7:
Our mishnah says that one has not fulfilled one's duty if the Shema is recited "out of order". What this means is that if one did not read the paragraphs in the accepted order - first Shema, then Vehaya and lastly Vayedabber. (We should recall that until well into the Middle Ages all the liturgy was recited by heart, and so mistakes such as this could occur.)

8:
Our mishnah also says that if one one makes a mistake one must return the place where the mistake occurred. The logical connection between the last two halakhot of the mishnah is explained by by last explanation, since they are both concerned with errors caused by reciting "by heart". We tend to think of errors in terms of mispronunciation, but this is not the subject of our mishnah. The term "error" here indicates, rather, "getting lost": what does one do if during recital by heart one forgets where one got up to? The Gemara [Berakhot 16a] gives the following indications: If you know you were in the middle of a certain paragraph, start that paragraph again. If you were between paragraphs (and you can't recall which) start all over again!

One particular problem (which can still dog even seasoned reciters!) is the fact that certain phrases recur in different paragraphs. The Gemara says that if you get stuck on the phrase And you shall write them on the doorposts etc (which occurs at the end of both the first and the second paragraphs - go back to the beginning of the first. However, if you already continued with the next phrase (in the second) you can assume that your seasoned tongue did not lead you astray!

Concerning "letting the ear hear what the lips pronounce", Ed Frankel sends the following comment:

Not only do we have this in the idea of whether Shema is to be physically heard, or learned and understood, we also have the same idea in the idea of how it is to be done. We refer to the passages as Keriat Shema: is this simply the "reading of Shema", which could be done with one's eyes alone, or "declaration of Shema", which at minimum requires moving of the lips if not some volume.

---

Jeffrey Smith has the following question which may be of more than just personal interest:

Over the weekend, I was questioned about the prayers collected under the rubric of "Bedtime Shema". Can you explain how far back they go, and just how "authoritative" the present text is? Not of course, the Shema itself, but the blessings and other recitations that accompany it in most siddurim.

I respond:

We have mentioned this a few times, in Berakhot 005 in particular. The source for the text of Keriat Shema al ha-Mittah ["Bedtime Shema"] is found in the Gemara [Berakhot 60b]. There it is written that "When one gets into bed to sleep one recites the first paragraph of the Shema and then says..." There then follows a text of the berakhah known as ha-mapil. This is all that there is there. In Talmudic times Babylonian Jewry in general was very superstitious, believing in devils and imps and other unseen malevolences with gusto. The religious leadership also took these matters very seriously. Nightime had its special terrors, and so other texts were gradually added to the "Bedtime Shema", particularly Psalm 91, which was considered to be a charm against harm (see Gemara Shevu'ot 15b, where it is called Shir shel Pega'im [The accidents psalm]. Another change was to move the recitation of the Shema itself from the beginning of the ritual to the end. This was based on information given in the Talmud of Eretz-Israel [Berakhot 1:1] that Rabbi Zeira would recite the Shema over and over until he fell asleep. (Well, its better than counting sheep, isn't it?)

---

Yiftah Shapir comments on the disqualification of a cantor who cannot pronounce Hebrew 'properly':

It is interesting to note that Yerushalmi (Berakhot, daf 16, actually the Gemara about the mishna we're discussing) mentions that the inhabitants of Haifa, Beth-Shean and Tivon should not be allowed to act as cantors because they do not pronounce Ayin and Ĥet (and Heh, too) properly. That's an interesting indication for differences in regional dialects of Hebrew within Eretz Yisrael - even then (4th century?), and an indication that some problems are perennial..

Yiftah is correct in everything he writes except his source: it is the Bavli (not Yerushalmi) and it is tractate Megillah 24b. (By the way: there is no Gemara such as Yerushalmi 16, since the tractate ends at 14d.)

---