The asiddha section
Most of the Aṣṭādhyāyī's sandhi rules are an an unusual section of the text. To understand that section, we must first understand how the Pāṇinian system works at a high level. We'll then discuss a major problem with its approach and how Pāṇini solves it.
When we use the Aṣṭādhyāyī, we start with an incomplete expression. We then apply one rule at a time. The output of one rule is the input to the next. And as we keep applying rules, our result gets closer and closer to a valid Sanskrit expression.
This full process, including the rules we apply and the results we get, is called a prakriyā (“procedure”, “derivation”). When we use the Aṣṭādhyāyī, we must know not only what rules mean but also how to apply them to generate a correct prakriyā.
What do we mean by a “correct” prakriyā? At each step in the prakriyā, multiple rules could apply, and we must choose which one to use. There are a few basic principles that help us here. (For example, we should prefer more specific rules to less specific rules.) And when we can no longer apply any rules, the prakriyā is complete.
Finishing a prakriyā
Let's focus on the phrase “when we can no longer apply any rules.” What does this mean? For example, we might have this incomplete expression that has had no sandhi rules applied:
If we use rule 6.1.78 from the previous lesson:
एचो ऽयवायावः। ६.१.७८
eco 'yavāyāvaḥ (6.1.78)
An ec vowel becomes ay, av, āy, or āv, respectively [when a vowel follows in saṃhitā].
then we can create a new result:
ते इच्छन्ति → तय् इच्छन्ति
te icchanti → tay icchanti
But if you know Sanskrit, you know that this isn't the typical result. The e at the end of a word usually becomes a when it is followed by a vowel. There is a vidhi rule that makes the appropriate change. And by applying that rule, we get the correct result:
तय् इच्छन्ति → त इच्छन्ति
tay icchanti → ta icchanti
A serious problem
The problem is that the prakriyā is not actually complete, because another rule can now be applied. Specifically, it's rule 6.1.87, which we saw in the previous lesson:
a [and the following vowel] become [a single] guṇa [in saṃhitā].
6.1.87 can apply, and there is no other rule that takes priority over it. So it must and will apply, which gives us a bad result:
* त इच्छन्ति → तेच्छन्ति
* ta icchanti → tecchanti
More abstractly, the problem is as follows. Suppose we have a good rule
GOOD that will give us a good result and a bad rule
BAD that will give
us a bad one.
GOOD matches the input, so we are able to apply it.
BAD does not match the input, so we cannot apply it.
We want to apply rule
GOOD so that we can get a good result. But if we
do, then rule
BAD might match the output of
GOOD. Then we might
be forced to apply
BAD, which will give us a bad result. This is what
happened in our example above.
We need to ensure that once
GOOD is applied,
BAD will never
be applied. If we can do that, then we can solve this serious problem.
Pāṇini's solution to this problem is to offer this rule:
… is asiddha (inert) in the previous [area].
Rule 8.2.1 is an adhikāra that lasts until the very end of the Aṣṭādhyāyī. So, all rules that follow rule 8.2.1 will be in its scope. But what does rule 8.2.1 mean, and what does it do?
Understanding rule 8.2.1
First, let's understand the two words pūrvatra and asiddham.
pūrvatra literally means “in the previous (area).” Here, it refers to all previous rules in the grammar. Every rule after 8.2.1 will inherit the word pūrvatra. So for each of these rules, every rule before it is pūrvatra.
asiddham literally means “not accomplished” or “not enacted.” Here, it essentially means that the rule cannot be used.
We can update our translation like so:
… is disabled with respect to prior rules.
What does this mean? Suppose we have our two rules
as before. If we put
GOOD after rule 8.2.1, then it is “disabled with
respect to prior rules.” So, no prior rule can use the result of
BAD is a prior rule, then it cannot use the result of
and will not be able to apply. Our problem is solved.
But rule 8.2.1 also has an interesting implication. Suppose that
BAD are both listed after rule 8.2.1. Does the basic idea still
hold? Yes: If
BAD comes before
GOOD, then once we apply
GOOD, we cannot go back and apply
BAD to the result of
problem is still solved.
The practical meaning of rule 8.2.1
Rules after 8.2.1 should be applied in order, and they should be applied after all other rules. This means that all of our prakriyās should have this structure:
First, we apply rules before 8.2.1.
Then, we apply rules after 8.2.1. But these rules must be applied in order; we cannot go back and apply an earlier rule.
To return to our example above, the vidhi rule that changes te to ta is after rule 8.2.1. So once we apply it, we cannot go back and apply 6.1.78 (ād guṇaḥ). We thus get the result we were aiming for:
ते इच्छन्ति → त इच्छन्ति
te icchanti → ta icchanti
and our prakriyā is complete.