Below are notes I made from an advanced class on Arabic grammar taught by Shaykh ‘Abid Husayn in Madrassah Imam al-Hadi (a) in Qom in 2020, based on the doctorate thesis al-Uṣūl al-Naḥwīyyah wa al-Ṣarfīyyah by Dr. Muhammad Abdullah Qasim, which is an analysis on the book al-Ḥujjah of Abū ‘Alī al-Fārisī (d. 377). Although the notes are incomplete as I was unable to continue the lessons, I am sharing what I have, as there is still ample material in it that can be of benefit to those interested in the subject. Furthermore, as these notes are for advanced students, many Arabic terms and phrases have been left untranslated.

—

1. What do we mean by the term al-‘Ulūm al-Lughawīyyah? The disciplines that study language from different perspectives are called linguistics. What is the difference between Fiqh al-Lugha, or Falsafa al-Lugha? Are they the same thing or inherently different? These are questions that we need to determine foremost. Jurjānī – the founder of the science of balāgha – for example puts forth the Naẓarīyyah al-Naẓm, even though this discussion has a lot of connection with naḥw.

2. ‘Ulūm are either ḥaqīqīyyah and i’tibārīyyah – linguistics are second, then what does i‘tibār mean? I’tibār does not mean a mirage, or pseudo imagination – rather it means ja‘l or a convention by the rational people of society, and this convention has has real effects. For example, the concept of milkīyyah (ownership) is an i’tibār, has no instance in extramental reality, but it has a very real effect.

3. Why are linguistics called i’tibārī? This is because the relationship between dāl and madlūl – lafẓ and ma‘na – is waḍ’īyyah and ja’līyyah. In this case, linguistics are divided into three: 1) dāl (lafẓ), 2) madlūl (ma‘na), 3) form of language (for example, morphology).

4. What is the definition of lugha? Lugha is from lagha-yalghū and laghw is something which has no benefit. Although some say it is taken from the sound of a bird which would then imply it is a word referring to a sound. Some say it means inḥirāf, in which case it would mean it is referring to a dialect (like lugha of Tamīm, lugha of Quraysh – this is how it was used in earlier works, not lihja). In the Qurān it is also used in the latter meaning, and the technical meaning of lugha today is not used in the Qurān at all. Rather the Qurān uses the word lisān for it – lisān qawmihi.

Some even say the word lugha is not even an Arabic word, rather it is originally Greek. Others say this is a weak opinion since the pattern for this word already existed in Arabic and so did the root-word. However, there is room for research here, since we have other words like this too. Raḥmān is from Raḥmah and the pattern for it also exists in Arabic, but yet some scholars believed Raḥmān is not an Arabic word, rather it is Hebrew. In other words, Raḥman is an ‘alam (proper noun), and hence it is ‘atf bayān for Allah in the basmalah, while al-Rahīm is na‘t.

The technical definition of lugha by Ibn Jinī in al-Khaṣā’iṣ is one of the best definitions:

حد اللغة أصوات يعبر بها كل قوم عن أغراضهم

The definition of Ibn Jinī was copied by many later scholars, and many argued against it, commented on different words of the definition. Ferdinand de Saussure (d. 1913) a Western scholar in his book Course in General Linguistics also gives a decent definition as follows:
Language is a well-defined homogeneous object in the heterogeneous mass of speech facts.”

Stages of Language

1) Language is essentially sound, and the study of sound can also be conducted from two perspectives:

i) sound qua sound (where does the sound come from and how it is uttered – makhārij – this is called phonetics / tajwīd), and

ii) sound qua other sounds (Western linguistics has spent a lot of time here – study on vowels, consonants, syllables / maqāṭi‘).

2) When these maqāṭi‘ are placed together you get words – this is called takwīn al-kalimah – and the study of it is called morphology (‘ilm al-ṣarf).

3) When these words are put together, you get sentences – this is called naḥw / syntax.

4) Then eventually you get into ‘ilm al-dilālah (semantics).

These are not the only disciplines, rather aforementioned disciplines are only related to the stages of the development of language. There are other disciplines like ‘ilm al-lugha, fiqh al-lugha, ‘ilm al-lisānīyāt, falsafa al-lugha etc. Are these all the same discipline? Since we are talking about i‘tibārīyāt, differences between disciplines means that in most cases they are different issues being addressed – it is not a mūjibah kullīyyah, rather it is aghlabīyyah. However, others say there is no difference, they all mean the same thing. We also have a discipline called ‘ilm al-mu‘jam which is a different study altogether.

‘Ilm al-Lisānīyyāt studies language qua language (whether it is Arabic, English etc.) – it is the study of human language and how it relates to societies, communities etc. In this field, it is often concluded that all human languages can go back to a few general original languages – Indo-European (includes English), Sino-Tibetan (Includes Chinese), and Afro-Asiatic (Includes Arabic).

‘Ilm al-Lugha on the other hand is divided into ‘ām (general) and khāṣ (specific)- ‘ām is exactly what we mentioned in ‘ilm al-lisānīyyāt. In lisānīyyāt we may say languages are broken down into Indo-European (Includes English), Sino-Tibetan (Includes Chinese), Afro-Asiatic (Includes Arabic) etc. However, when we want to study Arabic, we have to figure out which of these 3 original languages it falls into and how it works. In this context we see conflict between old and new naḥw – we see discussions on tajdīd al-naḥw (renewal of syntax), where they rely on istiqrā’ī (inductive) and descriptive approaches, unlike classical naḥw which goes into the study of ‘awāmil (word governance).

What is the definition of Fiqh al-Lugha? Fiqh is a precise and deep understanding of a subject matter and it was used to understand religious with precision. Lugha could possibly mean:

a) Dirāsah Mu‘jamīyya: Dictionary words and what those words mean. This is then very different to ‘ilm al-lisānīyyāṭ.

b) General meaning of lugha – language – words, sentences etc. A systematic study of this, in which case it is very similar to the specific (khāṣ) ‘ilm al-lisānīyyāt.

What is the definition of Falsafa al-Lugha? The West and the East had initially accepted Aristotelean philosophy which is based on burhān, however eventually the West began to concentrate on epistemology and human perception. The East did not go down this path; the West began to doubt certainty of human knowledge, in which case they began to question the value of human knowledge (philosophy of sciences – for example jurisprudence, ethics, religion etc.). In philosophy of language we discuss the nature of language, its reality, how its created, what are its objectives etc.

Principles of Language

When we study principles of language, it implies these principles are meant to be general and universal principles, as opposed to particular rules. The Arabic language has an apparent existence, such as the vowels, the sounds, or compositional and these are the particularities of the Arabic language.  However, just as we had an anti-philosophy and anti-logic trend amongst Muslim scholars, who believed these rules of logic and philosophy are man-made and then superimposed onto the Qurān; likewise we had an anti-grammar trend, as seen in the work of Ibn Muḍā and contemporaries like Shawqī Ḍayf who believed these grammatical rules were made by grammarians and later applied onto the Qurān. However, we say that a logician is not making up rules, rather they are identifying them and are conscious of these rules that humans are using and applying anyways. We say the same thing about the grammarians and linguists – they are doing kashf (uncovering) of these rules, not making them up. The question that still remains though, is what tools and methods were linguists using to uncover these rules? There are 4 tools:

a) Samā’ (Hearing)

b) Qiyās (Analogical Reasoning)

c) Istiṣḥāb (Principle of Continuity)

d) ‘Aql (Intellect)

The reason why we differentiate between the Kufan and Basran school of thought in grammar is not because they had some particular differences. Those kind of differences even exist amongst scholars of the same school, rather the difference between the Kufan and Basran school was in their methodology and approaches.

‘Ilm al-Naḥw

1) Al-Aḥkām al-Naḥwīyyah: Just like Naḥw was influenced by logic, it was also impacted by Fiqh. For example, a lot of syntatic discussions revolve around the concepts of aṣl and far‘ (taken from discussions on qiyās). You will find terms such as “the aṣl in a certain syntax is so and so”, and “the far‘ is so and so.” This is mostly taken from Sunni Fiqh. A second impact of Fiqh on Naḥw is that scholars derived rulings for it as well, such as what is wājib, mumtani‘, jā’iz etc. They will say, “It is wājib to prioritize fā‘il in such and such cases,” or “In some cases it is jā’iz” – this is the language of Fiqh.

Naḥw was a skill – something practical, but then eventually it become an organized discipline (‘ilm) especially due to the word governance theory of ‘āmil and ma‘mūl. Some classical grammarians – like Ibn Muḍa and Abū Alī Muḥammad b. al-Mustanīr (known as Quṭrub) – believed this theory had a negative impact on Naḥw. If it was not for this theory, which was taken from philosophy, Naḥw was a very simple and easy skill and in reality, a fallacy occurred due to this methodological confusion. For example, in the discussion of ishtighāl there are many things taken into taqdīr, and these critics will find this comedic as humans really do not have anything in taqdīr when they speak like this. Or for example mubtada is considered marfū‘ because of the non-existence of an ‘āmil, but philosophically speaking a non-existent entity cannot be a cause for an existent (effect)!

Contemporaries like Ibrahim Muṣṭafa in their works Iḥyā’ al-Naḥw also argue that this theory had a negative impact on Naḥw. He says Naḥw studies the final sign on a word, but this makes Naḥw a very narrow discipline. Furthermore, there are many things studied in Naḥw which do not concern this definition. For example, the cases where a mubtada is prioritized or a fā‘il is prioritized – what does this have to do with the final sign on a word? They have nothing to do with the structure (binā’) of a word. He instead offers many other definitions.

Unfortunately, though, these critics do not offer an alternative. The classical scholars were also aware of these critiques it appears, but their general response and understanding was to record and preserve the Arabic language. Of course, despite this, the Kufans believed the rules of the Arabic language had no relation with geography and time, hence you can derive principles that will be applicable on all regions and tribes; whereas the Basrans believed the Arabic rules have to revolve around the Qurānic, Hijāzī and specifically the Qurayshī conventions.

Principles of Naḥw and Ṣarf are of two kinds:

1) Reflective (naẓarī)

2) Application (tatbīqī)

We do not see the tatbīqī aspect in Uṣūl al-Fiqh itself, but you will see it in Fiqh, whereas when it comes to Naḥw and Ṣarf it happens in the same discipline. The book of Abū ‘Alī al-Fārisī is a tatbīqī book, as well as al–Ḥujjah of Ibn Khālawayh, or al-Mukhtaṣar of Ibn Jinnī. As for naẓarī books, we have al-Ighrāb of Ibn Barkat ibn Anbārī and Lum’a al-Adillah, al-Uṣūl by Dr. Tamam Ḥassān, al-Ashbah wa al-Naẓā’ir and al-Muz’ir fi ‘Ulum al-Qurān of Suyutī, and Kitāb al-Khaṣā’iṣ is a mixture of both reflective and application.

Principles of Naḥw which are reflective address the probativity (ḥujjīyyah) of the principles of Naḥw which will then eventually be applied. Ḥujjīyyah lexically just means evidence, it is a maṣdar ṣanā‘ī – like insānīyyah. Ḥujjīyyah means the validity and reliability of an argument (istidlāl) upon something and in iḥtijāj. Sometimes Ḥujjīyyah of something is inherent (dhātī) such as in analytical philosophy, but elsewhere it is conventional and agreed upon by society.

Ḥujjīyyah – which was validity and reliability on relying on something – is divided into different types:

1) Takwīnī – this is in external reality such as the relationship between feeling thirsty and drinking water.

2) Shar‘īyyah – ḥujjīyyah here means tanjīz (inculpatory) and ta‘dhīr (exculpatory).

3) Ma‘rifīyyah / Manṭiqīyyah – qaṭ‘ and burhān.

4) Lughawīyyah – ḥujjīyyah means to use and rely on some convention to derive a linguistic principle.

Where does the reliability come from? Either it is dhātī, or ja‘lī.

‘Ulūm al-Naqlīyyah (transmitted sciences) are i‘tibārīand, we do not deal with them as we do with ‘Ulum al-‘Aqlīyyah (rational sciences). Our discussion is going to be regarding the transmitters and narrators of language, and how they transmitted this knowledge down to us. This is going to be important to see whether a linguistic principle can be derived properly and whether it has ḥujjah or not. Our discussion concerns Naḥw Naẓarī at the moment and we have analyzed different discussions and definitions so far and have concluded that lugha is a naqlī discipline.

What Constitutes Evidence in Linguistics?

The question is, whose opinions are ḥujjah in lughawī discussions for us to use as evidence when we derive principles?

Scholars propose an ‘aṣr al-ihtijāj (era of dialectic argumentations), and its the conventions of that era that have to be used in one’s arguments. In the era of jāhilīyyah there was no need to develop a discipline, because the Arabic language was pure and it only became impacted when they interacted and mixed with other ethnicities. Later on there was essentially a war between languages, and in fact language is the basis of any civilization. For example, until the Ottoman Empire the language was Arabic and generally in the Muslim world. However, the first thing the colonizers did was to change language and prioritize dialects over classical Arabic – by this they are able to cut you off from your past and tradition.

In Arabic, we can even have a fāsiq from the era of ihtijāj whose words are ḥujjah, but a very pious scholar at a later period words in language will not be ḥujjah. Differentiating between a lughawī and rāwi of a lugha is also very important, and it seems Dr. Hādī al-Faḍlī seems to be the first one to have realized this, although he says there is no difference between a jammā‘ and a rāwī, whereas we believe that there is a difference. A lughawī is a scholar of language, but a rāwī doesn’t have to be a specialist, rather he is a narrator and transmitter. It is possible for someone to be a rāwī and lughawī, or a lughawī but not a rāwī, and a rāwī but not a lughawī. For example, al-Asma’ī was a rāwī and lughawī – if someone is both, then you gain greater conviction in their transmissions.

The question is, does anyone merely transmitting a language make their transmission ḥujjah? No – rather we look at the sīrah of the lughawīyīn of the classical period, and some scholars have said Bashār b. Burd is the last figure who you can rely on to do ihtijāj in language. For example, the Qurān was transmitted with tawātur and it can be used to derive principles, likewise the ḥadīth that are transmitted with lafẓ are ḥujjah (though they are very few quantitatively), but transmission with ma‘na cannnot used to derive principles in grammar since many transmitters were non-Arab.

The aforementioned discussion was regarding the time period, but there is also a discussion on geography. Farābī in his book al-Ḥurūf (a very important work) argues who you can use as evidence to derive principles from a geographical perspective.

The reason why the Arabic language was systemized was because of the Qurān and ḥadīth, and in fact preserving the language and its rules was essentially to preserve religion. How was it preserved though by the scholars – lughawī, rāwī and jammā’ (such as Ibn Manẓūr)? The Arabic language was preserved based on observing the ruwāt, but not with historical methodology, rather with the ḥadīth methodology, by looking at whether these ruwāt were reliable or not.

How Was Arabic Transmitted & Recorded?

How was Arabic transmitted to us? Usually by listening, memorizing and then oral transmission. This riwāyah (transmission) will either be mutawātir or a solitary report. A solitary report which does not have a contextual indicator (qarīnah), is either ṣaḥīḥ or ḍa‘īf (due to it being mursal or containing a majhūl). The latter is not ḥujjah for us to derive a grammatical principle. If it is ṣaḥīḥ, in ḥadīth the narrator has to be ‘ādil and thiqa, but in language is ‘adālah (whether it is being justice, or being a Shī‘a) a condition for a transmitter as well? Ibn al-Anbāri in Lum‘ al-Adillah says the narrator has to be ‘ādil and must have istiqāmah, can be a man or woman, free or slave. If a transmitter is a fāsiq, their transmission it not accepted. Bulūgh on the other hand was a condition for accepting a ḥadīth, but not in transmitting language. A person also has to determine whether their epistemic school is that of wuthūq or withāqa. As far as transmission is concerned, do full sentences, stanzas, poems have to be transmitted, or singular words? These are all questions that a scholar has to answer even after coming across a transmission.

Ibn Fāris says: Lugha is taken from listening, from the ruwāt and then he cites a chain of transmission. Al-Azharī in his Tahdhīb al-Lugha also relies on chains of transmissions and narrators who were known for their precision.

Ḍabṭ al-Lugha (Recording of Language) happens in two stages:

1) Mufradāt: works of ma‘ājim and are sources of single words; they were also called kutub al-lugha. Why wasn’t the Qurān known as a source for this? You need to know the meaning of the words to understand the sentences whose source is the Qurān. The mu‘jam words were compiled using a lot of qiyās – i.e. iqtiḍā ‘aqlī. Khalīl would think – for example – that words are made up of letters, and when he would think of different words that began with each letter, he would then go to the Arab society and determine its meaning.

On a side note, the reason why these works were called mu‘jam, which is from ‘ujmah and is the opposite of faṣāḥa, is because of what Ibn Jinnī says in his al-Khaṣā’īs. He says when you say a‘jamtū al-kitāb it means to “remove” the ‘ujma from it.

2) Tarākīb: the most important source is the Qurān, because it is protected due to tawātur. The second source are the earlier ḥadīth texts (Sunnī and Shī‘a). The third source are dīwān of poets and literary books (such as works of al-Jāḥiẓ).

What is the difference between mu‘jam and balāgha then, when you often find morphological discussions or discussions on differentiating metaphors from real meanings in both kind of works? We can divide ma‘ājim into two:

1) Awalīyyah: these were works that were written before disciplines were developed, and they often contain all raw material and sources for language. Hence you have mixed discussions in them. For example, when Khalīl or Asma‘ī wrote their works, and Sībawayh or other later scholars saw these works, they found a lot of words in these books. They then – perhaps due to influence of Fiqhi discussions – applied concepts like aṣl and far‘ on these words. For example they would say the aṣl is the root verb, whereas the present tense is the far‘. It is essentially an explanation of the multiplicity of so many words – this is how morphology developed and all the discussions such as ibdāl, idghām and so on. They also developed the idea of ḥurūf that are aṣlī and zā’id.

2) Thānawīyyah: these were works written after disciplines were developed, such as morphology and syntax.

So the difference between ma‘ājim and ṣarf books is that the former are concerned with different root-words used by Arabs, and they not care about derivatives (ishtiqāq). Ṣarf on the other hand is concerned with ishtiqāq. ‘Ilm al-Mu‘jam therefore means:

a) A discipline of writing and compiling a mu‘jam or,

b) Studying the mu‘jam.

This resulted in different schools of thought. To get a familiarity with these schools, Ṭabaqāt works of linguists such as Ṭabaqāt al-Naḥwīyīn by Zubaydī, Ṣuyūtī’s work, and also Ibn Qāḍī Shuhba’s book, can all be referenced to study the lives of these scholars. Before Khalīl, a Naḥwī was not just a grammarian, but they were also known to be a qāri of the Qurān. Sayyid Khū’ī and Hadi Ma‘rifat believed the qurrā’ were doing ijtihād, while we believe Dr. Hādī was correct and that it was not mere ijtihād. Before Khalīl there are two ṭabaqāt generally, the first contain Abu Aswad, second is Laythī, Yaḥya b. Ya‘mur, in the third ṭabaqa you have Ibn Abī Aqrab, and so on. In the fifth ṭabaqah you have Khalīl al-Farāhīdī, and the sixth contains Sībawayh.

It was only after Khalīl that the two schools of Basra and Kufa were developed and those later generations also contain a lot of big names.

Al-Uṣūl al-Naḥwīyyah wa al-Ṣarfīyyah

If Naḥw is to derive (istinbāṭ) principles that were present in the minds of Arabs, then speech (kalām) is a source, not an aṣl. Meaning, a lughawī arrives at those principles by listening to Arabs speaking. Prioritization of evidence in Naḥw has to happen in an accurate way. For example, we cannot enter the discussion as Muslims and say the most important source for Naḥw is Quran, then the Sunnah of the Prophet (p), then speech of Imams (a), and only then Islamic poetry and then Jāhilīyyah poetry. This is wrong. When we look at the sīrah of the grammarians, we see that they usually believe the Qurān is the most important source, then the speech of Arabs, and they do not usually rely on ḥadīth at all. In the speech of Arabs they prefer poetry over prose (nathar). Poetry or prose can be used as evidence to prove a grammatical principle. The reason why they did not rely on ḥadīth is because of naql bil ma’na, and many transmitters were also either non-Arab, or lived in cities. For example, you will not find find Abān b. Taghlib relying on ḥadīth of the Imams (a) to prove a grammatical principle. Another reason was because many people you would rely on were those from the past, not your contemporaries.

As mentioned earlier, the principles of syntax and morphology are divided into:

a) Samā’ (Hearing)

b) Qiyās (Analogical Reasoning)

c) Istiṣḥāb (Principle of Continuity)

d) ‘Aql (Intellect)

Samā‘ itself is divided into:

i) Qurān

ii) Ḥadīth

iii) Poetry

iv) Prose / Nathr (maqāmāt, khuṭab, amthāl)

There are many differences of opinions on what is ḥujjah in deriving grammatical principles. In all cases, one of the main sources is samā‘ which means formal Arab speech that has been transmitted authentically. Is samā‘ a source for uṣūl, or from one of the uṣūl itself?

The earliest source for which samā’ has been used as a source is the Qurān itself. With this definition, non-Arab speech is excluded. The word faṣīḥ is not the same definition as used in balāgha, rather it means that a person can simply speak without laḥn (grammatical mistakes) and in accordance with the principles. The transmission has to be a reliable transmission – tawātur or khabar wāḥid of a thiqa. The speech has to be common, not a rare statement uttered by some person here or there.

When it comes to using the Qurān as a source for samā‘, there is an intense debate on the different readings of the Qurān, and whether the Qurān can even be used as a source or not. We will not expand on that here, but we will address another point instead, and that is whether it is allowed to rely on ḥadīth to derive and defend a grammatical principle? There are two schools of thought here: the classical grammarians (until Suyutī), and post-classical grammarians.

The classical scholars themselves were divided into two: those who allowed, and those who prohibited. Some of those who prohibited the use of ḥadīth were Ibn Ḍā’i‘ al-Andalūsī (614-680) who was in fact the first one to reject the use of ḥadīth explicitly and after him we see Ḥayyan al-Andalūsī (the teacher of Ibn Hishām), and Suyuti himself in his al-Iqtirāḥ.

Proponents from the classical period were scholars such as Ibn Jinnī, Abu Ali al-Farisī and most of the classical grammarians. Sayyid Raḍī, Sayyid Murtaḍa, Shaykh Ja‘far Ṭūsī were Shī’a scholars who believed you could rely on ḥadīth in this matter.

The post-classical scholars divided the narrations into naql bil ma’na and bil-lafẓ. Naql bil-lafẓ was also only relevant when they were uttered in certain contexts, like during sermons or supplications. Hence, sermons recorded in Nahj al-Balāgha which are proven to be transmitted accurately can be used as evidence. Anything that is naql bil ma’na is not to be used as evidence.

When we look at the books of early classical grammarians who did not explicitly discuss this matter, we very clearly see that their reliance on ḥadīth is very rare, rather they are mostly relying on the Qurān and poetry. The reason why Ibn Ḍā’i‘ and Ḥayyan etc. would be against it is because they realized that the early grammarians would not rely on ḥadīth and as well as due to naql bil ma’na. They also say that the narrators of ḥadīth were mostly non-Arab and would barely narrate with lafẓ. They would also be cautious because one’s theological views would impact the ḥadīth and there was a real problem of fabrication. Also the lack of ḥadīth compilations in the beginning of Islam was a problem. Classical scholars also were Ahl al-Qurān, they were not generally considered muḥaddithīn, most of them were qurrā’ – in fact Naḥw developed amongst the qurrā’ not the ḥadīth scholars, such as Kisā’ī, ‘Āṣim (teacher of Khalīl), Abu Aswad al-Du’alī etc.

The third and fourth source used for samā‘ is Arabic poetry and prose. When it came to doing ihtijāj with Arab speech, there were a few conditions the scholars mentioned:

1) Location

2) Time

3) Type of Speech – whether it was a speech uttered in the bazars by merchants, as opposed to speech of poets.

Most of the times grammarians were interested in the transmission of pre-Islamic era poets, or poets who lived in the age of Ignorance but also witnessed Islam, and only then did they resort to Muslim poets, up until the end of the Umayyad dynasty.