S. R. Faruqi:

taklīf = to say, tell (to do some work)

An extremely beautiful [Persian] opening-verse of Ghalib's seems to be influenced by Mir:

'There is a fountain of blood, from the heart to the tongue, alas!
I have something to say to you, and no strength to speak, alas!'

Ghalib's verse has, with regard to 'mood', gone beyond Mir's; but with regard to meaning, Mir's verse has a great deal of depth/layeredness. In Mir's second line too, the image is extremely effective and heart-rending.

[The Persian infinitive] āġhashtan can mean (1) to to smear, saturate, pollute; (2) to knead; to braid, weave; (3) to moisten; (4) to defile, sully, immerse. It's clear that the first three meanings are extremely appropriate to the occasion. The speaker has not made clear what these hundreds of utterances are that are wet with blood or are smeared/saturated. He has also not made plain why those utterances have become bloody. The imagination of the reader or hearer is affected and astonished by the uncommon image in the second line; it creates several kinds of possibilities.

But the basic idea is that countless utterances came from the heart to the tongue, and having steadily turned to blood, kept collecting under the tongue. From the grammatical standpoint too, it is very fine that he has called himself 'one who has a hundred utterances under his tongue'; he has not said 'I keep a hundred utterances under my tongue'.

Here, taklīf is in the sense of its Persian idiomatic meaning. In Persian, taklīf is used for a task, or for telling someone to do a task. In Urdu, it's now used in the sense of 'trouble', 'sickness', 'suffering', etc. But the meanings of 'task' or 'telling someone to do a task' are even now concealed in idiomatic usage: for example, āp żarā itnī taklīf kareñ kih falāñ ṣāḥib se bāt kar leñ , or āp ne taklīf kī , us kā shukriyah .

Mir has used it in the Persian idiomatic sense several times. In addition, I've noticed this usage in Qa'im. But since two poets have used it, the suspicion arises that this meaning of taklīf would certainly have been current to some degree in eighteenth-century Urdu. From the first divan:

{7,11}.

Also from the first divan [{358,1}]:

taklīf-e bāġh kin ne kī tujh ḳhvush-dahāñ ke taʾīñ
detā hai āg rang tirā gulsitāñ ke taʾīñ

[who gave you, the fine-mouthed one, the trouble/requirement of going to the garden?
your color/mood sets/gives fire to the garden ]

Qa'im Chandpuri:

taklīf-e nālah kar nah mujhe ā ḳhudā ko mān
kyā jāne kyā ġhaẓab huʾā bhī ek dam ke bīch

[don't give me the trouble/requirement of laments-- come, fear the Lord!
who knows what devastation would occur amidst a single breath!]

In the urdū luġhat of the Taraqqi Urdu Board of Pakistan, one meaning of taklīf as 'incitement, putting in motion' has been entered, with the warrant [sanad] of Mir's {7,11}, and Qa'im's verse, both cited above. Farid Ahmad Barkati has, relying on ānand rāj , correctly written that in Persian the meaning of taklīf is also 'to do a task'; but in the verses under discussion he has erroneously given the meaning of taklīf , because he too has been deceived by the word's usual, present meaning ('trouble, suffering, etc.'). [A further discussion of definitions in various Persian dictionaries.]

This long discussion was necessary in order to make it clear that we must pay attention to every word of the old poets, and especially of the great poets. It would be incorrect to assume that they use every word in the very same meaning that is customary in our time, or that is in our knowledge.

Now please look once more at this line, from the point of view of meaning. Since the speaker is a poet, people have hopes that he will recite some poetry. In reply, the speaker says, 'Don't tell me to to recite anything'. Now the word āh , which is usually more or less conventional, becomes the bearer of a twofold meaning. On the one hand this word is a zila of taklīf , and on the other hand it makes clear the fact that I am ṣad-suḳhan-āġhashtah bah ḳhūñ , therefore I can't recite poetry, I can only sigh. Now the word suḳhan too is two-layered-- it means 'poetry and verse', and also 'speaking'. It's a powerful verse.

This same theme, treated a bit lightly, he has composed in the second divan like this [{877,4}]:

maṣlaḥat hai merī ḳhāmoshī hī meñ ai ham-nafas
lohū ṭapke bāt se jo hoñṭ apne vā karūñ

[there is prudence in only/emphatically my silence, oh confidant,
blood would drip from the utterance, if I would open my lips]

[See also {1010,2}.]

FWP:

SETS == GROTESQUERIE; POETRY
MOTIFS
NAMES
TERMS

Well, I am unable to take the split between the two meanings of taklīf as seriously as SRF does. It seems to me that they have a strong middle ground in the idea of the imposition of something burdensome on the speaker.

The āh is indeed perfect-- it's a sound that expresses one or another kind of taklīf , and also a sound that can be made without moving the lips. Though of course, it's also true that the speaker is saying the words that convey his inability to speak, so that he has to be moving his lips. This paradoxical situation might incline us to take suḳhan as referring particularly to poetry, not to ordinary speech.

The image of a speaker who doesn't want to open his lips, for fear of a gush of blood from beneath his tongue-- does that constitute 'grotesquerie'? His every word is 'mixed' or 'polluted' with blood-- ugh! I know, I know all the reasons it's supposed to fit perfectly into the transgressive abstractness of the ghazal universe. But I still don't care for it.

Note for grammar fans: In the second line, to make the sense work, we need to consider the grammar to be something like, 'I am [in a state of having] a hundred verses mixed with blood under my tongue'.