The HifdhAlgorithm

Most guides to hifdh are a list of encouragements. This one is a system. It tells you what to do on Monday morning, how many times to do it, what to do with it on Tuesday, and what to do with it in five years when you have twenty juz to protect.

That system rests on something, though, and the something is not technique. The Companions did not treat the Qur’an as information to be stored. They treated it as an instruction to be obeyed, and the memorising happened along the way. Parts one to four of this guide are about that, and they are not decoration. A person who skips them can still memorise. Whether the memorising will do him any good is a separate question, and the Prophet, peace and blessings be upon him, answered it plainly.

Parts five to eight are the practical core. Part five is the gate: what you must be able to read before you memorise anything. Part six is the memorisation method itself, step by step, with a tip on every page. Part seven is the revision system, which is the part that decides whether you keep any of it. Part eight puts the two together into a day you can actually live.

There is a peculiar danger in memorising the Qur’an that does not attach to most acts of worship. Fasting is invisible. Charity can be hidden. Hifdh cannot be hidden for long, because it is performed, in front of people, in the most public act a Muslim does. You will be asked to lead. You will be praised. You will feel the difference in a room when it becomes known.

This is why the texts on the virtue of memorisation come with a condition attached so consistently. The scholars who wrote about the reward of the hafiz almost never wrote about it without adding that it is for the one who did it for the sake of Allah, and not for wealth, position, or the pleasure of being called a reciter.

Note what the verse promises: a trade that will never perish. Trade is the right image. You are putting in years, and the return is real. But a trade only pays if the goods are genuine, and the goods here are the intention. Everything else is packaging.

The Prophet, peace and blessings be upon him, described the first three people judged on the Day of Resurrection. A man killed in battle, a man who gave wealth generously, and a man who learned knowledge and recited the Qur’an. All three are asked what they did. All three answer truthfully in the outward sense. All three are told they have lied about why.

It is worth sitting with the fact that the reciter is in that list. Not the thief, not the drunkard. The man who spent his years with the Book. What condemned him was not his recitation but the reason underneath it, and he had made his reason come true: it was said. He got exactly what he worked for, and there was nothing left to pay him with.

The remedy is not to hide your hifdh or refuse to lead prayer. The Companions taught publicly. The remedy is the small, private, repeated act of turning the deed back to Allah, especially at the moments when praise arrives.

When a student today says he cannot retain what he memorises, the modern answer is technique: more repetition, better spacing, a different app. Those things matter, and most of this book is about them. But they are not the first answer our scholars gave, and it is worth knowing what the first answer was.

Malik ibn Anas said the same thing without the poetry: if anything will improve your memory, it is giving up sin. This is not superstition and it is not a threat. It is a claim about what kind of thing knowledge is. If the Qur’an is a light, then the question of retention is partly a question about the condition of the room you are asking it to sit in.

There is a practical version of this too, and it is not mystical at all. Sin occupies the mind. It brings anxiety, and things to hide, and conversations replayed at night. A head full of that has less room for a page of al-Baqarah, and anyone who has tried to memorise while carrying something heavy knows exactly how true this is.

There is a famous du’a said to have been taught by the Prophet, peace and blessings be upon him, to Ali ibn Abi Talib for memorising the Qur’an. It circulates widely and is printed in a great many books. Ibn Kathir called it ghareeb, indeed munkar. Adh-Dhahabi called it shadhdh munkar. It is not established, and a book that is honest with you will say so rather than hand you something to lean on that will not hold.

What is established is better anyway, because it is general and it is unlimited. Seek refuge with Allah before you recite, as the Qur’an itself instructs. Ask Him for knowledge in your own words, in sujud and in the last third of the night, and ask persistently. He taught His Prophet a du’a for exactly this need.

It is difficult to overstate this. Of all the ways a group of human beings could be described, they are described as belonging to Allah, in the way the Ka’bah is called the House of Allah: not because He needs it, but as an honour conferred by attribution. Al-Munawi explained the phrase exactly that way, and then immediately narrowed it.

Memorise it and act by it. Not memorise it. Every scholar who commented on this hadith added the second clause, and some were sharper still. Al-Hakim at-Tirmidhi held that a person is only counted among them if his heart is free of disease and his conduct free of wrongdoing. Ibn al-Qayyim went the other way and widened the door in a manner worth thinking about carefully.

Abdullah ibn Mas’ud, who learned seventy surahs directly from the mouth of the Prophet, peace and blessings be upon him, left a description of what a carrier of the Qur’an should look like from the outside. It is not a list of rules. It is a portrait, built entirely out of contrasts, and every line of it is about being different from the room you are standing in.

Read it again slowly and notice that not one of the six markers is about recitation. Not the beauty of the voice, not the number of juz, not the qira’at held. Night prayer, fasting, seriousness, weeping, silence, humility. A person could satisfy every one of them with three juz, and a person could hold all thirty and satisfy none.

The classical works on the etiquette of the carriers of the Qur’an, from Imam an-Nawawi’s at-Tibyan onward, are surprisingly specific. They do not deal in general exhortation. They deal in the tongue, the temper, the wallet and the household. Here is the substance of what they list, grouped.

What is striking about the list, read as a whole, is how little of it is religious in the narrow sense. There is no item here about recitation, or about how much a person holds. Almost every one is about how he behaves in the ten minutes after he closes the mushaf, which is where his hifdh is actually visible to everybody else.

This is the most precise thing ever said about the relationship between reciting the Qur’an and being changed by it, and it is precise because it separates two variables that we constantly confuse: what a person is, and what a person gives off.

Scent travels. It is what reaches other people: the recitation heard, the halaqah taught, the benefit that spreads outward. Taste is interior. It is what the thing actually is. And the hadith refuses to let us collapse them. A believer with no recitation is still sweet, and a hypocrite with beautiful recitation is still bitter.

Ten verses. Then a full stop, until they had learned the knowledge in them and how to act on them. Think about what that pace implies. It implies that Ibn Mas’ud and those like him could have gone faster and chose not to. It implies that the bottleneck was never memory. And it implies a definition of learning the Qur’an in which memorising the words is one of three components, not the whole of it.

They learned knowledge and action together, as one thing. Not the words this year and the meaning later, when there is time. This is worth being honest about, because the modern arrangement is almost exactly the reverse: memorise thirty juz first, quickly, ideally as a child, and understand it afterwards if life permits.

Sa’d ibn Hisham came to Aisha and asked her about the character of the Prophet, peace and blessings be upon him. Her answer was four words in Arabic, and it remains the highest thing ever said about a human being.

An-Nawawi explained it as acting in accordance with it, keeping its limits, taking on its manners and heeding its lessons. Al-Qadi Iyad put it as: whatever the Qur’an praised, he embodied, and whatever it condemned, he avoided. Ibn Rajab said the same. The explanations converge, because the statement is not really metaphorical. It is a description of a life in which there was no gap between the text and the person carrying it.

Set against that, the ordinary experience of hifdh is humbling. You can spend a morning drilling a verse about restraining anger and lose your temper in the car at noon. This is not hypocrisy. It is being a human being. But it is the gap, and being aware of the gap is what keeps a hafiz from the worst version of himself, which is the one who thinks the memorising was the achievement.

The verse does not ask whether they recited. It asks whether they reflected, and it offers only one alternative explanation for not doing so, which is that something has been locked. That is a severe way of putting it, and it is put that way deliberately.

Hudhayfah ibn al-Yaman described praying behind the Prophet, peace and blessings be upon him, at night. He recited slowly, and when he passed a verse mentioning mercy he stopped and asked for it, and when he passed a verse mentioning punishment he stopped and sought refuge from it. He was doing in real time what the verses were saying. That is what tadabbur looks like when it is not a technique but a reflex.

In the long hadith on purification, after describing prayer as light and charity as proof, the Prophet, peace and blessings be upon him, said something about the Qur’an that has no middle position in it at all.

There is no third outcome. It is not proof for you, or neutral. Everything you memorise takes a side, and which side it takes is decided by what you did with it. Shaykh Ibn Uthaymeen put it directly: if a person disrespects the Qur’an, neglects its wording and its meanings, and fails to act by it, then on that Day it stands against him.

The Prophet, peace and blessings be upon him, was also shown a man in a dream whose head was being crushed with a rock, and was told who it was.

Ibn Battal held that the warning is for the one who neither retains its words nor acts on its meanings, and that a person who does not memorise but does act is not included. Others noted that the failure described is abandonment: not reciting it at night, not weighing its commands. A person who falls short through weakness or incapacity is not who is being described.

Umar ibn al-Khattab narrated this, and the context is instructive: he had appointed a man over Makkah who was a freed slave, and the nobility of Quraysh were surprised. The reason given was his knowledge of the Book of Allah. That is the mechanism the hadith describes, and it operated in the first generation exactly as stated.

The one with most knowledge of the Book leads the prayer. Not the oldest, not the wealthiest, not the most senior. In the burial of the martyrs of Uhud, the Prophet, peace and blessings be upon him, asked which of two men had taken more of the Qur’an, and placed that one first in the grave.

Your rank is at the last verse you recite. Which means the ascent is not a fixed reward handed to a category of people. It is measured out, verse by verse, against what you actually carried. Every page is a step, and the person beside you who took ten more pages seriously will be ten steps further up. It is difficult to think of a stronger argument for the extra half hour.

And then the narration that undoes most people, because it moves the reward to someone else.

If you have ever sat with a page for an hour and produced nothing, and then watched somebody take the same page in fifteen minutes, this hadith was narrated for you specifically.

Notice how strange the arrangement is. The proficient reciter gets the higher station, with the angels. The one who stumbles gets two rewards. The Prophet, peace and blessings be upon him, did not say the struggling one gets less, or gets encouragement, or should try harder. He said the difficulty itself is paid for, separately, on top.

This is not a consolation prize handed to people who will not finish. It is a statement about how the accounting works, and the accounting does not reward talent. It rewards the recitation, and then it rewards the cost. A person for whom this comes easily is being given a gift. A person for whom it does not is being given a wage, and wages are earned.

You will very likely have been told, at some point, that forgetting a surah after memorising it is among the greatest sins, and that this comes from a hadith. That report exists. It is not authentic.

Both al-Bukhari and at-Tirmidhi rejected its chain. Its weaknesses are specific: Abd al-Majid ibn Abi Dawud is alone in narrating it, and there is a break between Ibn Jurayj and al-Muttalib and Anas. A weak report cannot be used to establish that something is a sin, and it should not be used to frighten people either.

What replaces it is stronger and kinder. The authentic texts do not threaten the person who forgets. They instruct the person who neglects. If you strive to revise and still lose some of it, through nature, or age, or a season of life that took your time, there is no sin on you. The blame attaches to abandonment, not to a fading memory.

The Qur’an was not delivered as a document. Jibril recited it to the Prophet, peace and blessings be upon him, who recited it to his Companions, who recited it to those after them, mouth to ear, without a break, until it reached whoever taught you al-Fatihah. That is not a historical detail. It is the mechanism by which the recitation itself, and not only the text, has been preserved.

Which is why the scholars are unambiguous about it. The best way to learn is to recite to a shaykh qualified in recitation. If there is none where you are, it is encouraged that you travel for it. If you cannot travel, learn from those around you and from recordings, and know that a person listening to his teacher directly will learn the Book of Allah more precisely than one listening to a recording, because a recording cannot hear you back.

That last point is the whole argument in one line. Audio is output. What you need is input: someone who hears the exact sound you are producing and tells you it is not the sound on the page. You cannot detect your own errors, because the error and the ear that would catch it are the same organ.

Work down this list honestly with your teacher. Every item you can tick is an item that will not have to be repaired later, at ten times the cost, in verses you have already drilled two hundred times into the wrong shape.

For most adults who can already read Arabic script slowly, closing the remaining gaps on this list takes something between six weeks and six months of short, regular sessions with a teacher. That sounds like a long delay when you are impatient to begin. Set it against the years the memorisation itself will take, and against the cost of repairing a thousand verses learned with a wrong letter in them, and it stops looking like a delay at all.

This is the shortest chapter in the book and the one most often ignored, so it has been given its own page in the hope that it will be harder to skip.

Memorisation is habit formation. Every repetition deepens a groove, and the groove does not care whether what is in it is right. Twenty repetitions of a correct verse builds it. Twenty repetitions of a wrong vowel builds that instead, and it builds it just as firmly.

The asymmetry is what makes this so expensive. Learning a verse correctly takes forty passes. Correcting a verse learned wrongly takes several hundred, spread over months, because you are not learning something new, you are competing with something already installed. And it will still come back under pressure, in the middle of taraweeh, when you are tired.

There is a second cost, and it is the one nobody warns you about. A verse learned wrongly is not only wrong. It is uncertain, and uncertainty is what makes people hesitate. A hafiz who is not sure whether it is an alif or a ya slows down at that spot for the rest of his life, and slowing down at a known weak point is how a whole page comes apart under the pressure of leading a prayer.

One portion, four moves. You will run this loop once per third of a page, three times per page, six days a week, for as many years as it takes. Everything in the rest of this part is detail on one of these four boxes.

The two things that make this method work are both counter intuitive. The first is that the unit of work is a third of a page rather than a page, so that you finish something completely every session instead of touching everything partially. The second is that verses are joined in overlapping pairs as you go, rather than being learned separately and stitched together at the end, which is where most self taught memorisation falls apart.

Before anything else, the page is read aloud to someone qualified to hear it. Not the verse you are about to memorise. The whole page, so that the corrections arrive before the repetition does, and so that your teacher can point at the places on this particular page where students go wrong.

If your teacher is not available every day, this step still happens: read the page to them the session before you intend to memorise it. Working one page ahead of your teacher, rather than one page behind, is the single easiest scheduling change most students can make.

Take the page and split it into three roughly equal blocks. In the standard fifteen line mushaf, that is five lines each. Do not split by verse count and do not agonise over the boundary. Put a light pencil mark at each division and treat those three blocks as three separate pieces of work.

This is doing something specific. A page of the Qur’an is too large to be finished in one sitting at this level of repetition, and a target you cannot finish is a target that gets abandoned halfway. Splitting it into thirds gives you three complete, sealed units per page, and completion is what makes memorisation hold.

If the thirds are very unequal because of where the verses fall, adjust by a line. If a single verse is longer than a third by itself, as happens in al-Baqarah, treat that verse as its own unit and break it at a place your teacher approves of.

Look again at the shape of the ladder. Verse one, verse two, then one joined to two. Verse three, then two joined to three. Verse four, then three joined to four. The joins overlap, like links in a chain, and each one is drilled as hard as the verses themselves.

This solves the failure that ruins most self taught hifdh. A student memorises ten verses individually and finds he can recite any of them if you give him the first word, but cannot get from the fourth to the fifth without a prompt. He has ten memories, not one. The transitions were never learned, because they were never practised.

The symptom of missing joins is easy to recognise, and most people have had it. You can recite the page perfectly at home, then stand in prayer, reach the end of a verse, and find nothing waiting. Somebody says the first word and the whole of the next verse arrives instantly, which proves the verse was never the problem. What was missing was the road into it.

It is worth knowing exactly what you are signing up for, because a plan built on a fantasy about time is a plan that collapses in the third week. Here is the arithmetic of one third of a page.

Per verse, that works out at roughly one hundred and twenty passes for the first and last verse of a third, and one hundred and sixty for the ones in the middle, which appear in two joins rather than one. The middle of a passage gets the most work, which is correct, because the middle is where people get lost.

Read those timings and then look at your actual day. If a full page is two hours and you have forty minutes, the answer is not to run the method at a quarter of the repetitions. The answer is to memorise a third of a page and run it properly.

The final rung of the ladder is the one people skip when they are tired, and it is the rung that decides whether the previous three hundred repetitions turn into anything. When the last pair has been joined, close the mushaf and recite the entire third from memory, twenty times.

It will be uncomfortable for the first three or four passes. That discomfort is the point. Until now every retrieval has been of a verse or a pair, with the surrounding context freshly in your ear. Reciting the whole block cold is the first time you are asking your memory to hold the structure rather than the pieces.

One warning about the seams between thirds. They are the weakest points on the page, because each third was learned as a self contained unit and the transitions between them were never drilled the way the transitions between verses were. Treat the last verse of one third joined to the first verse of the next as its own item, and give it the same forty passes any other join gets.

The classical recommendation is unanimous and it is dawn. The verse describes the recitation of dawn as witnessed, the scholars of IslamWeb and others point to the time before or just after Fajr as the best slot, and there is a practical reason underneath the textual one: your memory is cleanest before the day has put anything into it.

In the mahadhir of Mauritania, students do not begin by reciting. They begin by writing the day’s portion onto a wooden tablet, in their own hand, from the text, and only then start the repetition. Teachers in that tradition will tell you that a large part of the memorisation happens during the writing, before a single repetition has been counted.

Whether or not the proportion they quote is exact, the mechanism is not mysterious. Writing forces you through the verse letter by letter at a pace you cannot rush. It exposes the words you were gliding over. And it engages a third channel, the hand, on top of the eye and the ear.

This is optional in a way that nothing else in part six is optional. Plenty of huffadh have never written out a page in their lives. But if you are someone who finds that verses feel slippery, or that you cannot picture the page when your eyes are closed, this is the first thing to try, and it costs ten minutes.

A page you memorised this morning is not yet a page you have memorised. It is in a fragile state that lasts about a day, and what you do at the two edges of that day matters more than another thirty repetitions in the afternoon.

None of these three passes is long. Together they are perhaps ten minutes, and they are spread across a day you were living anyway. Set against another thirty repetitions squeezed into the same session that produced the first three hundred, they are worth several times more, because they are spaced and the session is not.

Every other instruction in this part can be bent. This one cannot, and breaking it is the single most common reason people spend five years memorising and end with nothing they can rely on.

Before you take a new portion, recite yesterday’s from memory. If it comes out cleanly, take the new one. If it does not, yesterday’s portion is today’s work, and there is no new portion. Not a reduced one. None.

The scholars at IslamQA put the same rule the other way round: never move to a new portion until the previous one is perfectly memorised, so that it settles. Easy to agree with in principle, and very hard to obey on a Tuesday when you are behind and the new page is right there.

The image is exact, and it repays attention. Tethered camels are not gone. They are standing right there, valuable, entirely yours, and requiring nothing except that somebody checks the rope. The loss is never dramatic. It is the absence of a small, repeated, boring act.

What the curve below shows is why the schedule matters more than the effort. Left alone, a memorised page falls off a cliff within the first week. Touched on a schedule, each return restores it and the next decline is slower, until the intervals can stretch to weeks and it still holds.

This is the entire justification for the three tier system on the next page. Fresh material needs to be touched daily because it is on the steep part of the curve. Old material can be touched every few weeks because it is not. Treating both the same way is how people end up spending two hours a day on revision and still losing ground.

The hifdh schools have run the same three tier system for centuries, under these three names. It divides everything you know by how recently you learned it, and gives each tier a different frequency. It is the most effective arrangement anybody has found, and it maps almost exactly onto the curve on the previous page.

The two rules that make it work: sabqi is never skipped, because that is the material sitting on the steepest part of the curve, and manzil is a fixed quota rather than a feeling. You do not revise until you feel done. You revise the number of pages your cycle requires, and then you stop.

Abstractions are easy to agree with and hard to run. So here is a real day for a student who has five juz memorised, is adding five lines a day, and has one hour. Yesterday he finished page 147.

Notice the proportions. Twenty five minutes go on new material and thirty five on material he already holds, which is almost exactly the forty against sixty split that part eight prescribes for somebody at five juz. That ratio feels wrong to every beginner and is obvious to everybody who has been doing this for five years. And notice the first row: the whole plan is conditional on it.

Manzil is a number, not an intention. Work out how many pages you hold, decide how often you want to come back round to any given page, and divide. The standard mushaf is 604 pages, and a juz is about twenty.

To convert those into time: a page you know well takes about ninety seconds to recite from memory. So twenty pages is half an hour, and eighty six pages is a little over two hours. Choose the column your life can actually sustain, because a quota you abandon is worse than a slower one you keep.

Set the cycle once, write the number in the front of your mushaf, and keep a note of which page you reached. A quota you have to reinvent every morning is a quota you will negotiate downwards by the end of the month.

For someone who holds the whole Qur’an and wants a weekly cycle, there is a ready made division, reported as the practice of the Companions themselves. They were asked how they divided the Qur’an into portions, and gave a sequence of numbers.

One portion a day gives a complete recitation of the Qur’an every seven days, which is the pace the texts single out. The portions are uneven in surah count and even in length, which is the point: the early surahs are long and the last portion, the mufassal, holds sixty five short ones.

Of everything in this part, this is the highest yield instruction, and the scholars at IslamQA describe it as the way that has been tried and tested: repeat what you have memorised in your prayers, particularly in the sunnah and nafl prayers, and above all in qiyam al layl.

There are four separate reasons it works better than sitting revision, and they compound.

Ask any hafiz where he goes wrong and he will not describe random gaps. He will name specific places: the openings of the Ha Mim surahs, the stories of Musa told with small differences in a dozen surahs. Verses that closely resemble other verses cause a large share of every error made by people whose memorisation is otherwise sound.

The mistake is to treat these as ordinary weakness and respond with more repetition of each passage separately. That makes it worse: each is reinforced in isolation and the two stay confusable. They have to be worked on together.

Self assessment in hifdh is unreliable in a very specific way. When you revise alone, you are the reciter and the examiner at once, and the examiner has already heard the answer. Small slips get waved through. Whole half verses get skipped without registering. The feeling of knowing a page and the fact of knowing it come apart, quietly, over months.

The remedy is unglamorous: recite to a human being holding the mushaf, regularly, and let them stop you. The scholars make this an explicit part of the method, that the memoriser recites to another hafiz or to someone who can follow along and point out his errors.

If there is genuinely nobody available, the next best thing is to record yourself reciting from memory and then listen back with the mushaf open. It is uncomfortable, which is the point: hearing your own recitation from the outside removes the examiner who already knew the answer. Most people find two or three errors on the first page they try this with.

These three are constantly confused, and confusing them is how people end up doing one of the three and believing they have done all of them. They are different activities, with different purposes, and each has its own claim on your day.

The third one is the one people drop, on the reasonable sounding grounds that they are already reciting for an hour a day. But revision is performance and reading is reception. In revision your attention is on retrieval and accuracy. In wird your attention is on the meaning, and it is the wird that keeps the Qur’an a living address to you rather than a body of material you maintain.

A beginner spends most of his hifdh time memorising, and that is correct: he has almost nothing to lose. A person holding twenty juz who still spends most of his time memorising is heading for a collapse, and usually gets there in about a year.

The percentages are of hifdh time, not of your day, and they are a guide rather than an instruction. What is not a guide is the direction. Every juz you add makes revision more expensive and new memorisation no cheaper, so the split has to keep moving, and the person it moves against is the person who most enjoys moving forward.

There is a reason this is so hard to accept, and it is not laziness. New memorisation produces a visible result: yesterday you did not have that page and today you do. Revision produces nothing you can point at. The whole of its output is the absence of a loss you will never be able to measure, and human beings are very poor at valuing that.

These are built from the numbers in part six, so the time given for new memorisation is enough to actually run the ladder rather than a wish. Choose by the time you genuinely have on your worst week, not your best one.

Note what happens as the hours increase: the new portion grows, but revision grows faster, and the time recited to a teacher appears as its own line rather than being assumed. The full time plan spends more of its day on old material than on new, and that is not a compromise forced by the schedule. It is what the method requires.

There is a second speed limit in this subject that has nothing to do with memorisation and everything to do with reading, and it is worth knowing because it puts a ceiling on the whole enterprise. Abdullah ibn Amr wanted to complete the Qur’an constantly. He was negotiated down.

The majority of scholars take from this that a complete recitation every seven days is the recommended pace, and that going faster than three days is disliked, because at that speed understanding is gone. Ibn Mas’ud instructed the same: read it in seven, and do not read it in less than three. There are recognised exceptions for Ramadan, for i’tikaf, for Makkah, and for the purpose of revising memorisation, taken occasionally rather than as a habit.

No plan survives a year intact. What separates people who finish from people who stop is not whether they were interrupted, but what they did in the first week afterwards. Here is what to do in each case.

A plan in your head is an intention. A plan on paper is a commitment you can be held to. Fill this in, in pen, and put it inside the front cover of your mushaf.