To the great chagrin of the frum community, it has recently been discovered that many of the sifrei Torah coming to America from Eretz Yisrael have serious pesulim. No matter how much one has paid for the sefer Torah or whether it was computer-checked before its arrival, many have been found to have missing letters and other serious deficiencies.
To understand the underlying reasons for this problem and to explore what can be done to rectify it, I spoke to the spokesperson of the largest international organization that has been established to ensure the halachic integrity of sifrei Torah, tefillin and mezuzos, Vaad Mishmeres Stam. Founded in 1976 by Rabbis Dovid Leib Greenfield and Shmuel Eliyahu Granatstein, it has gained the support of prominent rabbanim, including Rav Shmuel Halevi Wosner, and has partnered with the Eidah Hacharedis.
Some kehillos have reported that most if not all of the sifrei Torah they sent out to be checked were pasul. Is that accurate?
In recent years, a lot of sifrei Torah have been found to have extra or missing letters. There are also other types of pesulim, such as when one letter has been substituted for another. That’s a pesul that no one can debate because it’s black and white. We’re finding a lot of those, and the reason seems to be that there’s a problem with the computer check.
What’s happening with the sofrim who are writing them?
There’s almost no such thing as a sofer who can produce a sefer Torah without a single error. To remedy that, around 45 years ago a computer program was created to find added, missing or substituted letters. Of the first 100 sifrei Torah that were checked, only 18 were found to be fully kosher. Since then, another computer check was devised to not only check the accuracy of the letters but also whether or not they are written according to halachah. Once this program came into existence, not only were there far more problems coming up on the screen, but the proprietors also sold the software to anyone who wanted to make money, who was then able to claim that he was qualified even if he didn’t have a basic understanding of Hilchos Stam. The end result was that there are now many sifrei Torah in use that have pesulim that were never found.
Can you explain the problem with the newer computer check in more detail?
Of course. When the first computer check was invented, all of the gedolim set three conditions: it could only be used to find extra, missing or substituted letters; it could only be used in addition to a regular check by a qualified sofer; and it could only be used by an organization that specializes in Hilchos Stam. Five years later, the second computer program was created, and it was sold to laymen. It also looked for more than only the original problems, and no one was responsible for ensuring that there would be a regular check afterward. I would say that over the past 30 years, some 80% of Stam is being checked by that new program, and it was sold to between 1,500 and 2,000 individuals. Of those, maybe 2% to 3% really know how to use it properly.
What else does the newer program do?
It checks the kashrus of each individual letter: Is the beis rounded or squared? Is the chof rounded or squared? It also checks if any letters are touching or if there are any other possible disqualifications.
You said that the rabbanim set three conditions. Who were those rabbanim?
The Badatz of the Eidah Hacharedis put out a kol korei that included the signature of Rav Ovadia Yosef. I had never seen his signature on a kol korei from the Eidah Hacharedis before. Along with the members of the Badatz, there were signatures from Rav Shlomo Zalman Auerbach, Rav Nissim Karelitz, the Shevet Halevi and Rav Elyashiv, among others.
What’s wrong with being mehadrin min hamehadrin and using the program to pick up every kind of pesul?
First and foremost, a computer can’t pasken on the halachic definition of a letter.
Yes, but it can at least flag potential problems.
A computer will never be able to catch 10% of the issues that are really going to be in a sefer Torah.
So the problem isn’t that they’re using the program but that they’re using it instead of a human magiah.
That’s one part of it. The other part is that it might create a warning that letters seem to be touching when there’s really a letter missing. If the program doesn’t get that right, the person who is double checking will see that they aren’t touching and will just put an X on the flagged note—not noticing that the letter is missing. I have hundreds of pictures of missing letters that were flagged by the computer as “seemingly touching letters.”
In other words, the added checks have created problems because there’s confusion between the issues.
There’s also a third aspect to the problem, which is the sheer number of things being flagged. Every amud can have 1,000 potential problems that the computer thinks might be there. Someone should really go through each and every one of them, but they aren’t going to actually do that, so they’ll just look at it quickly to see if there are any added, missing or substituted letters, which are the most important components of a computer check.
To read more, subscribe to Ami
Recent Comments