Argthtjtr

The pronunciation of words spelled with "-ti-" in German seems to be a little complicated. I read in a book that "-ti-" is pronounced as "-zi-" (I believe only when a vowel follows) in words like Aktie or Patient, but that words from French such as Aristokratie may have stressed /tiː/ instead.

It seems to me that it would be simpler to just use spellings like Akzie and Pazient instead. That kind of use of Z has precedent in some other languages, such as Italian ("paziente"). I know that "ce" and "ci" in Latin-based words are often replaced with "ze" and "zi" respectively in German, and I am wondering why the use of the letter Z did not also become usual to represent the same sound in words that have "ti" in Latin. I wonder whether the spelling "ti" has been kept out of some concern that using "zi" would obscure the etymological distinction between words that had -ci- and -ti- in Latin. Did any scholars or spelling reformers ever comment on this topic?

Yes. That is actually a significant part of one of the recent spelling reforms:

Es werden neue Varianten eingeführt: Differenzial, Potenzial, potenziell, substanziell, parallel zu den schon eingebürgerten finanziell, tendenziell;

Neuerungen der deutschen Rechtschreibreform von 1996

Whereas previously the 'correct' forms would have been Differential, Potential…

This is just an ongoing process of customs and bureaucratic negotiations, and resulting in always imperfect and inconsistent outcomes.

By the way, while now it is Aktie, in older books you may find Akzie:

Allgemeiner anzeiger und nationalzeitung der Deutschen, 1818

Background:

I'd say that there is no real rhyme nor reason behind it. Reformers as well as opponents argue with 'rules' to apply (if…then), but those rules are then inconsistent as well. So yeah, has to be memorised, Duden has to be bought in a new edition… lather, rinse, repeat in a few years. You know: a camel is just a horse designed by a commission.

A conservative critic of the reform efforts up to 1997 was Theodor Ickler, who wrote a ghastly attack on plans, reformers, contradictions, inconsistencies, idiosyncrasies of the participants, proponents, rules and results in "Die sogenannte Rechtschreibreform – ein Schildbürgerstreich" (PDF). It's a delightful read.

Given the one example from the question and the reforms applied: one change that is apparently unthinkable, despite ample prior art in that regard would be Nazion. Looks like only a 'special' group of people that is currently still seen as a minority really favours such a change. Given the supposed rules that should govern such a decision for change in orthography, there would be absolutely no obstacle at all. Politically or historically however…

Looking for more comments on the proposed or 'accomplished' reforms, or mere: changes? A quite brilliant analysis of the battles over the 'correct' spelling, in English but covering the German tohu wa-bohu and looking on much more aspects than just -ti-/-zi- are in

Sally A Johnson: "Spelling Trouble? Language, Ideology and the Reform of German Orthography", Multilingual Matters: Clevedon, Buffalo. Toronto, 2005.

The decision to reform German orthography had not been taken lightly. Given that the first and hitherto only set of official guidelines for all the German-speaking countries had been agreed in 1901, the final proposal for their revision in 1996 was the result of almost a century of often heated debate among linguists, politicians, educationalists, lexicographers, writers, journalists, and other interested parties. […]

In this book I propose that the 1996 reform of German orthography, together with the public protests it inspired, constitute a prime example of what Jan Blommaert (1999) has referred to as a language ideological debate.

The 1996 Reform as Language Ideological Debate – returns to an analysis of the disputes surrounding the 1996 reform of German orthography specifically in relation to the concept of a language ideological debate. The discussion focusses on the views of the three main groups of ideological brokers in this dispute: the complainants, the judges of the Constitutional Court, and the linguists, exploring both the differences and similarities between them vis a` vis perceptions and/or rationalisations of the re-standardisation process. Against this backdrop, Chapter 6 – The Trouble with Spelling? Discussion and Conclusions – draws the analysis to a close by considering, in both theoretical and practical terms, possible alternative approaches to the reform and its overall implementation.

Finally, the spelling of foreign loans – a category that, strictly speaking, includes a range of items that are merely perceived as such – has, as we have seen throughout, been a traditional bone of contention in debates over orthographic reform. On the one hand, the retention of loaned spellings tends to favour educated language users familiar with the source language in question. Germanicisation, on the other hand, prioritises those unfamiliar with the source language, while creating two spellings for its users and/or learners. Although these issues are especially relevant to school-aged learners of modern languages such as French or English (for whom double spellings can be a genuine irritant), they are also closely tied in with the question of classical languages such as Greek and Latin, from which German has derived a substantial part of its vocabulary. In the latter case, germanicisation may well be interpreted as a form of ‘dumbing down’ linked to a perceived denial of the classical heritage of German. That said, germanicisation is just as likely to be associated, conversely, with an inappropriate and potentially distasteful concern to emphasise the ‘Germanness’ of the language thereby indexing the political, historical and cultural baggage of German nationalism. As such, it is closely tied to ongoing debates over language loyalty on the part of German speakers (Klein, 2001).

It was in an attempt to reconcile these considerations that the reformers opted to germanicise the spelling of foreign loans only in those cases where a process of integration was perceived to be well advanced in current usage (Table 3.11). However, it was decided that in all cases a system of optional variability should apply, according to which a series of main and subsidiary variants were specified.

The real problem with changing the spelling from the historic (Latinate) "ti" to the phonological "zi" is that you would have to write "Nazion", "nazionalisieren" and the like. Most Germans would think this is a step too far. (I didn't invent this. Brecht wrote it somewhere, but I cannot trace it at the moment.)

EDIT: The source is here (towards the bottom of the page):
https://www.cicero.de/kultur/reform-der-rechtschreibung/44502