This coming week we begin the three-week countdown to Tish’ah BeAv, commemorating the three weeks between when the Romans broke through the walls of Jerusalem on 17 Tammuz in the year 70 CE (which we believe was the year 3830) and Tish’ah BeAv (the 9th of Av that same year) when they destroyed the Second Temple. On 17 Tammuz we fast from dawn to dusk - no food, no water. During the following three weeks many refrain from going to concerts and public celebrations, and some don’t even take a haircut. And in shul, we read the three haftarot of rebuke.
After Tish’ah BeAv, we read seven haftarot of consolation (though in the opinion of your correspondent the first few sound much more like rebuke and over the seven they progress toward consolation).
Now to the source of the question. Should one of the Shabbatot on which we read a haftarah of rebuke fall on Rosh Ḥodesh (the first of the month, which happens this year for the second haftarah of rebuke), we still chant the haftarah of rebuke. Sefardim actually insert a reminder that it is Rosh Ḥodesh by adding both the beginning and the ending lines of the haftarah they are not reciting at the end of the haftarah of rebuke, but Ashkenazim (which is our model) do not do so.
However, when it comes to the haftarot of consolation, if it is Rosh Ḥodesh we do skip the consolation for a week, we chant the Rosh Ḥodesh haftarah, and then we add the third haftarah of consolation to the end of the fifth (which makes sense, given that they follow each other in the book of Isaiah, and also the haftarah for Ki Teitze, the fifth, is very short at only ten lines). Thus all get read, but deference is paid to the Rosh Ḥodesh haftarah as well. (For erev Rosh Ḥodesh, when the new month begins on Saturday evening rather than Friday evening, there is also a special haftarah, but we do not substitute that one for a haftarah of consolation. The Sefardim do add the first and last lines to the end, though.)
Why? Why are we inconsistent? It seems discussion of this occurred sometime in the late 1300s to early 1400s, and the recognition that the Rosh Ḥodesh haftarah also contains words of consolation led to rabbis deeming it appropriate for that period, thus allowing them to recognize both rules for haftarot - the rule about Rosh Ḥodesh haftarah (set forth in the Gemara, Megillah 31a) and the rule about consolation (set forth in Gemara, Megillah 29b-31a) (but not for rebuke). To be complete, they combined the two aforementioned haftarot for one week. There is, they decided, no consolation contained in the haftarah mahar hodesh - the one for erev Rosh Ḥodesh.
These rulings are apparently codified in several locations, including the Mishnah Berurah and Tukachinsky Calendar. The Sefardim follow the Shulhan Arukh’s prescribed rule of never pushing away any of the haftarot of consolation.
In other words, if two rules can be satisfied, it should be done, rabbis said, rather than having to justify making a choice between them. And yet we still get our full measure of rebuke meted out in the coming three weeks, in a timely way.