prepared by Johnson et al. (2015). This corpus is partitioned into test, development and training subsets, and has minor clean-up from the original data collected by F. Smith (2010).

It is also uses by B. Smith et al (2016), and Otter et al (2016).

Note that since "Smith corpus" is a name of a corpus, rather than a citation, it can remain without the initial.

Also note that in APA style, all citations to any Smith papers have to have the initial. So even if this is the only time you mention B. Smith et al 2016, all of your references to F. Smith 2010 have to have the initial. This is a little annoying but nicely unambiguous.

Answer 2

I would go with:

We use the Smith Corpus (Smith 2010), as prepared by Johnson et al (2015). This corpus is partitioned into test, development and training subsets, and has minor clean-up from the original data collected by Fred Smith (2010). It is also uses by Bill Smith et al (2016), and Otter et al (2016).

'Smith Corpus' is the name of a dataset, not a person, and is to be found in the paper referenced as 'Smith 2010'. The other two instances of 'Smith' are names, not references. The reference is in brackets afterwards, and is made up of just a date because the name part of the reference is not required when the authors have just been named in the sentence.

If this feels a bit too odd, you could add in 'Sam'.

Answer 3

The way to handle this is to refer to F. Smith and B. Smith explicitly in the text. The editors will probably not like it but you can point out to them that adding the first initial removes any confusion about which author you are referring to.

I had so deal with a similar situation many years ago, and after clarifications the editors let me use the first initials (indeed I used both the first and the middle initial) even if it was against the "house rules".