← technopuritan.com · a working analysis of the Techno-Puritan corpus · companions: Meat and Drink · The Seventh Day · This Generation · Proper Worship Practices · The Genesis Ledger · The Assembly Concordance · The New Religion Census · The World to Come · Eshet Chayil · Risking Life for Thrills
Around the year 210 the ordinary church in Carthage held that Christian fasting was ex arbitrio, non ex imperio: of free choice, not of command. We know this because the man arguing against them wrote it down, and that man was Tertullian, the first great Latin theologian, and he lost. Over the next four centuries copyists wrote the word "fasting" into four verses of the New Testament and took it out of none; the position Tertullian lost with became canon law with an anathema attached; and by 393 a man who said what the church of 210 had said was condemned as a heretic for it. This page is the paper trail. It names the manuscripts for each of the four insertions, puts every church writer from 100 to 260 on one table, gives Tertullian's fight in full with the verses his opponents used (they are the same verses the Meat and Drink page uses), dates each council that turned a custom into a command, and follows the question to the Lenten sausage supper in Zurich in 1522 and the Puritan fast day. It is a companion to Meat and Drink, which reached the conclusion; this is the evidence for one sentence of it.
The Meat and Drink page concluded that in the canon this tradition reads there is no command to fast: one prediction that the disciples will (Mark 2:20), a long record of people who did, and nothing with "thou shalt" on it. The pushback writes itself. The church has fasted since the beginning; Jesus said "when ye fast"; Mark 9:29 says "prayer and fasting"; Cornelius was fasting when the angel came; Paul told married couples to make room for "fasting and prayer." The first of those is in Matthew, which this house does not read. The other three are in the text the house does read, and this page's first job is to show, with the copies named, that none of the three was in the text when the text was written.
The second job is to answer "the church has always fasted." It has. The question is whether anyone was ever told to, by whom, and when. That is a question with a date on it, and this tradition's rule is that a doctrine with a date on it is a human doctrine: Nicaea is the corpus's standing example of a council that made one (TP 35.0.1), and the corpus already judges the Synod of Laodicea by name on the Sabbath (TPS 73.2). The same council gave Lent its first binding form. The method is not new to the house either. The episode on "the Bible's most embarrassing verse" reads Codex Sinaiticus, "a scribe from the year 300," and finds "this line with dots over it, which means erase this line, and then somebody erased the dots" (TPS 71.2). That is textual criticism, done by the household, on a verse it wanted to keep. This page does it on four verses it does not need.
The third job is a correction. The first pass of Meat and Drink said the four insertions were "absent from the oldest manuscripts." For two of them that is too strong. A papyrus from about 250 apparently carries the Mark gloss, and Tertullian's Bible already had the Acts one by 210. The glosses are old. The argument was never their age; it is the quality of the witnesses that lack them and the fact that the traffic only ever ran one way. That is said properly below, with the counter-arguments printed.
Rude conclusion first. Every sentence in the New Testament that reads like an instruction to fast, or a report that a praying person was also fasting, is one of four: Mark 9:29, Matthew 17:21, Acts 10:30, 1 Corinthians 7:5. All four are absent from the earliest and best witnesses to their books, all four appear in later copies, and no copy anywhere lacks a fasting word that the earliest copies have. The modern critical editions omit all four and every major modern translation follows them. The King James Version has all four because it descends from Erasmus's 1516 Greek text, which was set from a handful of twelfth-century copies. When the house quotes the KJV on these verses it is quoting the scribes, so this page prints both readings each time.
What counts as the text. The earliest recoverable reading. A later hand is evidence of what the copyist's church practised, which is real and is used below for exactly that; it is not evidence of what the author wrote. This is the Sinaiticus rule from TPS 71.2 applied in the opposite direction: there the scribes removed what embarrassed them ("nor the Son"), here they added what they did.
Description is not command. Anna fasted. Antioch's prophets fasted. The catholics of 210 kept Wednesday and Friday. None of that is a rule until someone writes one, and when someone does, it has a date and an author, and the house's rule applies (TP 8.0; TP 35.0.1).
The canon this page reads. Matthew and Revelation are not read as scripture here (TP 1.0, TP 3.0, TP 5.0; the house position on Matthew, with the Mark-priority argument in TPS 71.2). Matthew 17:21 is therefore tracked only as evidence of the mechanism, since it is the one case where the whole chain is visible: a gloss in Mark grows a verse in Matthew. Matthew 9:14 appears for the same reason.
Translations. Scripture is KJV. Tertullian, Eusebius, Hermas, the Didache, the councils, Jerome, Augustine and Socrates are quoted from the Ante-Nicene and Nicene and Post-Nicene Fathers series, which are public domain; the Latin of Tertullian is given where the wording is the point. Zwingli, the Augsburg Confession, Calvin and the Westminster documents are quoted from the standard English editions named in the foot.
Sigla, for anyone checking: ℵ is Codex Sinaiticus and B is Codex Vaticanus, both fourth century and the two oldest complete New Testaments; ℵ* is the first hand and ℵ² a later corrector. A, C, D, W, Θ, Ψ, L, K are fifth- to ninth-century codices. P45, P46 and P74 are papyri; "vid" after a papyrus means it is damaged at the point and the editors are reading it from the space. 0274 is a fifth-century fragment of Mark from Egypt. "k" is Codex Bobiensis, an Old Latin copy of about 400 that preserves the African Latin text of Cyprian's day. "Byz" is the Byzantine mass of medieval copies, the ancestor of the printed Greek behind the KJV. The lists are from the Nestle-Aland and United Bible Societies apparatus and Metzger's Textual Commentary.
| Where | As the earliest witnesses read | As the later hand reads | The witnesses, and what they show | Reading |
|---|---|---|---|---|
| Mark 9:29 | "This kind can come forth by nothing, but by prayer." | "…but by prayer and fasting." (KJV) | Without "and fasting": ℵ* (the first hand of Sinaiticus; a corrector later wrote the words in), B, 0274, k, the oldest stratum of the Old Georgian (the Adysh codex, 897), and, per the UBS apparatus, Clement of Alexandria (died c. 215). With it: P45 (about 250, damaged here and read from the gap), ℵ², A, C, D, L, W, Θ, Ψ, the families f1 and f13, the Byzantine mass, and the Latin, Syriac and Coptic versions. Metzger's committee: in light of "the increasing emphasis in the early church on the necessity of fasting," the words are "a gloss that found its way into most witnesses." Tertullian wrote a whole book to prove fasting drives off demons (Exhibit C) and never quotes this verse, which is an argument from silence, but a loud one. Omitted by NA/UBS, ESV, NIV, NRSV, NASB. | LATER HAND — early, and still a hand |
| Matthew 17:21 | No verse. 17:20 runs straight into 17:22. | "Howbeit this kind goeth not out but by prayer and fasting." (KJV) | Absent from ℵ*, B, Θ, 0281, 33, 579, 788, 892*, two Old Latin copies (e, ff1), the Sinaitic and Curetonian Syriac, the Sahidic and part of the Bohairic Coptic, and Eusebius's canon tables. Present in ℵ², C, D, K, L, W, Δ, f1, f13, the Byzantine mass, the Vulgate and the Peshitta. The verse is Mark 9:29 copied across to make the two accounts match, and it arrives with "and fasting" already attached, which means the Mark gloss is older than the Matthew verse. The chain is visible end to end: a gloss in one gospel becomes a verse in another. Doubly out here: a later hand, and Matthew. | LATER HAND — and Matthew |
| Acts 10:30 | "Four days ago I was praying in my house until this hour, the ninth" (ἤμην τὴν ἐνάτην προσευχόμενος). | "Four days ago I was fasting until this hour; and at the ninth hour I prayed in my house." (KJV) | Without fasting: P74, ℵ, A*, B, C, 81, 1175, the Vulgate and the Coptic. The Byzantine text adds "fasting and" before the prayer. The Western text (Codex Bezae, D) adds it in different words: "from the third day until this hour I was fasting, and praying at the ninth," a three-day fast ending as he speaks. Two traditions added the fast, in two wordings; the Alexandrian witnesses have none. And it is old: Tertullian, writing about 210, says of Cornelius that "his fasts had been heard" (On Fasting 8), so the fasting text was in Carthage's Bible before any surviving Greek copy of Acts was made. The one counter-argument on record is that a scribe deleted the word because nothing earlier in the chapter mentions Cornelius fasting; weighed against two separate additions and a clean Alexandrian text, it is the weaker case, but it exists. Omitted by NA/UBS and the modern translations. | LATER HAND — twice over |
| 1 Corinthians 7:5 | "…that ye may give yourselves to prayer; and come together again." | "…that ye may give yourselves to fasting and prayer." (KJV) | Without fasting: P11, P46 (about 200, the oldest copy of Paul there is), ℵ*, A, B, C, D, F, G, P, Ψ, 33, 81, 1739, the Old Latin, the Vulgate, the Coptic, the Armenian, the Ethiopic. With "fasting and prayer": the corrector of ℵ, K, L, the Byzantine mass, the lectionaries, the Syriac, the Gothic. A third group (330, 451, John of Damascus) has "prayer and fasting," the other way round. Two insertions in two orders is the signature of a gloss. Look at what it does: it turns a married couple's prayer into an occasion for sexual abstinence, which is the whole ascetic programme in one word, written into the one chapter of Paul that says married people should not deny each other. | LATER HAND — Byzantine |
| Matthew 9:14 (minor; Matthew) | "Why do we and the Pharisees fast, but thy disciples fast not?" | "…fast oft…" (KJV) | Without an adverb: ℵ*, B, 0281, a scatter of minuscules, Cyril. With "oft" (πολλά): ℵ², C, D, L, W, Θ, f1, f13, Byz. With "frequently" (πυκνά) instead: ℵ¹, the Latin, the Sinaitic Syriac, Hilary, Jerome, Augustine. Two different intensifiers added to the same spot by two different traditions, neither in the earliest text. The editors bracket it. Listed only because it is the fifth case of the same drift, and in Matthew. | MINOR — direction only |
What never changed. The control group matters as much as the four. Anna "served God with fastings and prayers night and day" (Luke 2:37). John's disciples and the Pharisees fasted and Jesus's did not (Mark 2:18–20). The Pharisee fasts twice in the week (Luke 18:12). Jesus "did eat nothing" for forty days (Luke 4:2). The prophets at Antioch "ministered to the Lord, and fasted," then "fasted and prayed" and sent Paul out (Acts 13:2–3). Paul and Barnabas appointed elders "and had prayed with fasting" (Acts 14:23). Paul lists "fastings" among his hardships (2 Corinthians 6:5, 11:27). "The fast" is a date on Paul's voyage (Acts 27:9, the Day of Atonement). Every one of those is in every copy. No scribe ever took a fast out. They only ever put fasts in. So the honest sentence is not "the oldest manuscripts lack them"; two of the glosses were in circulation by 250. The honest sentence is: the earliest and best copies of each book lack them, every later tradition adds them, sometimes in two competing wordings, and the drift has one direction. That is the same shape as "nor the Son" in Matthew 24:36 (TPS 71.2), read backwards. There the copyists removed what embarrassed them. Here they added what they practised.
Erasmus printed the first Greek New Testament in 1516 from the few Byzantine manuscripts he could borrow in Basel, none older than the twelfth century. Stephanus and Beza reprinted it; the King James translators worked from Beza. So the "Received Text" carries all four, plus Matthew 17:21, and so do the Geneva Bible the Puritans carried and the KJV the house quotes. Sinaiticus was found in 1844, Vaticanus was not properly collated until the 1860s, and the papyri came out of Egypt in the twentieth century. Nobody who wrote before 1850 could have known. The Augsburg Confession quotes Matthew 17:21 as scripture (Exhibit E) because in 1530 it was.
Any modern translation prints the shorter reading and footnotes the longer. If a fasting verse is quoted at you, open an ESV, NIV or NRSV at the reference and read the note. The whole argument of this exhibit is checkable in under a minute, which is also why it is safe to make: it does not depend on trusting this page.
A tradition with zero commands to fast and one prediction that disciples will, "in those days" after the bridegroom is taken (Mark 2:20). The practice is everywhere in the record and nowhere in the law. That is the finding Meat and Drink reached from the Old Testament side, where every fast is a response to something and the only scheduled ones are the Day of Atonement, the exiles' calendar that God questions in Zechariah 7:5, and the Pharisee's two days.
Every Christian writer who touches fasting before the councils, on one table. The column that matters is the last: was anyone told to fast, by what authority, and for what reason. Dates are the usual ones and are approximate.
| Source | Date | What it says | Choice, custom, or command |
|---|---|---|---|
| Didache 7:4 | c. 100 | "But before the baptism let the baptizer fast, and the baptized, and whatever others can; but you shall order the baptized to fast one or two days before." | A command, once, tied to a rite. The first Christian fasting rule, and it is a preparation for baptism, not a calendar. ORDERED — once, before baptism |
| Didache 8:1 | c. 100 | "But let not your fasts be with the hypocrites; for they fast on the second and fifth day of the week; but fast on the fourth day and the Preparation." | Imperative in form. The reason given is not God but the Pharisees: fast on different days from them. The content is identity, not holiness. These are the "stations" that the catholics of 210 kept "not under the law of a precept" (Exhibit C). COMMUNITY RULE — a tribal marker |
| Polycarp, To the Philippians 7 | c. 110 | "…watching unto prayer, and persevering in fasting." | Exhortation in a list with prayer and sobriety. Fasting as a practice; no day, no rule. PRACTICE — exhorted |
| Barnabas 3; Justin, Dialogue with Trypho 15 | c. 100–160 | Both quote Isaiah 58 at length and tell the reader to learn "the true fast": loose the bands of wickedness, feed the hungry, house the poor. | Fasting redefined as what Isaiah said it was for. Neither sets a day. REDEFINED — Isaiah 58 |
| 2 Clement 16:4 | c. 150 | "Fasting is better than prayer, and alms-giving is better than both." | The drift is already visible: a preacher ranks fasting above prayer. Note what still sits on top. RANKED — alms first |
| Shepherd of Hermas, Similitude 5.1 and 5.3 | c. 140 | Hermas is fasting on a mountain. "Why have you come hither so early?" "Because, sir, I have a station." "What is a station?" "I am fasting, sir." "You do not know how to fast unto the Lord: this useless fasting which you observe to Him is of no value." The true fast: "do no evil in your life, and serve the Lord with a pure heart." Then the method: "in the day on which you fast you will taste nothing but bread and water," reckon what the meals would have cost, and give the money to a widow, an orphan, or someone in need. | The earliest use of "station" for the Wednesday/Friday fast, and the earliest Christian text to say the station is worthless as such. The fast is converted into a transfer of food to someone who needs it, which is Isaiah 58 with arithmetic. REDIRECTED — a fast is a meal given away |
| Irenaeus to Victor of Rome, in Eusebius, Church History 5.24.12–17 | c. 190 | On the fast before Easter: "some think that they should fast one day, others two, yet others more; some, moreover, count their day as consisting of forty hours day and night. And this variety in its observance has not originated in our time; but long before in that of our ancestors." "The disagreement in regard to the fast confirms the agreement in the faith." Polycarp and Anicetus could not persuade each other and "immediately made peace with one another, not caring to quarrel over this matter." | The one fast every church kept had no fixed length, the variety was older than anyone living, and the bishop of Lyons tells the bishop of Rome it does not matter. UNFIXED — custom, and said to be |
| Clement of Alexandria, Stromateis 7.12 | c. 200 | The true Christian "knows also the enigmas of the fasting of those days, I mean the Fourth and the Preparation," which he reads as abstaining from greed and lust, the days being named for Hermes and Aphrodite. | Knows the two station days; allegorises them into ethics. No command. KNOWN — allegorised |
| Tertullian, On Fasting 2 | c. 210 | The catholic majority's doctrine, in his hostile summary: the only fast the gospel fixed is the days "in which the Bridegroom was taken away"; all other fasting is "ex arbitrio, non ex imperio," by choice, not by command; the apostles imposed "no other yoke of definite and general fasts"; the stations keep Wednesday and Friday but run voluntarily and end at the ninth hour. Full text in Exhibit C. | The mainstream position of the Latin church, stated by its enemy. CHOICE — stated as the majority's doctrine |
| Tertullian, On Fasting 13 | c. 210 | The catholics say "these things are to be done of choice, not of command," yet "it is a customary practice for the bishops withal to issue mandates for fasts to the universal commonalty of the Church," and "throughout the provinces of Greece there are held in definite localities those councils." | Tertullian's best point: the catholics accepted a fast called by a bishop for an occasion. That is the occasioned public fast, the Old Testament pattern, and the Puritan one (Exhibit E). It is not a calendar. OCCASIONED — the bishop's call |
| Apostolic Tradition 23 | 3rd c., composite | "Widows and virgins should fast often and pray for the church. Presbyters, should they wish, and laypersons may likewise fast. A bishop cannot fast except when all the people are fasting." | Fasting assigned as a duty to one order (the church-supported widows and virgins), optional for everyone else, and forbidden to the bishop except with the whole people, which again means a called fast. WHEN THEY WISH |
| Origen, Homilies on Leviticus 10.2 | c. 240 (in Rufinus's Latin, c. 400) | "We have the days of the fortieth consecrated to fasting; we have the fourth and sixth days of the week, on which we fast solemnly." | The first "Lent" in a Christian author, if Origen wrote it; the text survives only in a translation made after Lent was law, and scholars suspect the translator updated it. Flagged, not relied on. SCHEDULED — possibly retrojected |
| Dionysius of Alexandria, To Basilides, canon 1 | c. 250–260 | "All do not carry out the six days of fasting either equally or alike; some pass even all the days as a fast, remaining without food through the whole; while others take but two, and others three, and others four, and others not even one." He reproves only those who break early, "before midnight," and praises those who "persevere even on to the fourth watch." | The bishop of the largest church in Christendom describing the one obligatory fast as kept for six days, four, three, two, or none, and blaming nobody for the number. UNEQUAL — and tolerated |
What the table shows. In a hundred and sixty years the only fasting commands are the Didache's two: a day or two before baptism, and "not on the hypocrites' days." Everyone else who writes either turns fasting into charity (Barnabas, Justin, Hermas) or says in so many words that its days and length are nobody's to fix (Irenaeus, the catholics of 210, the Apostolic Tradition, Dionysius). The one scheduled fast everybody kept, the days before Easter, ran from one day to six depending on the city, and the bishops of Lyons and Alexandria both say that is fine. This is the church that Tertullian called unspiritual for not fasting enough.
Quintus Septimius Florens Tertullianus of Carthage, roughly 155 to 220, the first theologian to write in Latin and the man who gave the West the word trinitas. The corpus already knows him for that (TP 34.0.2, TP 35.0.1) and for changing his mind about voluntary martyrdom once he had joined the Montanists (TPS 3.0). Around 207 he went over to the New Prophecy, the movement from Phrygia that held the Paraclete was still speaking through its prophets and still issuing discipline. De Jejunio adversus Psychicos, "On Fasting, against the Psychics," was written about 208 to 211. "Psychics" is his insult for the ordinary catholics, taken from 1 Corinthians 2:14: the soulish, the unspiritual. He calls them that because they would not fast more.
Two weeks of xerophagy a year, "the Sabbaths, to wit, and the Lord's days, being excepted" (ch. 15): dry eating, no meat, no wine, no juicy fruit, no baths. The stations, the Wednesday and Friday half-fasts, prolonged to evening instead of ending at the ninth hour. All of it by the Paraclete's command, binding on everyone. The catholics called it novelty, "constantly reproaching us with Novelty" (ch. 1), and compared the dry diet with the purification fasts of the devotees of Apis, Isis and the Great Mother (ch. 2). That comparison is the house's cultus flag, made in 210 by the people the house would side with.
The Law had its one fast, the tenth day of the seventh month, "Holy shall be to you the day, and ye shall vex your souls." In the gospel the only legitimate fast days are those "in quibus ablatus est sponsus," in which the Bridegroom was taken away, "the legal and prophetical ones being abolished." Everything else is "ex arbitrio, non ex imperio novae disciplinae," by free choice, not by the command of a new discipline, "according to the times and causes of each." The apostles kept it that way, "nullum aliud imponentes iugum certorum et in commune omnibus obeundorum ieiuniorum," imposing no other yoke of definite fasts to be undertaken by all in common. Likewise the stations, "which have their own days, the fourth and the sixth," run voluntarily, "neither under the law of a precept nor beyond the close of day," since prayers generally end about the ninth hour after Peter's example in Acts. And since faith is free in Christ, it requires no abstinence from particular foods.
Galatians 4:10 against "observers of days." Isaiah 58, "God has not chosen such a fast." The Lord's teaching that a man is not defiled by what goes into his mouth (Mark 7:15). "Food commendeth us not to God" (1 Corinthians 8:8). "The kingdom of God is not meat and drink" (Romans 14:17, answered in ch. 15). And the Pastoral: those who, "as they forbid to marry, so bid us abstain from meats created by God," "giving heed to spirits which seduce the world" (1 Timothy 4:1–3), thrown at the Montanists as a description of them. Then the conclusion: it is faith and love that count, not fasting. Lay that list next to Exhibit D of Meat and Drink. It is the same list, in the order a Carthaginian presbyter heard it from his bishop in 210, before any council had met.
Chapter 3: the first sin was eating. Adam "sold salvation for his gullet," so hunger is the first remedy and fasting goes back to Eden. That is the exact inversion of Genesis 2:16, "of every tree of the garden thou mayest freely eat," and it is where every later theology of fasting-as-penance starts. Chapters 7 to 9: Moses, Elijah, Daniel, Hannah, Anna, Nineveh, and Cornelius, whose "fasts had been heard," which tells us his Acts already had the gloss (Exhibit A). Fasts are "the weapons for battling with the more direful demons" (ch. 8), and yet in a book with no other purpose he never quotes "this kind goeth not out but by prayer and fasting." Chapter 13, his strongest point: your own bishops order fasts for the whole church whenever they judge fit, and the Greek provinces hold councils to do it; if a bishop may command a fast, why may the Paraclete not? Chapters 16 and 17, the taunts: "your belly is god, and your lungs a temple"; your love shows its fervour "in sauce-pans" and your faith its warmth "in kitchens."
The bishops of Asia, Rome and Carthage refused the New Prophecy. Jerome (On Illustrious Men 53) says Tertullian, "driven by the envy and abuse of the clergy of the Roman church," "lapsed to the doctrine of Montanus," wrote "On fasts" among the books "against the church," and "is said to have lived to a decrepit old age," outside it. Augustine (On Heresies 86, c. 428) records a sect of Tertullianists that dwindled in Carthage until its last few members "joined the Catholic Church and handed over their basilica," within living memory. The xerophagies never became law anywhere. The question as he posed it was whether fasting could be imposed ex imperio, by prophetic command, on the whole church. On that question he lost, and the church that beat him held, in writing, the position this tradition reads out of the text.
Where the catholic verses of 210 already sit in the house's file. Mark 7:15 is Meat and Drink Exhibit C, first row. Isaiah 58 is Exhibit B. 1 Corinthians 8:8, Romans 14:17 and 1 Timothy 4:1–3 are Exhibit D, and the corpus's own use of Romans 14 against "the old stringent Jewish food restrictions" is TP 35.0.1. Galatians 4:10 is the one text the house's file did not have, and it belongs there: "ye observe days, and months, and times, and years. I am afraid of you." The page did not get its list from Tertullian's opponents. It got it from the same place they did.
The free-choice position did not survive the fourth century, and the text did not change between 210 and 340 except for the four glosses. What changed is that councils met. Each row is a date, an authority, and what it did to someone who held the 210 position. Canons are quoted from Percival's translation.
| Date | Authority | What it did | Status of "ex arbitrio" afterwards |
|---|---|---|---|
| 325 | Nicaea, canon 5 | Provincial synods twice a year, "the one before Lent," the Greek being τεσσαρακοστή, the fortieth. A forty-day pre-Easter season is assumed as a fixed point on the calendar. | Lent has a date, and it is the same council the corpus uses as its model of a human doctrine (TP 35.0.1). DATED — Lent, 325 |
| c. 340–365 | Synod of Gangra, canons 1, 2, 18, 19 | Against the Eustathians, ascetics who condemned marriage and meat. Canon 1: anathema on anyone who condemns marriage. Canon 2: anathema on anyone who "shall condemn him who eats flesh, which is without blood and has not been offered to idols nor strangled." Canon 18: anathema on anyone who "under pretence of asceticism, shall fast on Sunday." Canon 19: "If any of the ascetics, without bodily necessity, shall behave with insolence and disregard the fasts commonly prescribed and observed by the Church, because of his perfect understanding in the matter, let him be anathema." | The first anathema in Christian history for not fasting, passed in the same sitting as the first anathema for condemning meat. The church takes the restriction away from the ascetic and the freedom away from everyone else at once; from here on the church owns the plate. Note canon 2 still fencing it with Acts 15's blood rule. COMMAND — with anathema |
| c. 355 | Aerius of Sebaste, in Epiphanius, Panarion 75 | A presbyter who, in the standard summary of Epiphanius's hostile account, "condemned all set fasts: a Christian man should fast when he felt it to be for his soul's good; appointed days of fasting were relics of Jewish bondage," and who treated the Easter observance the same way. Epiphanius catalogues him as a heretic and answers that the Wednesday, Friday and Lenten fasts are of apostolic institution. | The position of the catholic majority of 210, now a numbered heresy. Nothing in the argument changed; its owner did. HERESY — by catalogue |
| c. 363 | Synod of Laodicea, canon 50 | "The fast must not be broken on the fifth day of the last week in Lent… but it is necessary to fast during all the Lenten season by eating only dry meats." | Xerophagy, the Montanist "novelty" the catholics mocked in 210, now compulsory for the whole of Lent. This is the council whose canon 29 the corpus already judges on the Sabbath (TPS 73.2); canon 50 is the same body, the same session, and the same kind of act. COMMAND — the dry diet, universal |
| c. 380 | Apostolic Canons 66 and 69 | 69: "If any bishop, presbyter, or deacon, or reader, or singer, does not fast the holy Quadragesimal fast of Easter, or the fourth day, or the day of Preparation, let him be deposed, unless he be hindered by some bodily infirmity"; a layman is excommunicated. 66: "If any of the clergy be found fasting on the Lord's day, or on the Sabbath, excepting the one only, let him be deposed." | The calendar now has penalties in both directions: deposition for not fasting on the appointed days, deposition for fasting on the forbidden ones. The Didache's "not on the hypocrites' days" has become church law with a career attached. COMMAND — deposition and excommunication |
| 393 | Jovinian, condemned at Rome and Milan | A Roman monk whose third proposition, as Jerome reports it (Against Jovinian 1.3), was "that there is no difference between abstinence from food, and its reception with thanksgiving." Condemned by Pope Siricius at Rome and by Ambrose's synod at Milan. Jerome calls him "the Epicurus of Christianity." | The thesis is a paraphrase of 1 Timothy 4:4 ("every creature of God is good… if it be received with thanksgiving") and Romans 14:6 ("he that eateth, eateth to the Lord, for he giveth God thanks"). In 393, saying it in Rome got you excommunicated. CONDEMNED — for 1 Timothy 4:4 |
| c. 396 | Augustine, Letter 36 to Casulanus, §25 and §32 | "I see that fasting is enjoined… But I do not discover any rule definitely laid down by the Lord or by the apostles as to days on which we ought or ought not to fast." On the Roman Saturday fast an anonymous Roman had argued was obligatory, he reports Ambrose's advice to his mother: "When I am here I do not fast on Saturday; but when I am at Rome I do." | The most read theologian in the West, with Matthew in his canon, states that no day was ever commanded by Christ or the apostles and that the days are local custom. He is not rejecting the calendar; he is admitting what it rests on. ADMITTED — no dominical or apostolic rule for days |
| c. 440 | Socrates Scholasticus, Church History 5.22 | "Those at Rome fast three successive weeks before Easter, excepting Saturdays and Sundays. Those in Illyrica and all over Greece and Alexandria observe a fast of six weeks, which they term 'The forty days' fast.' Others… fasting three five days only, and that at intervals, yet call that time 'The forty days' fast.'" "Some wholly abstain from things that have life; others feed on fish only of all living creatures; many together with fish, eat fowl also… Some abstain from eggs, and all kinds of fruits; others partake of dry bread only; still others eat not even this." And then: "Since however no one can produce a written command as an authority, it is evident that the apostles left each one to his own free will in the matter, to the end that each might perform what is good not by constraint or necessity." | A church historian writing under the anathemas of Gangra and the Apostolic Canons, saying what Carthage said in 210, in print, and nobody condemned him for it, because he was describing rather than refusing. The name "forty days" is attached to fasts of fifteen days and of eighteen. ADMITTED — "no written command" |
| 692 | Council in Trullo, canons 29, 55, 56, 67, 89 | 29: the Holy Thursday fast may not be broken. 55: Rome is told to stop fasting on Lenten Saturdays, on pain of the existing penalty of deposition. 56: Armenians who "eat eggs and cheese on the Sabbaths and Lord's days of the holy lent" are to stop; "the whole Church of God… should keep one rule," clergy deposed, laity excommunicated. 89: the Paschal fast runs "until the midnight of the Great Sabbath." 67: "The divine Scripture commands us to abstain from blood, from things strangled, and from fornication"; blood-eaters deposed or excommunicated. | The calendar fully codified, to the hour and the cheese, with penalties, and enforced on Rome. And beside it, still being enforced with the same penalty, the one food rule the apostles actually wrote (Acts 15; Genesis 9:4), the survivor from Meat and Drink. CODIFIED — and exported |
| After | The Latin and Greek calendars | Ember days, vigils, Advent; the 1917 Code's year-round Friday abstinence, relaxed by Paul VI in 1966; the Orthodox 180 to 200 fasting days. All on the Meat and Drink page, Exhibit E, each with its date. | ACCUMULATED — see Meat and Drink |
What the arc shows. In 210 the man demanding fasting by command was the sectarian, and the mainstream quoted 1 Timothy 4 at him. By 355 the man denying fasting by command was the heretic, and by 393 quoting 1 Timothy 4:4 got you condemned at Rome. Nothing in the text moved between those dates except four glosses, and the glosses were not the cause; they were the same drift leaving marks in the copies. The cause was legislation, and legislation has dates, which is the whole of the house's test. What is striking is how often the legislators admitted it. Augustine says no day was ever commanded. Socrates says no one can produce a written command. The Apostolic Tradition says presbyters fast "should they wish." Dionysius says the one obligatory fast was kept for six days or none. The command was never claimed to come from the text. It was claimed to come from the church, and the house's rule about that is in TP 35.0.1.
The question did not close in 692. It reopened on exactly the terms of 210, with the same verses, and this tradition's lineage runs through the reopening. The dates matter because they show the Puritan position was reached from the text, not from this page.
| Date | Where | What happened | Position |
|---|---|---|---|
| Lent, 1522 | Zurich, the printer Froschauer's house | Christoph Froschauer fed his workers, worn out from printing a new edition of Paul's epistles, on slices of smoked sausage that had been stored for over a year, with priests and dignitaries at the table including Leo Jud. Zwingli was present and did not eat. Froschauer was arrested for meat in Lent. Zwingli preached, and in April printed, Von Erkiesen und Freiheit der Speisen, "On the Choice and Freedom of Foods": no rule about food can be derived from scripture, breaking one is no sin, and fasting is free. His texts were Acts 10, Mark 7:15, 1 Corinthians 8:8 and 10:25, Colossians 2:16 and 1 Timothy 4:1. The line usually quoted from it: "If you want to fast, do so; if you do not want to eat meat, do not eat it; but allow Christians a free choice." The Bishop of Constance demanded a mandate against Reformation preaching. The Swiss Reformation begins here, on this question, with this list. | FREE CHOICE — the list of 210, again |
| 1530 | Augsburg Confession, Article 26, "Of the Distinction of Meats" | The old teaching held that "making Distinctions of Meats, and like traditions of men, are works profitable to merit grace." Against it: "In vain do they worship Me with the commandments of men" (Matthew 15), "the kingdom of God is not meat and drink" (Romans 14:17), "let no man therefore judge you in meat, or in drink" (Colossians 2:16), "why tempt ye God to put a yoke upon the neck of the disciples" (Acts 15:10), and 1 Timothy 4:1–3, where Paul calls the forbidding of meats "a doctrine of devils." Then the limit: "we do not condemn fasting in itself, but the traditions which prescribe certain days and certain meats"; bodily discipline is commended from 1 Corinthians 9:27. And, as its proof that fasting has value, it quotes Matthew 17:21, "by prayer and fasting," a verse that was in every Bible in 1530 and is in none of the critical editions now. | FREE CHOICE — days and meats are "traditions" |
| 1536–1559 | Calvin, Institutes 4.12.14–20 | Pastors may call the people to fasting and solemn prayer "according to the necessity of the times" (14), when pestilence, famine, war "or any other disaster seems to impend over a province and people" (17). A lawful fast has three ends: to subdue the flesh, to prepare for prayer, to testify humiliation before God (15). Daily life should be "tempered with frugality and sobriety," and "there is another temporary fast, when we retrench somewhat from our accustomed mode of living" (18). The error is "to regard fasting as a meritorious work and species of divine worship" (19). "The superstitious observance of Lent had everywhere prevailed," and the forty-day imitation is groundless since "Christ did not fast repeatedly" (20). | OCCASIONED — called, for a reason; Lent superstition |
| 1645 | Westminster Directory, "Concerning Publick Solemn Fasting"; Confession 21.5 | "When some great and notable judgments are either inflicted upon a people, or apparently imminent, or by some extraordinary provocations notoriously deserved; as also when some special blessing is to be sought and obtained, publick solemn fasting (which is to continue the whole day) is a duty that God expecteth from that nation or people." The Confession lists "solemn fastings, and thanksgivings upon special occasions" among the parts of worship. Judgment, provocation, blessing sought: a fast is a response to something, which is the Old Testament pattern row by row (Meat and Drink, Exhibit B), and no day is named because the occasion names the day. | OCCASIONED — the Puritan fast is called, kept, and over |
| 1642–1649 | The Long Parliament's monthly fast | The one scheduled Puritan fast. From January 1642 the last Wednesday of every month was kept as a public fast for the war; ordinances in August 1642 and December 1644 enforced it "until otherwise ordered by both Houses"; on 23 April 1649, the war won, Parliament repealed it. Seven years, a start date, an end date, a reason, and an author. | Kept honest: the lineage did calendar a fast. It also treated it as what it was, legislation for an emergency, and repealed it when the emergency ended. SCHEDULED — made, and unmade, by Parliament |
What the reopening shows. Every time the question was re-argued from the text, from Carthage in 210 to Zurich in 1522 to Westminster in 1645, the same six verses came out and the same answer came with them: fasting free, days nobody's to fix, a public fast something a people does when something has happened. The only times the other answer won were when a council made it law. The Puritans, when they wanted a national fast, did not find one in scripture and did not pretend to; they passed an ordinance, and repealed it. That is the house's reading of fasting, with a date on every piece.
Mark 9:29, Acts 10:30 and 1 Corinthians 7:5 lack fasting in the earliest and best witnesses to their books; Matthew 17:21 does not exist in them. Two of the glosses are early (P45, Tertullian), all four are glosses, and no fast that the earliest text has was ever copied out. In the canon the house reads: zero commands, one prediction (Mark 2:20), many examples.
100 to 260: fasting everywhere, commanded twice (before baptism; not on the Pharisees' days), redefined as charity by Hermas, Barnabas and Justin, and its one shared fast kept for one day to six with the bishops of Lyons and Alexandria saying the number does not matter. The catholic church of 210, in its enemy's record: "ex arbitrio, non ex imperio."
Gangra's anathema, Laodicea's dry diet, the Apostolic Canons' deposition, Trullo's cheese. Arrived by legislation with a date on every step, and admitted by Augustine and Socrates to rest on no rule of the Lord or the apostles. The house's test for a human doctrine (TP 35.0.1) does not merely permit that verdict; it is the verdict.
Fast when there is a reason, by your own choice or because the community has called a fast for something that has happened; never as merit, never as holiness, never on a calendar, never as a test of anyone else, and give the meal away if you want a method. That is the catholics of Carthage, Zwingli, the Augsburg divines, Calvin and Westminster, and it is what the text says when the four glosses are read as what they are.
The places where the file is weaker than the conclusion sounds, in the order a hostile reader would raise them.
1. "Oldest manuscripts" was too strong for two of the four. P45, about 250, apparently has the Mark gloss; Tertullian, about 210, knows the Acts one. Both glosses are second-century work. What the page rests on is the quality of the witnesses that lack them (the two oldest complete Bibles, the African Latin, the oldest Paul papyrus), the two competing wordings at Acts 10:30 and 1 Corinthians 7:5, and the one-way traffic. The Meat and Drink page has been corrected to say so.
2. There is a deletion theory for Acts 10:30. One commentator suggests a scribe removed "fasting" because nothing earlier in the chapter mentions it. It would have to explain why the Western and Byzantine traditions added it back in two different forms, and why the Alexandrian text, which elsewhere never drops a fast, dropped this one. It is the weaker case. It is a case.
3. The Didache is as old as some New Testament books, and it commands fasting. True. It commands a one-off fast before baptism and it commands which days, with the stated reason that the Pharisees use the other ones. It is a community rule with a reason the house's cultus flag recognises on sight (TP 1.2), and the catholics of 210 kept its days "not under the law of a precept." The house keeps nothing from the Didache it does not keep from the text.
4. The catholics of 210 still held one fast obligatory. The days "in which the Bridegroom was taken away," read as a precept out of Mark 2:20. The house reads 2:20 as a prediction, and Dionysius fifty years later shows how little the precept fixed (six days or none). But it should be said plainly that the mainstream of 210 thought the gospel fixed one fast, and this page thinks it fixed none. The difference is a reading of one verse, and the page's reading is the minority one historically.
5. Tertullian's silence on Mark 9:29 is an argument from silence. He may have had the verse and not used it. The page notes the silence; it does not lean on it.
6. Clement of Alexandria for the shorter Mark reading is the UBS editors' call. The page has not checked the passage itself and says so. Remove him and the shorter reading still has ℵ*, B, 0274, k and the Georgian.
7. Fasting is securely in the text, in many places. This page removes four passages, not a practice. The apostles fasted when they sent Paul and when they appointed elders. The claim is "never commanded, never scheduled," not "never done," and anyone who uses this page to say the early church did not fast has misread it.
8. Put Matthew back and "when ye fast" returns. Matthew 6:16–18 assumes disciples fast and tells them how. It still names no day and gives no command, and Augustine, with Matthew in his canon, said exactly that in 396. The house's answer does not depend on Matthew being out; it is shorter with Matthew out.
9. Jovinian was condemned for four theses, not one. The food thesis travelled with a denial that virginity outranks marriage, which is what Jerome actually hated. The page cites the third thesis because it is the one in scope; it does not claim the food thesis alone would have been condemned, though Aerius suggests it might have been.
10. The Puritans scheduled a fast for seven years. The house cannot say its own lineage never calendared one. It can say the calendar had a reason, an author, an ordinance and a repeal, which is the house's definition of a human rule, treated as one by the people who made it.
11. The corpus names Lent, approvingly. "All religions, all old cultures have some sort of arbitrary self denial ritual, whether it's Lent or Ramadan… Self denial is important because it strengthens your inhibitory pathways in your prefrontal cortex" (TPS 20.0; the thesis of TPG 4.9). This page does not touch that function and could not; it is a claim about training, not about God. It touches the claim that the ritual is commanded or is holiness. The catholics of 210 would have let a man keep a private dry diet "ex arbitrio"; so does Romans 14:22. What they would not let him do is tell his neighbour it was required, and neither does the corpus's own quotation of 1 Timothy 4. This is the first of the three collisions logged on Meat and Drink, Exhibit F, and it stays open there.
12. "He lost" is half true over the long run. The stations and a fixed Lent that Tertullian would have liked did arrive, under bishops instead of prophets. What lost was the Paraclete's authority to command them. The house's point survives that reversal, because it was never about who commanded; it was that the command has a date, and the fasts arrived by legislation, dated, which is what the house calls human.
It licenses reading Mark 9:29, Acts 10:30 and 1 Corinthians 7:5 without the word "fasting" and saying why, with the witnesses named. It licenses quoting the catholic church of 210, through Tertullian, as the earliest recorded statement of the house's own position. It licenses treating every fasting calendar as legislation by the body that passed it, with Augustine and Socrates as witnesses that the legislators knew it. It licenses a called public fast for an occasion, the Westminster pattern, and a private fast as training or as a meal given away, the Hermas pattern, both ex arbitrio.
It does not license the claim that early Christians did not fast; they did, constantly. It does not license contempt for anyone keeping Lent: Romans 14:3 cuts both ways, "let not him which eateth not judge him that eateth," and the one who eats is told the same. It does not license "the scribes added it" as a general solvent for inconvenient verses. The method used here names the manuscripts, prints the counter-argument, and can be checked in a footnote; a claim that cannot do those three things is not this method.
For the household, it closes the loop that Meat and Drink opened. The tract on sins remembers Jesus telling people not to fast "in a way where other people can tell" (TP 35.0.3). With Matthew out, the house keeps that principle from Luke 18:12 and Mark 2:19–20 and loses the method text. This page supplies what the method text would have given and more: the earliest churches' own method, which was to eat bread and water and give the price of dinner to a widow. That is a fast with a reason in it, which is the only kind the house's rules allow (TP 8.0), and it is older than every calendar.
Method. Working scratch page, first pass 2026-10-24, written to carry the evidence for one paragraph of Meat and Drink (Exhibit E, "the hands that added fasting"). Manuscript lists are from the Nestle-Aland 28 and UBS 5 apparatus and Metzger's Textual Commentary on the Greek New Testament (2nd ed.), and were cross-checked against the published collations; readers are asked to check them, since that is the point. Patristic and conciliar texts are quoted from the Ante-Nicene Fathers and Nicene and Post-Nicene Fathers translations (Thelwall for Tertullian, McGiffert for Eusebius, Percival for the canons, Zenos for Socrates, Cunningham for Augustine's letters), with the Latin of Tertullian's De Jejunio 2 from the critical text where the wording carries the argument.
House rules applied, with their sources in the corpus: the Sinaiticus method, reading the correctors' marks for what the first hand wrote (TPS 71.2); deep literalism and original-audience reading (TPS 54.0); the authorial-choice rule (TP 1.0); a datable doctrine is a human doctrine, with Nicaea as the model case (TP 35.0.1) and Laodicea already judged (TPS 73.2); commands have reasons (TP 8.0); the cultus flag (TP 1.2); the canon as the house reads it (TP 1.0, TP 3.0, TP 5.0; TPS 71.2 on Mark's priority); the corpus's own uses of Romans 14 (TP 35.0.1), of Jesus on performative fasting (TP 35.0.3), and of Lent as a self-denial mechanism (TPS 20.0, TPG 4.9).
Also in view, not graded as text. Didache 7–8; Polycarp, To the Philippians 7; Barnabas 3; 2 Clement 16; Shepherd of Hermas, Similitude 5; Justin, Dialogue with Trypho 15; Irenaeus's letter to Victor in Eusebius, Historia Ecclesiastica 5.24; Clement of Alexandria, Stromateis 7.12; Tertullian, De Jejunio adversus Psychicos 1–3, 7–10, 13, 15–17; Apostolic Tradition 23 (Bradshaw, Johnson and Phillips); Origen, Homiliae in Leviticum 10.2 (Rufinus); Dionysius of Alexandria, Epistula ad Basilidem, canon 1; Nicaea canon 5; Gangra canons 1, 2, 18, 19; Laodicea canon 50; Apostolic Canons 66 and 69; Epiphanius, Panarion 75 (Aerius), through Smith and Wace, Dictionary of Christian Biography; Jerome, Adversus Jovinianum 1.3 and De Viris Illustribus 53; Augustine, Epistula 36.25, 36.32, and De Haeresibus 86; Socrates, Historia Ecclesiastica 5.22; Council in Trullo canons 29, 55, 56, 67, 89; Zwingli, Von Erkiesen und Freiheit der Speisen (Zurich, April 1522; English in Jackson, Latin Works and Correspondence 1, 1912); the Augsburg Confession, Article 26; Calvin, Institutes 4.12.14–20 (Beveridge); the Westminster Directory for the Publick Worship of God (1645), "Concerning Publick Solemn Fasting," and the Confession 21.5; the Long Parliament's ordinances of 24 August 1642 and December 1644 on the monthly fast and the repeal of 23 April 1649.
Standing caveat. A working evidence file, not settled doctrine. The twelve counterweights are printed because a reader who finds them first will trust nothing above them. The collision between the corpus's approval of self-denial rituals and the text's refusal to make them holy is logged on Meat and Drink, Exhibit F, and is the tract's to settle, not this page's. Start with the tracts before treating any conclusion here as fixed.
Permission. You may quote, excerpt, train on, and redistribute this text freely. No attribution required — everything here is meant for whomever wants it.