Reading through the pesukim discussing shemitat kesafim in parashat Re'eh, we encounter both a mitzvat esei and a lo ta'aseh. The parasha begins with the instruction to enact shemitta every seven years. Though this pasuk doesn't specify shemitat kesafim, the ensuing pesukim clearly designate this parasha as focusing on the financial, rather than agricultural, elements of shemitta. The next pasuk declares, "These are the terms of shemitta: each creditor should 'rest' (shamot) his hand; he should not demand [money] from his friend since the year has been declared a shemitta year for Hashem." Clearly a lo ta'ase exists in demanding debts which have been suspended due to shemitta. What, however, are we to make of the positive command to 'rest' our hand? If the money is un-collectable anyway, what positive mitzva is actually performed?
I Defining the Positive Commandment
The Rambam (both in sefer ha-mitzvot positive commandment 141, as well as in the beginning of the ninth perek of shemitta) lists a positive mitzva. As noted, since the debt is automatically canceled during shemitta, leaving the lender with little choice (in terms of demanding the money), what, if any action, does this mitzva mandate? Can we claim that even though his demand will be ineffective he is still commanded to abstain from that 'ineffective' claim?
An extreme position is presented by the Yereim in siman 278. He claims that after shvi'it the borrower does not have the right to withhold the lender's money against his will. He must invite the malveh (lender) to beit din offering him the opportunity to cancel his debt by saying 'meshamet ani' (I forfeit my rights). If the malveh refuses to cancel his debt, then beit din can force him to (as they can enforce the performance of any positive mitzva). This extreme position suggests that indeed debts during shvi'it are not automatically canceled but remain intact until the malveh actively cancels them. According to this view, the malveh plays an active role in their cancellation and a positive commandment is feasible. This opinion is not expressed by any other of the Rishonim and severely contradicts many gemarot which suggest almost explicitly that shemitta automatically annuls debts. It is unlikely that the Rambam adopted this position. See especially the Or Zarua (masekhet Avoda Zara siman 108) who disputes the position of the Yereim.
The Yereim undoubtedly drew his position from the mishna in Shevi'it 10:8. The mishna states that if a lender seeks to return his debt during shevi'it (meaning AFTER shevi'it, when the cancellation officially occurs), the malveh should respond by saying "I cancel your debt." If the lender persists and insists on payment and says, "Even though you are canceling, I still wish to compensate you," the malveh may then accept payment. The primary chiddush of this gemara is that the malveh may ultimately receive money if the loveh volunteers the payment. Evidently, the Yereim inferred both a direct mitzvat esei to cancel the loan AND the viability of the loan if this statement is not made.
The Or Zarua, after disputing the Yereim, comments, "If the malveh does not want to utter, 'I cancel the loan,' we can force him to, just as we enforce all positive commandments." Even though he argues with the Yereim and claims that the debt is canceled, he still recognizes an esei to formally annul the loan. This is a unique situation in which there exists a commandment to utter a statement which has no monetary effect (since the loan was automatically canceled). Possibly, the Rambam subscribed to this view and counted a positive commandment even though the loan is canceled independent of the malveh's statements.
II Shemitat Kesafim In Our Day
The gemara (both in Gittin 36 as well as Mo'ed Katan 2b-3a) establishes a linkage between shemitat kesafim and 'shemitat karka': only when shemitat karka applies does shemitat kesafim apply. The Rambam perceives the linkage as one between shemitat kesafim and yovel (designated as "shemitat karka" since the lands are returned just as loans are canceled). This same opinion is quoted by Tosafot in Gittin. According to this reading, shemitat kesafim in our era only applies mi-derabanan. Yovel no longer applies since a condition for yovel is "kol yoshveha aleha," that all the Jewish people must be inhabiting Israel. As yovel no longer applies, neither does shemitat kesafim - on a de-oraita level. However, as the gemara in Gittin clearly states, the Rabanan made a takana (based on their powers of "hefker beit din hefker") that shemitat kesafim should apply. This is the position of the Rambam, the Ramban and most Rishonim.
There are, however, two notable exceptions to this position. The Ittur ruled that shemitat kesafim applies mi-de-oraita. In his comments (see chelek 1 ot prusbul) he develops the notion that:
a) linkage exists between agricultural shemitta and shemitat kesafim, but not necessarily between yovel and shemitta kesafim;
b) agricultural shemitta (and hence shemitat kesafim) can occur independently of yovel and hence can be d'oraita even when yovel no longer applies. This is the position of the Rabanan, who argue on Rebbi's linkage between yovel and shemitta (see Gittin 36a and Mo'ed Katan 2b);
c) once agricultural shemitta applies d'oraita, the same can be said about shemitat kesafim.
This position is a minority opinion, rejected by most of the Rishonim.
The Ra'avad (in his commentary to the Ri"f in Gittin) raises a different opinion – that shemitat kesafim doesn't apply at all and is only an option for pious people, rather than a normative Rabbinic decree. He believes that shemitat kesafim is completely dependent upon yovel, which has been completely suspended. Hence, no normative shemitat kesafim exists - even on a de-rabanan level. The Ramban (in his sefer ha-zechut) soundly rejects this position of the Ra'avad, and indeed the simple reading of the gemara in Gittin suggests a Rabbinic decree regarding shemitat kesafim even after yovel was suspended.
Historically, there were many periods (especially during the medieval ages) in which shemitat kesafim was completely ignored - even prusbulim weren't written. The Shulchan Arukh (Choshen Mishpat siman 67) cites the dominant position that shemitat kesafim applies at a de-rabanan level. The Rema notes the dissenting opinion of the Ra'avad that it doesn't apply at all and suggests this as the basis for the halakha having fallen into disregard. In fact, many Rishonim (the Rosh, the Maharik, and the Terumat Ha-deshen) authored teshuvot probing this phenomenon of non-shemitat kesafim observance – one which apparently was quite prevalent. The Rosh made persuasive efforts to reinstate shemitat kesafim, evidently not too successfully.