Citations are used in papers for various purposes, including giving credit to past papers for career points, providing evidence and support for the citing author, claiming that the work is grounded in or contributes to existing literature, and referring readers to previous useful work. However, this can confuse the scientific record, especially when authors do not clearly identify the purpose of their citations and what they argue the cited papers are claiming.
One proposal to address this issue is to include coded tags with different categories in citations, such as tags for claims, evidence, related sources of numbers, and disagreements with the results or methods of the cited papers. For example, a paper may make a claim such as “Dogs are often seen as man’s best friend” (claim: Fido et al, 2022). Another paper may provide evidence of some type, such as “Dogs often help wounded humans” (survey evidence: Spot et al, 1975; lab experiments: Rex, 1990). A related source of numbers might be “Dogs helped 10,342 people in 1988” (figure: Bowzer and Fifi, 2018). And a paper may express disagreement with the results or methods of another paper, such as “While Brutus et al (1997 - disagreement) use whispering to ask dogs, we argue, and provide evidence that shouting at a dog is more likely to elicit their true beliefs.”
Has this idea been implemented? When I posted this on Twitter, some agreed this would improve the clarity and transparency of the scientific record. One suggested that the purpose of the citation should be included in the text of the paper, while another proposed that the tags could be included as hover-over pop-ups.
I suggested that we could trial this system at The Unjournal, although this journal would only apply to the evaluations and responses, rather than the original research.
This post came from chucking this Twitter thread into ChatGpt and giving it some TLC.