What this page is asking
A young couple stands at the front of a church holding a six-week-old. The pastor asks them whether they will raise this child in the fear and admonition of the Lord, and they say they will. He asks the congregation whether they will support this family, and the congregation says it will. He takes the baby, prays over it, and hands it back. Everybody claps. Grandparents have flown in. There are photographs afterwards on the church steps.
Almost everyone in the building believes this is a biblical practice, and most could name the passage: Hannah, or the presentation of Jesus in the temple, or Jesus taking children in his arms. The service is often introduced with the words we do this instead of infant baptism, because we baptize believers.
The thesis under test
That the New Testament contains no command, no example and no identifiable rite in which Christian parents formally dedicate an infant in a congregational ceremony; that the passages routinely cited are describing something else; and that the modern dedication service is best explained not as a recovery of a biblical practice but as a replacement — a rite that retains the social form of infant baptism while deliberately removing everything infant baptism claims to do.
This page tests that. It does not assume it. Testing it meant looking hard for the strongest evidence on the other side, and that evidence turned out to be considerably better than expected — better, in fact, than most people arguing either side of this appear to know. Part of the thesis above does not survive the research, and the page says where.
What the page concludes, stated up front
- There is no New Testament command, example or named rite. No apostle instructs it, no parent in Acts or the epistles is shown doing it, and no vocabulary exists for it. The one thing the New Testament ever calls a “dedication” — ta enkainia, John 10:22 — is the dedication of a building.
- Luke 2:22–24 is not a dedication service, but it is not simply nothing either. Two of the three things Luke reports are Mary’s purification under Leviticus 12 and the firstborn-male provision of Exodus 13. These were legal obligations, not parental initiatives. But Luke conspicuously does not record the five-shekel redemption payment that Numbers 18 requires, and he tells the whole scene in language borrowed from 1 Samuel. Something is being said about Jesus being given. That is the strongest thing on the dedication side of the ledger, and Part II sets it out honestly.
- Hannah is a vow, not a rite. One woman, one child, an unrepeatable set of circumstances, a Nazirite-shaped promise binding for life — and she left him at the sanctuary and went home without him. Every element that makes the modern service what it is (the congregation, the parental promises about upbringing, the child going home in the car seat) is absent, and the one element that makes Hannah’s act what it is (permanent surrender of the child) is absent from the modern service.
- The early church has no infant-dedication rite offered as an alternative to baptism — but it does have a rite for unbaptized infants, and this page gives up part of its thesis there. Children of Christian parents whose baptism was deferred were signed with the cross, given salt, and enrolled as catechumens. Augustine describes it happening to him at birth. A 2022 article in a Baptist journal argues that this amounts to infant dedication, and it is a serious argument. Part VI states it at full strength before answering it.
- The continuity claim is strong but not airtight. From roughly the fifth century to 1525 there is no organized Christian body in East or West that did not baptize infants. But the fourth century is a real exception: Basil, both Gregories, Chrysostom, Ambrose, Jerome and Augustine were all children of Christian households and none was baptized as an infant. “Unbroken and universal” is too strong for that window, and the window matters.
- The modern service is simultaneously much older and much newer than either side usually says. Balthasar Hubmaier describes gathering the congregation, presenting the child, expounding Matthew 19 and praying on bended knee — in a letter dated 16 January 1525, before the first believers’ baptisms in Zurich, and before his own. The rite is therefore not a twentieth-century invention. But the printed order of service is: it enters the Mennonite conference handbooks between 1830 and 1843, the Church of the Nazarene Manual in 1936, and the United Brethren Discipline in 1945, under the title “Dedication of Infants (Without Baptism).”
- The asymmetry is the real finding. Set the two rites side by side and what dedication keeps from infant baptism is the entire social architecture — the front of the church, the infant held up, the officiant, the questions, the answers, the congregation, the prayer, the register. What it removes is every claim: the water, the Triune name, the assertion that God is doing anything, membership, and irreversibility. The 1945 United Brethren service was commissioned to accompany legislation establishing a Children’s Membership Record. They built the register too.
The prompt behind this page
Everything below was produced by putting the following brief to a frontier language model with web search, which assembled and verified the sources. It is published so you can run it yourself, change it, sharpen it, or reverse it and see what comes back. If you get a materially different answer, that is information, and worth publishing too.
Write a detailed, historically grounded, biblically rigorous and well-sourced
investigation of the modern Christian practice of "baby dedication" or "child
dedication" - the congregational ceremony in which parents present an infant,
make promises, and receive prayer, in churches that do not baptize infants.
Test - do not assume - this thesis:
There is no New Testament command, example, or identifiable rite in which
Christian parents formally dedicate an infant or child to God in a
congregational ceremony. The modern dedication service is not a recovery
of a biblical practice but a replacement for infant baptism: it retains
the social form of the older rite while deliberately removing its
sacramental and covenantal claims.
Cover, in this order:
1. The New Testament evidence, worked from the texts themselves. Is there a
command? An example? A named rite? Examine Mark 10:13-16 and parallels
(Matt 19:13-15; Luke 18:15-17), Acts 2:39, 1 Cor 7:14, Eph 6:4, 2 Tim
1:5 and 3:15, Luke 1:59, and the household baptisms (Acts 16:15, 16:33,
18:8; 1 Cor 1:16). State plainly what each does and does not establish.
2. Jesus' presentation in Luke 2:22-24. Determine whether this passage
describes a dedication of Jesus or compliance with (a) Mary's
purification under Leviticus 12 and (b) the Mosaic law concerning the
firstborn male in Exodus 13 and Numbers 18. Consult the Greek. Note the
textual variant at Luke 2:22 ("their" vs "her" purification). Address
directly the fact that Luke records no five-shekel redemption payment
(Num 18:15-16) and that his language echoes 1 Samuel 1-2. Do not resolve
this in whichever direction the thesis needs; report the strongest case
on each side.
3. Old Testament "dedication" precedents. Test the examples commonly cited -
above all Hannah and Samuel (1 Sam 1), and also Samson (Judg 13) - to
determine whether they establish a general rite for children or describe
exceptional vows involving lifelong sanctuary service or Nazirite
separation. Work from the Hebrew, including the sha'al wordplay in
1 Sam 1:20, 27-28. Ask what Hannah actually did with the child
afterwards.
4. Vocabulary. Determine whether modern baby-dedication terminology is being
read backwards into biblical passages whose historical and covenantal
meaning was substantially different. Examine hanak, qadash, nazar,
and the Greek paristemi, anatithemi and enkainia. Ask what word the New
Testament actually uses for "dedication" and what it is applied to.
5. Early-church evidence. Search the first five centuries for any distinct
Christian infant-dedication ceremony offered as an alternative to
baptism. Search church orders, liturgies, canons, homilies, catechetical
lectures, funerary inscriptions and the fathers. Compare the documentary
footprint with the evidence for infant and child baptism. Engage
seriously with the argument that the making of a catechumen - the
signing with the cross, the salt, the imposition of hands, described by
Augustine in Confessions 1.11.17 - constitutes an infant dedication.
Look up Andrew Messmer, "Infant Dedication in the Early Church",
Journal of European Baptist Studies 22:2 (2022), and state his case at
full strength before answering it.
6. Historical continuity of infant baptism. Evaluate the strength of the
claim that infant baptism was practised throughout most of Christian
history - continuously enough that abandoning it created a ritual vacuum
in churches rejecting paedobaptism. Test the claim against the
fourth-century pattern of deferred baptism (Basil, the Gregories,
Chrysostom, Ambrose, Jerome, Augustine) and against medieval dissenting
movements. Say clearly where the claim is too strong.
7. Origin of the modern service. Trace when a recognizable service - parents
bringing an infant before the congregation, making promises, receiving
prayer, with baptism explicitly postponed until personal profession -
actually became common. Establish the earliest documented instance. Check
Balthasar Hubmaier's letter to Oecolampadius of 16 January 1525, Pilgram
Marpeck's confession of 1531, the Nordlingen agreement, the Palatinate
and Hesse Mennonite conference decisions of 1830-1843, the Church of the
Nazarene Manual of 1936, and the United Brethren Discipline of 1945.
Distinguish the practice from the printed order of service; they have
different dates.
8. Relationship to credobaptist theology. Investigate whether dedication
developed specifically because churches rejecting infant baptism still
perceived a pastoral need to mark a birth, name parental responsibility
and involve the congregation. Look for what nineteenth-century Baptists
actually did instead - consult Spurgeon's sermon 581, "Children Brought
to Christ, and Not to the Font" (1864), and the Baptist church manuals
of the period.
9. Theological function comparison. Compare what the rites actually claim to
accomplish, tradition by tradition: Catholic, Orthodox, Lutheran,
Reformed, Anglican, Methodist, and credobaptist dedication. For each,
identify who the principal actor is, what is claimed to be conferred,
whether the rite is repeatable, whether it confers membership, and what
the parents are promising. Historic infant baptism is ordinarily
understood as God acting toward the child - with different traditions
explaining that differently - whereas modern dedication usually
emphasizes the parents' commitment.
10. The form question. Determine whether dedication retains the social form
of infant baptism while removing its sacramental and covenantal claims.
Compare the physical and liturgical elements one by one: location,
officiant, posture, questions, congregational participation,
registration, certificates, naming, timing after birth. Then ask
whether the traffic runs the other way too - whether infant baptism in
some paedobaptist churches has itself become a dedication in practice.
11. Construct the strongest case against the thesis before concluding: that
an argument from silence proves too much; that Christians rightly do
many things without New Testament command; that Jesus laid hands on
infants and blessed them; that the practice is five hundred years old
rather than modern; that dedication addresses the parents, whom the New
Testament actually does command; and that infant baptism's own
documentary footprint before AD 200 is thinner than its defenders
usually concede. Give it every reasonable opportunity, then evaluate it.
12. End on the pastoral question. If a couple in a credobaptist church asks
what they should do when a child is born, what is the honest answer -
and what should the church say the ceremony is, and is not?
Requirements: quote primary sources verbatim with citations a reader can
check, and prefer editions that are free online. Distinguish what the record
establishes from what it does not. Where the evidence undercuts the thesis,
say so on the page rather than working around it. Treat both paedobaptist
and credobaptist positions as held by serious people for serious reasons,
and do not caricature either. Make no claim about the sincerity or standing
of anyone who practises either rite.
If you re-run this, the interesting experiment is not whether you get the same conclusions. It is whether you get the same primary sources — and in particular whether the model finds the January 1525 letter, which almost everything written on this subject in English appears to have missed.
What do you currently think?
Reading changes very few minds. Noticing that you were wrong about something changes rather more. So before the evidence: seven questions. Answer them honestly — nothing is scored yet and nothing is sent anywhere. You will see your answers again at the end, next to what the sources say.
Your answers stay in this browser tab. Nothing is transmitted.
What follows
What the New Testament contains
Start with the negative, because it is unusually clean. Search the New Testament for a ceremony in which Christian parents bring a child before the assembled church and commit that child to God. There is no command to do it. There is no instance of it being done. There is no word for it. There is no letter correcting how it was being done, which is often the surest sign that a first-century practice existed at all — Paul corrects the Lord’s Supper, head coverings, tongues, collections, lawsuits, meat, widows’ lists and the order of a service, and never mentions this.
That is worth stating carefully, because it is easy to overstate. The New Testament also contains no command to hold a wedding in a church, no order of service for a funeral, no instruction to build a building, and no Sunday School. Silence is not by itself a verdict. The question is not whether the practice is commanded; it is whether the passages routinely produced as its warrant actually contain it. Six passages do nearly all the work.
Mark 10:13–16 and parallels
“And they brought young children to him, that he should touch them: and his disciples rebuked those that brought them. But when Jesus saw it, he was much displeased, and said unto them, Suffer the little children to come unto me, and forbid them not: for of such is the kingdom of God… And he took them up in his arms, put his hands upon them, and blessed them.”
Mark 10:13–16, King James Version. Parallels at Matthew 19:13–15 and Luke 18:15–17, where Luke uses brephē, infants.
This is the strongest text on the blessing side, and it is genuinely strong: Jesus takes infants, lays hands on them, and blesses them. A church that lays hands on an infant and blesses it is imitating something Jesus did. That much is real, and this page does not take it away.
What the passage does not contain is everything that makes a dedication service a dedication service. No parent makes a promise. No congregation is asked anything. Nothing is dedicated. Nobody is enrolled. The one thing Jesus is recorded as saying about the children concerns the kingdom of God, not the obligations of their parents. It is also, notoriously, the favourite proof text of the other side: paedobaptists have cited this passage for infant baptism since at least the third century, with as little textual warrant. Spurgeon’s verdict on that usage, delivered from the Metropolitan Tabernacle pulpit in 1864, is worth having in front of you, because it applies with exactly equal force in both directions:
“NOT THE SHADOW OF THE SHADE OF THE GHOST OF A CONNECTION WITH BAPTISM.”
C. H. Spurgeon, “Children Brought to Christ, and Not to the Font,” sermon no. 581, preached 24 July 1864 on Mark 10:13–16. Capitals are Spurgeon’s own, as printed.
The other four
Acts 2:39 — “For the promise is unto you, and to your children, and to all that are afar off.” A statement about the scope of the gospel promise, produced in the middle of a sermon whose imperative is repent and be baptized. Used by paedobaptists for covenant continuity and by credobaptists for the universality of the call. It is not about a ceremony.
1 Corinthians 7:14 — “else were your children unclean; but now are they holy.” Paul’s subject is a mixed marriage and whether the believing spouse should leave. The children appear in an argument about the unbelieving spouse, and whatever “holy” means here it is predicated of them already, without any rite being performed.
Ephesians 6:4 — “bring them up in the nurture and admonition of the Lord.” This is the clearest New Testament instruction to Christian parents about children, and it is the substance of what a dedication service asks parents to promise. It is addressed to fathers, continuously, for the length of a childhood. It prescribes no event.
2 Timothy 1:5, 3:15 — Timothy’s grandmother, his mother, and his knowing the scriptures “from a child.” The New Testament’s one extended portrait of a child raised in the faith. It describes a household and a lifetime of teaching. No ceremony is mentioned at either end of it.
The pattern is consistent and it is the same pattern in every case: the New Testament’s interest in Christian children is durative. It concerns what happens over years in a house. The modern service is punctiliar: it concerns what happens in four minutes on a Sunday. Neither is wrong, but they are not the same kind of thing, and the second is not derived from the first.
The one candidate nobody wants
If you are looking in the New Testament for an occasion on which infants in a Christian household underwent a religious act at the initiative of a parent, there is exactly one class of candidate: the household baptisms — Lydia’s (Acts 16:15), the Philippian jailer’s (Acts 16:33), Crispus’s (Acts 18:8), Stephanas’s (1 Corinthians 1:16). Whether any infants were present in those households is exactly the disputed question, and this page does not pretend to settle it. But note what the argument is about. Nobody, on either side, has ever proposed that the household passages describe a dedication. The only rite in the vicinity is baptism.
Interactive: the presentation, line by line
Luke 2:22–24 is the passage. It is printed in the bulletin, read from the platform, and named in the certificate. It is worth reading it against the four Mosaic texts it is actually quoting and alluding to. Turn the highlights on and off and watch which parts of Luke correspond to which law.
Three observations follow from the text, and only the third is contested.
First, the trip to Jerusalem is dated by Mary’s ritual calendar, not the child’s. Leviticus 12 gives a mother of a son forty days; Luke starts the clock there. The offering Luke records — a pair of turtledoves or two pigeons — is the concession in Leviticus 12:8 for a woman who cannot afford a lamb. It is the purification offering. It has nothing to do with the child at all.
Second, the words “present him to the Lord” are Luke’s summary of Exodus 13, which he then quotes: every male that openeth the womb shall be called holy to the Lord. That statute is not a parental option. It is a claim God makes on every Israelite firstborn male at the Exodus, and it applied to every family in Israel whether they wanted it or not. Reading it as a voluntary act of consecration by Joseph and Mary inverts it.
Third — and this is where the honest reading gets interesting — Luke never records the payment.
The five shekels that are not in Luke
Numbers 18:15–16 is explicit: the firstborn of man shall be redeemed, from a month old, for five shekels of the sanctuary. That is what an Israelite family did with a firstborn son. They bought him back. The whole legal machinery exists so that the child does not stay with the sanctuary — the Levites having been substituted for the firstborn wholesale in Numbers 3.
Luke describes the journey, quotes the consecration clause, and reports the purification offering. He does not report the redemption. And the vocabulary he uses of Jesus in the temple — and of Simeon, and of Anna who “departed not from the temple” — is saturated with 1 Samuel 1–2, down to the canticle Mary sings, which is patterned on Hannah’s. Something is being said here about a firstborn who is not bought back.
Read one way
Luke assumes the redemption and does not bother to mention it, exactly as he does not mention the circumcision fee or who carried the birds. He is compressing. Ancient narrative omits routine legal detail constantly, and the passage is explicitly framed three times as doing “according to the law of the Lord.” On this reading nothing whatever is being suggested about dedication; Luke is establishing that this family kept Torah.
What it costs: it makes the 1 Samuel echoes accidental, which is hard to sustain, given that Luke has just written the Magnificat.
Read the other way
Luke omits the redemption deliberately, because he is presenting Jesus as a firstborn who is given and not reclaimed — a new Samuel, handed over to God permanently. On this reading there is a giving in Luke 2, and it is the strongest biblical footing the dedication tradition has ever had.
What it costs: the thing being described is unique to Jesus by definition. A parent who reads Luke 2 this way and then applies it to their own child is claiming their child is not to be redeemed either — which is precisely what the modern service does not mean, and would be alarming if it did.
This page does not need to choose. Either way, the passage will not carry a general rite. On the first reading it is Torah compliance and describes nothing repeatable. On the second reading it is Christological and describes nothing repeatable. The one thing it is not is a template for what a Christian family does forty days after a birth.
A note on the text: most modern critical editions read tou katharismou autōn at Luke 2:22 — “their purification” — while the Textus Receptus behind the KJV reads autēs, “her purification.” The variant does not affect the argument here, but it is visible in any parallel Bible and worth seeing once.
Hannah, and what she did next
1 Samuel 1 is the other passage, and it is quoted in dedication services more than any other. It is worth reading past verse 28.
“And she vowed a vow, and said, O LORD of hosts, if thou wilt indeed look on the affliction of thine handmaid, and remember me, and not forget thine handmaid, but wilt give unto thine handmaid a man child, then I will give him unto the LORD all the days of his life, and there shall no razor come upon his head.”
1 Samuel 1:11, KJV. The clause about the razor is the language of the Nazirite separation in Numbers 6:5. The Septuagint and the Qumran manuscript 4QSama have a fuller form of the vow, adding that he shall drink neither wine nor strong drink.
Six things are true of Hannah’s vow, and each of them is absent from the modern service that cites it.
| In 1 Samuel 1 | In the modern service |
|---|---|
| A conditional bargain made before conception, in desperation, by a woman who had been childless for years | An act performed after a healthy birth, unconditionally, as a normal part of church life |
| A Nazirite separation binding the child for life — a specific legal status under Numbers 6 | No status of any kind is conferred or claimed |
| Made by the mother alone; Elkanah’s role is to decide not to annul it (1 Sam 1:23, cf. Numbers 30:6–8) | Made by both parents together, and often required to be |
| Performed at the sanctuary at Shiloh, before one priest, with no congregation present or involved | Performed before the assembled congregation, whose participation is the point |
| Sealed with a sacrifice: a three-year-old bullock, an ephah of flour, a bottle of wine | No sacrifice |
| She left him there. “And he worshipped the LORD there” — and Hannah went home to Ramah, and thereafter brought him a little coat once a year | The family goes home together, which is the entire assumption of the ceremony |
The last row is the one that matters. Whatever Hannah did, the modern service is not a smaller version of it. It is a different act in every material respect, and the resemblance rests almost entirely on a shared English word.
The Hebrew, briefly
The chapter turns on a pun that English cannot carry. Hannah asked — שָׁאַל sha’al — and names the child Samuel, שְׁמוּאֵל, explaining “because I have asked him of the LORD.” Then at 1:28 she says הִשְׁאִלְתִּהוּ לַיהוָה, hish’iltihu la-YHWH: I have lent him, or handed him over as one asked for, to the LORD — and adds “as long as he liveth he shall be lent to the LORD.” The word for what she does with the child is built from the word for what she wanted. The transaction is total and it is permanent. There is no vocabulary here for a temporary, symbolic, or renewable giving, because nothing of that kind is happening.
Samson (Judges 13) is the other case, and it makes the same point more sharply: the Nazirite status is announced by an angel to the parents before conception, is binding on the child from the womb, and again is lifelong. Two extraordinary births, two lifelong separations, two children who do not live a normal life. This is not a body of precedent for a general practice. It is a pair of prodigies.
Interactive: does this text establish a rite?
Ten passages, each of which is cited somewhere in print as biblical warrant for dedicating an infant. For each, judge one question and one question only: does this text establish a repeatable ceremony that Christian parents are to perform for a child? Answer all ten, then reveal the analysis.
The word “dedicate”
Much of the confusion in this area is lexical, and it is the kind of confusion that only exists in translation. English has one broad word, dedicate, that covers a spectrum from “this book is dedicated to my wife” to “the temple was dedicated with a hundred and twenty thousand sheep.” The biblical languages do not have that word. They have several narrower ones, and they are not applied to children.
None of this means a church may not use the English word. It means that when a service is announced as “the biblical practice of dedicating children,” the biblical warrant being appealed to is an artefact of translation. The Hebrew and Greek behind the passages cited are doing something else in every instance: legislating a divine claim on the firstborn, describing a lifelong Nazirite vow, or naming a temple festival.
The early church, and where this page gives ground
The question for the first five centuries is narrow and answerable. Is there a Christian ceremony in which parents formally dedicate an infant, offered instead of baptizing it? Search the church orders, the liturgies, the conciliar canons, the catechetical lectures, the homilies, the funerary inscriptions and the fathers. There is not.
What there is, in the same period, is a substantial and increasingly unambiguous record of infants and small children being baptized — Irenaeus around 180, Tertullian around 203 (objecting to it, which is how we know it was happening), the Apostolic Tradition, Origen around 244 saying flatly that the church received the practice from the apostles, Cyprian and sixty-six African bishops in 253 ruling that it must not even wait until the eighth day. Dossier 01 sets that record out in full with the texts, and this page does not repeat it.
But the honest question is not whether infant baptism is attested. It is whether the absence of dedication is as clean as it looks. It is not, and the best argument that it is not deserves to be stated properly.
The case that the early church did have an infant rite
In 2022 the Journal of European Baptist Studies published an article by Andrew Messmer arguing that early Christians incorporated the unbaptized children of Christian parents through a two-stage practice: a dedication shortly after birth, followed by enrolment in the catechumenate, with baptism deferred until later. The evidence he assembles is real, and the first exhibit is Augustine describing his own infancy.
“I was signed with the sign of the cross, and was seasoned with His salt even from the womb of my mother, who greatly trusted in You.”
Augustine of Hippo, Confessions I.11.17. Augustine goes on to describe how, when he fell dangerously ill as a boy, his mother arranged for his baptism — and then, when he recovered, it was deferred. He was baptized at thirty-two.
That is not baptism. It is a distinct rite, performed on an infant, by the church, with lasting effect on his status — and Augustine considers himself to have been, from that moment, in some real sense a Christian. Messmer adds Gregory of Nazianzus, who says he was “dedicated” immediately after birth and baptized around the age of thirty; Basil’s language about catechumens being conceived but not yet born; Cyril of Alexandria on the chrism of the catechumenate given to newborns; Jerome on being reared on Catholic milk; and Siricius on children vowed to the church’s service from infancy but not baptized until puberty. And he points out something that anyone can check and almost nobody does: six of the eight figures traditionally called Doctors of the Church were not baptized as infants, and several were children of devout Christian parents.
Gregory of Nazianzus is the clearest witness to what the fourth century actually thought about timing:
“Certainly, if any danger presses. For it is better that they should be unconsciously sanctified than that they should depart unsealed and uninitiated… in respect of others I give my advice to wait till the end of the third year, or a little more or less, when they may be able to listen and to answer something about the Sacrament.”
Gregory of Nazianzus, Oration 40 (On Holy Baptism), 28. Answering the question of when infants should be baptized.
This page accepts the substance of that case. There was a rite for the infants of Christian families that was not baptism. It involved the cross, salt, the laying on of hands, and prayer. It was widespread across Carthage, Alexandria, Constantinople and Milan. It was performed in infancy. And any credobaptist who says “the early church had nothing for babies but baptism” is factually wrong.
Why it is still not the modern service
Three differences, and they are not quibbles.
1. It pointed at baptism rather than away from it. Making a catechumen is the first step of the baptismal process. The sign of the cross and the salt were the opening moves of a rite of initiation whose destination was the font; the catechumenate is defined by what it leads to. The modern dedication service is defined by what it replaces. A church that enrolls an infant as a catechumen is saying this child will be baptized. A church that dedicates an infant is saying this child will decide later.
2. It happened in churches that also baptized infants. Nobody in the fourth century held that infants must not be baptized. Gregory advises waiting until about three — unless there is danger, in which case baptize at once. Augustine’s mother tried to have him baptized as a boy. The deferral was a matter of prudence about post-baptismal sin, not a doctrine about who is eligible, and the moment the Pelagian controversy forced the question the deferral collapsed. It is an argument for delayed baptism, not for believers’ baptism.
3. Nobody promised anything. The modern service is constituted by parental vows. The fourth-century rite is done to the infant by the church: signing, salt, hands, prayer. Where promises for infants do appear in the early record they appear in connection with baptism — and Tertullian is already complaining about them in 203, which tells you the sponsors were promising on behalf of the baptized.
Tertullian is worth a last word here, because he is the first person in surviving Christian literature to argue that infants should not be baptized, and he is therefore the first person who had the opportunity to propose a substitute. He did not propose one. His alternative was delay and instruction:
“let them ‘come’ while they are growing up; let them ‘come’ while they are learning, while they are learning whither to come; let them become Christians when they have become able to know Christ. Why does the innocent period of life hasten to the ‘remission of sins’?”
Tertullian, On Baptism 18, c. AD 203. In the same passage he objects that sponsors are “thrust into danger” because they may fail to fulfil their promises — the earliest surviving complaint about godparents.
The first credobaptist argument in Christian history proposes catechesis, not a ceremony. That is a real datum, and it holds for the next thirteen hundred years.
Interactive: weigh the documentary footprint
Historical arguments about early Christian practice are arguments about evidence, and evidence has types. Some kinds carry much more weight than others: a liturgical rubric telling a bishop what to do is worth more than a preacher’s metaphor, and an opponent’s complaint is worth more than a supporter’s claim. Below are the categories of evidence that exist for the first five centuries. For each, decide which practice it attests. Then reveal what is actually in the record.
How continuous was infant baptism?
The claim that matters for this page is not that infant baptism is apostolic. It is the weaker and much better attested claim that infant baptism was practised so continuously, and so nearly universally, that abandoning it left a hole in the life of a church — a moment in a family’s life for which there was suddenly nothing to do. Test that claim in four parts.
Before 200: thin
Everett Ferguson, whose Baptism in the Early Church is the standard modern survey and who is himself from a tradition that does not baptize infants, puts it plainly: “There is general agreement that there is no firm evidence for infant baptism before the latter part of the second century.” The paedobaptist case for the first century rests on the household baptisms and on inference. Anyone who tells you it is documented is overstating.
200 to 400: attested, contested only as to timing
Anthony Lane’s 2004 study in the Tyndale Bulletin makes the observation that decides the question for this page. Across the whole period in which baptismal timing was disputed — and it was disputed, vigorously — “no one claimed that anyone else’s practice was unapostolic or wrong in principle.” Tertullian wants delay. Gregory wants three years. Cyprian’s council wants the second day. Nobody says infants are ineligible. There is no party in the ancient church that corresponds to the modern credobaptist position, and therefore no ancient church that needed a substitute rite.
The fourth-century exception, which is real
This is the part of the continuity claim that does not hold, and it is worth being precise about it. Through much of the fourth century, deferring baptism — often until adulthood, sometimes until the deathbed — was normal among the devout and the well-connected. Constantine was baptized as he was dying. Basil, Gregory of Nazianzus, Gregory of Nyssa, John Chrysostom, Ambrose, Jerome and Augustine were all raised in Christian or partly Christian households and were all baptized as adults. For perhaps a century, in significant parts of the empire, a large number of Christian children were not baptized as infants — which is exactly why the catechumenal rite described in Part VI mattered so much. The pattern reverses after Augustine and the Pelagian controversy, and by the fifth and sixth centuries infant baptism is, in Ferguson’s words, “the usual practice.”
500 to 1525: as close to universal as anything in church history
For roughly a millennium there is no organized Christian body, East or West, that did not baptize infants. The medieval dissenting movements are sometimes produced as counterexamples; the honest position is that the sources for what Waldensians, Cathars, Lollards and Hussites actually did with babies are fragmentary, hostile, or both, and that the claim of a continuous underground of believers’ baptism from the apostles to the Reformation is a nineteenth-century construction that its own best historians have abandoned. Whatever the exceptions, they are exceptions.
So the vacuum is real. When the Swiss Brethren rejected infant baptism in January 1525, they were not removing an optional extra. They were removing the one thing the church had done for a newborn for a thousand years — a public, dated, recorded event, attended by family, marking the arrival of a child into the community. That hole had to be filled, and the striking thing is how fast it was.
Interactive: where the service came from
The chronology is the heart of this page, and it does not say what either side expects. Click any entry to open it.
The five days that matter
The first believers’ baptisms of the Radical Reformation are conventionally dated to 21 January 1525, in Felix Manz’s house in Zurich, when Conrad Grebel baptized George Blaurock. Balthasar Hubmaier’s letter to Oecolampadius describing his practice of presenting infants to the congregation instead of baptizing them is dated 16 January 1525 — five days earlier, and months before Hubmaier himself was baptized at Easter.
If that dating is right, and it is the date given in the standard Mennonite reference work, then the substitute rite is not a later development at all. It arrives with the rejection itself, or fractionally ahead of it. Nobody had to wait for the vacuum to be felt. The man who created it had already designed the thing to fill it.
Then three centuries of very little
What follows is the part of the story that nobody tells, and it is the part that decides whether “replacement” is the right word. Hubmaier’s practice does not become a liturgy. It survives among continental Anabaptists as a custom — the standard Mennonite reference describes it as “an old, though not universally observed custom” — and it is not until 1830–1843 that Mennonite regional conferences in the Palatinate and Hesse formally adopt it and put forms for it into printed handbooks.
In the English-speaking Baptist world it is later still, and the negative evidence is worth weighing. When Spurgeon preached in 1864 on the very passage that dedication services now use, and set out at length what Baptists should do with their infants instead of bringing them to the font, he described no ceremony at all:
“We take them up in the arms of our prayer. I hope many of us, so soon as our children saw the light, if not before, presented them to God with this anxious prayer, that they might sooner die than live to disgrace their father’s God… So soon as they become of years capable of understanding the things of God, we endeavour to bring them to Christ by teaching them the truth. Hence our Sabbath-schools, hence the use of the Bible and family prayer, and catechizing at home.”
C. H. Spurgeon, sermon no. 581, 24 July 1864. Emphasis Spurgeon’s. This is the fullest nineteenth-century Baptist answer to the question “what do you do instead?” and the answer is private prayer, Sunday school, family worship and catechizing — not a service.
Edward Hiscox’s Standard Manual for Baptist Churches (1890) devotes an entire chapter to arguing against infant baptism and nowhere proposes a substitute observance. The nineteenth-century English-speaking Baptist answer to the vacuum was catechesis and the Sunday School, not a rite.
The service arrives in print in the 1930s and 1940s
The printed order of service — the thing a minister opens a book and reads — enters the English-speaking denominations remarkably late, and it enters through the holiness and Wesleyan bodies rather than the Baptists.
A ritual for “The Dedication or Consecration of Children” appears alongside the existing ritual for infant baptism, provided for use in those cases where the parents “do not care to have children baptized but simply dedicated.”
A prescribed order for infant dedication appears for the first time, under the title “Dedication of Infants (Without Baptism).” The General Conference commissioned it to accompany legislation establishing a Children’s Membership Record.
The merged denomination carried the dedication service over, then noticed the problem: its Committee on Ritual observed that a service for the dedication of infants without baptism was “somewhat contradictory” to its own baptismal doctrine. The 1951 revision gave both services parallel formats — the same readings, the same address to parents, the same congregational questions.
From there the practice spreads through American Baptist, Southern Baptist, Pentecostal and independent evangelical churches in the second half of the twentieth century until it is simply what those churches do, and is assumed by most of the people doing it to be very old.
Interactive: whose words are these?
Eight quotations, all authentic, all findable. Guess who said each one and when. Most people get two or three. The point of the section is not the score; it is which way you guess wrong.
Interactive: what each rite claims to do
This is the comparison the whole question turns on, and it is almost never made carefully, because the two sides usually argue about whether infants should be baptized rather than about what either rite is asserting. Pick a tradition and read what it says it is doing. The differences among the paedobaptist traditions are as large as the difference between any of them and dedication, which is itself a finding.
The one structural difference
Step through the seven traditions and one field changes while the rest hold their shape. Every historic baptismal tradition, however much they disagree about what the water effects, answers the question who is the principal actor? with God. Trent and Geneva and Wittenberg and Canterbury fight bitterly over what God is doing, and agree entirely that the sentence has God as its subject. Dedication answers that question with the parents. It is not a weaker version of the same claim. It is a different kind of sentence.
Interactive: what do you think happens?
Not what your church teaches — what you actually believe is true of a child after the ceremony. Answer six questions and the page will tell you which historic position your answers correspond to. It is common for the result to be a tradition the reader does not belong to, and that is the useful outcome.
The form and the claim
Set the two services beside each other as physical events, ignoring theology entirely, and ask what an anthropologist with no interest in doctrine would write down.
| Element | Infant baptism | Baby dedication |
|---|---|---|
| Timing after birth | Days to weeks | Weeks to months |
| Location | Front of the church, main service | Front of the church, main service |
| Officiant | Ordained minister | Ordained minister |
| Infant held up before the congregation | Yes | Yes |
| Child’s name spoken aloud | Yes | Yes |
| Questions put to the parents | Yes | Yes |
| Promises made about upbringing | Yes | Yes |
| Congregation formally asked to respond | Yes, in most modern rites | Yes |
| Prayer over the child | Yes | Yes |
| Extended family attend; photographs; a meal after | Yes | Yes |
| Certificate issued | Yes | Yes |
| Entry in a church register | Yes — the baptismal register | Yes — the children’s record |
| Water | Yes | No |
| Trinitarian formula pronounced over the child | Yes | No |
| Something claimed to be done to the child by God | Yes | No |
| Child becomes a member of the church | Yes | No |
| Unrepeatable | Yes | No |
Twelve rows identical, five rows different, and the five are precisely the sacramental and covenantal claims. That is the finding of this page stated as a table. Dedication is what remains of infant baptism when you subtract every assertion it makes and keep everything else.
Two objections to that framing deserve answers, and one of them is good.
“Convergent form doesn’t prove derivation.” Correct in general — two communities can independently arrive at standing at the front of a room with a baby, because there is a limited number of ways to mark a birth publicly. But it is not a live objection here, because the derivation is documented rather than inferred. Hubmaier describes his rite in 1525 as what he does instead of baptism. The Nazarene Manual introduces it in 1936 for parents who do not want their children baptized. The United Brethren Discipline of 1945 calls it “Dedication of Infants (Without Baptism)” on the page. Nobody has to reconstruct the relationship. Every document states it.
“The traffic runs the other way too.” This one lands, and it is uncomfortable for paedobaptists rather than for credobaptists. Larry Ball, a Presbyterian minister writing in 2015, complained that “in many PCA churches covenant baptism has become little more than a wet dedication service” — that the sacramental claim has quietly drained out of the rite in churches that still perform it, leaving parents making promises and God doing nothing in particular. If that is right, then dedication has not merely borrowed baptism’s form; its theology has migrated back into the churches that kept the water. Which raises a question this page cannot settle: whether the rite that is really doing the work in modern Protestantism is a dedication with an optional wet version.
The case against this page
Stated at its strongest, by someone who thinks the page is wrong. Six arguments, in ascending order of force.
- The argument from silence proves far too much. The New Testament contains no order of service for a wedding, no funeral liturgy, no church building, no Sunday School, no midweek meeting, no membership class, no ordination service in anything like its modern form, and no instruction to sing anything written after AD 100. Christians do all of these by liberty and nobody calls them unbiblical innovations. A page that says “there is no command” and treats that as damaging has to explain why the same argument does not delete most of what every church does on a Sunday.
- Jesus blessed infants, and dedication does that. Mark 10:16 is not a proof text for a liturgy, but it is a record of what Jesus did when infants were brought to him: he took them in his arms, laid hands on them and blessed them. A church that brings infants to the front and prays over them is doing that thing. If the objection is that no words of institution accompany it, the objection applies to a great deal that Christians rightly imitate.
- The practice is five hundred years old, not modern. The page’s own research destroys the popular version of its thesis. Hubmaier is describing the rite in January 1525, before he had been baptized as a believer himself. It is contemporaneous with the birth of the tradition, not a late accretion or a marketing device, and it was defended from Matthew 19 by a first-generation Reformer with a doctorate in theology.
- The early-church footprint of infant baptism is thinner than paedobaptists concede. If the standard for a legitimate rite is documentary attestation in the first two centuries, infant baptism does not clear it either — Ferguson, no friend of the practice’s antiquity, and Lane, who thinks it probably was apostolic, agree that firm evidence begins in the late second century. And the fourth century shows a whole generation of eminent Christians raised in Christian homes and not baptized as infants, receiving instead a non-baptismal infant rite. A credobaptist can reasonably say: our practice has better ancient parallels than you allow, and yours has worse credentials than you claim.
- Dedication addresses the people the New Testament actually addresses. Ephesians 6:4, Deuteronomy 6, 2 Timothy 1:5 — the biblical weight on this subject falls on parents and their conduct over years. A rite that makes parents state their obligations publicly, before witnesses who will hold them to it, is aimed at the New Testament’s actual target. Infant baptism aims at the child; the New Testament’s child-rearing material aims at the parents.
- “It keeps the form and removes the claims” can be a compliment. That is what the whole Reformation did with a great deal of medieval practice, and what the Reformers thought they were doing on purpose. Keeping a good social structure while deleting a theological claim you believe to be false is not a confidence trick; it is the ordinary work of reform. If a credobaptist believes infant baptism asserts something untrue about an infant, retaining the family occasion and dropping the assertion is exactly the right thing to do, and describing it as “the social form of a sacrament with the sacrament removed” makes a virtue sound like a vice.
The last argument is the strongest and this page does not defeat it. What it does is narrow the disagreement to something worth arguing about: not whether a church may bless a baby — it plainly may — but whether the resulting service should be described to the people in the pews as the biblical practice that infant baptism displaced. On the evidence assembled here, it should not. It is a five-hundred-year-old Christian custom with a good pedigree and a bad genealogy, and it would lose nothing by being introduced as what it is.
So what should a church do?
The question that actually arrives in a pastor’s office is not historical. It is a couple with a newborn asking what happens next. Three things follow from the material on this page, and none of them requires anybody to change what they practise.
Say what it is. The service is a blessing of a child and a public assumption of responsibility by the parents, supported by the congregation. That is a good thing and needs no inflation. Introducing it as “the biblical practice of dedication, as Hannah dedicated Samuel and Mary and Joseph dedicated Jesus” asserts two things about two passages that neither passage will bear, in front of people who mostly cannot check — and the ones who can check will conclude that the pulpit is unreliable about texts.
Say what it is not. Nothing is conferred on the child. The child is not saved, not regenerated, not made a member, not marked, and not in a different condition at 11:15 than at 11:05. Most parents in the room believe some diffuse version of the opposite — which is the thing that ought to worry a credobaptist church most, because it means the borrowed form is doing exactly the work the borrowed theology used to do.
Notice which promise is load-bearing. The parental vows in a dedication service are the whole content of the rite, and they are almost never revisited. The historic traditions that ask parents to promise something attach the promise to a programme — catechism, instruction, a date of admission to the table, an examination. A promise made once before a congregation with nothing downstream of it is the weakest part of the modern service, and it is the part that could be strengthened tomorrow without anyone conceding a thing on baptism.
Your answers, revisited
Here is what you said before you read the page, next to what the sources show.
You did not lock in answers at the top. If you would like to, go back and do it — it takes two minutes and it is the part of the page most likely to be useful.
Check it yourself
Everything on this page is built from sources you can read without buying anything. Several are one page long. If you read only three, read the ones marked.
- 1. The biblical texts themselves, read in sequence and in one sitting: Exodus 13:1–16; Leviticus 12; Numbers 3:11–13 and 3:40–51; Numbers 6:1–21; Numbers 18:15–16; 1 Samuel 1–2:11; Judges 13; Luke 2:21–40; Mark 10:13–16. Read this first. It takes about thirty minutes and it settles more of this question than any commentary. Any translation will do; a parallel Bible is better for Luke 2:22.
- 2. C. H. Spurgeon, “Children Brought to Christ, and Not to the Font,” New Park Street / Metropolitan Tabernacle Pulpit, sermon no. 581, preached 24 July 1864: romans45.org; also at The Spurgeon Library. The best single document for what nineteenth-century Baptists actually did instead of a ceremony.
- 3. “Dedication of Infants,” Global Anabaptist Mennonite Encyclopedia Online, by Christian Neff and Harold S. Bender (1953), with a 1990 addition by David Janzen: gameo.org. Read this first. Two screens long, and it contains the Hubmaier letter of 16 January 1525, the Nördlingen agreement, and the Palatinate and Hesse conference decisions of 1830–1843. It is the single most important source on this page and it appears to be almost unknown outside Mennonite scholarship.
- 4. Andrew Messmer, “Infant Dedication in the Early Church: Texts, Commentary, and Present-Day Application,” Journal of European Baptist Studies 22:2 (2022), 61–89: jebs.eu, open access. Read this first. The strongest argument against this page’s thesis, made by a Baptist scholar from the primary texts. Read it before deciding this question is settled.
- 5. Tertullian, On Baptism 18: CCEL or newadvent.org. Two paragraphs. The first credobaptist argument in Christian literature, and it proposes catechesis rather than a ceremony.
- 6. Gregory of Nazianzus, Oration 40 (On Holy Baptism), especially section 28: newadvent.org. The clearest ancient discussion of when to baptize a child, from a bishop who was himself not baptized until adulthood.
- 7. Augustine, Confessions I.11: newadvent.org; parallel text at CCEL. The salt, the sign of the cross, and the deferred baptism, in his own words.
- 8. Anthony N. S. Lane, “Did the Apostolic Church Baptise Babies? A Seismological Approach,” Tyndale Bulletin 55.1 (2004): tyndalebulletin.org, PDF, open access. The most careful treatment of what the ancient silence does and does not permit you to conclude, in either direction.
- 9. Everett Ferguson, Baptism in the Early Church: History, Theology, and Liturgy in the First Five Centuries (Eerdmans, 2009). Not free, but in most theological libraries, and the standard survey. Ferguson’s own tradition does not baptize infants, which makes his concessions worth more than most.
- 10. “Dedication of Infants: A Ritual History,” United Methodist Discipleship Ministries: umcdiscipleship.org. The 1945 United Brethren Discipline, the Children’s Membership Record, and the 1951 Committee on Ritual noting the contradiction in its own service book.
- 11. The Manual of the Church of the Nazarene, ritual section, and its history: current text at nazarene.org. Compare the dedication and infant baptism rituals side by side; the denomination prints both, which makes the comparison in Part XIII unusually easy to make for yourself.
- 12. Your own church’s service. Ask for a copy of the words used, read them against the table in Part XIII, and count the rows. This is the least glamorous item on the list and the most informative.
Want this in twelve minutes instead?
There is a plain-English version of this dossier with the same findings, the same honest objections and one short quiz — no Hebrew, no Greek, no footnotes.
Read the quick version → Simple English →