# Better handling of multiple publishers

**URL:** <https://discourse.citationstyles.org/t/better-handling-of-multiple-publishers/1571>\
**Category:** CSL Evolution\
**Created:** [October 26, 2019, 9:59pm UTC](https://discourse.citationstyles.org/t/better-handling-of-multiple-publishers/1571 "2019-10-26T21:59:56Z")\
**Posts on this page:** 4\
**Page:** 1

<div class="post-metadata">

**Author:** ![Denis\_Maier](https://avatars.discourse-cdn.com/v4/letter/d/8491ac/32.png) [@Denis\_Maier](https://discourse.citationstyles.org/u/Denis_Maier)\
**Post date:** [October 26, 2019, 9:59pm UTC](https://discourse.citationstyles.org/t/better-handling-of-multiple-publishers/1571/1 "2019-10-26T21:59:56Z")

</div>

This is probably more for CSL 1.2 than for CSL 1.1, but anyway: The way multiple publishers are dealt with is not ideal. SBL Handbook of Style, p. 83 requires this formatting `Gottingen: Vandenhoeck & Ruprecht; Fribourg: Presses Universitaires`, which is not possible with standard CSL. (CSL-M includes a [“hack”](https://citeproc-js.readthedocs.io/en/latest/csl-m/#subgroup-delimiter-subgroup-delimiter-precedes-last-and-extension), to quote @Frank_Bennett here.) Perhaps this requires a deeper change in the data model, like treating `publisher` and `publisher-place` as pairs, something like name parts.

---

<div class="post-metadata">

**Author:** ![dstark](https://yyz1.discourse-cdn.com/flex029/user_avatar/discourse.citationstyles.org/dstark/32/296_2.png) [@dstark](https://discourse.citationstyles.org/u/dstark)\
**Post date:** [April 14, 2025, 8:19pm UTC](https://discourse.citationstyles.org/t/better-handling-of-multiple-publishers/1571/2 "2025-04-14T20:19:51Z")

</div>

For this scenario, could we not use

- an empty `original-date` variable and
- the place and name of the publisher to be listed first in `original-publisher` and `original-publisher-place`?

See the example and linked pull request in [this post](https://www.jdavidstark.com/how-to-properly-handle-facts-of-publication-with-zotero/#extra).

---

<div class="post-metadata">

**Author:** ![Denis\_Maier](https://avatars.discourse-cdn.com/v4/letter/d/8491ac/32.png) [@Denis\_Maier](https://discourse.citationstyles.org/u/Denis_Maier)\
**Post date:** [April 15, 2025, 12:19pm UTC](https://discourse.citationstyles.org/t/better-handling-of-multiple-publishers/1571/3 "2025-04-15T12:19:06Z")

</div>

Nice. I guess that would work, but I still think `publisher` should probably be a name variable consisting of `name` and `place`. But in the meantime your solution looks very good.

---

<div class="post-metadata">

**Author:** ![bwiernik](https://yyz1.discourse-cdn.com/flex029/user_avatar/discourse.citationstyles.org/bwiernik/32/438_2.png) [@bwiernik](https://discourse.citationstyles.org/u/bwiernik)\
**Post date:** [April 16, 2025, 3:02pm UTC](https://discourse.citationstyles.org/t/better-handling-of-multiple-publishers/1571/4 "2025-04-16T15:02:02Z")

</div>

For a personal workaround, `original-publisher` could work, but I’d generally like to avoid hacks like this in styles in the central CSL repository. One of CSL’s strengths over BibTeX is avoiding style-specific data entry. This workaround works for the case of 2 publishers, but doesn’t extend to more publishers.

I’ll also note that citeproc-js does have capability built in to separate publisher and publisher-place fields on a semicolon and re-join them if appropriate attributes from CSLm are set.

The best solution would be to get `publisher` as a name variable incorporated into the spec, processors, and clients. It would be worth aligning with processor maintainers on prioritizing for CSL 1.1 feature implementation.