Walk into any Hebrew Roots community and you will find people who are certain. Certain that they have recovered the original faith. Certain that the church has been wrong for centuries. Certain that Torah observance is what God requires of all believers. The certainty is real, it is sincere, and it radiates from people who have studied hard and care deeply about getting scripture right.

Now put two of those certain people in a room together and ask them what "fulfil" means in Matthew 5:17. Or which parts of Torah are currently binding. Or what Paul meant by "works of the law" in Galatians. Or what the Sabbath commandment actually requires. Or which version of the church's corruption history is the correct one.

They will disagree. Not about peripheral details. About the foundational claims of the movement itself.

This article maps those disagreements across six of the HRM's most central theological claims. In every case, the movement holds multiple incompatible versions of the same position - versions that cannot all be true, that often directly contradict each other, and that are deployed in different conversations depending on which one is most resistant to the challenge at hand. The mathematics of this situation is simple and devastating: if there are six versions of a claim, at best one is right. That means at least five are wrong. And the person holding the wrong version is not holding it tentatively as a live question - they are holding it with the same certainty as the person holding the version that contradicts it.

Before examining the contradictions, one clarification is necessary. The point of this article is not that internal disagreement disqualifies a theological movement - all traditions have internal debates. The point is that these particular disagreements are not debates about secondary matters between people who share a common foundation. They are disagreements about the foundation itself. They reveal that the HRM's certainty is not grounded in a clear, consistent reading of scripture but in a shared emotional and cultural orientation toward Hebrew identity - an orientation that different teachers then fill with different, incompatible theological content. And that distinction matters enormously.

Area One: What "Fulfil" Means in Matthew 5:17

Matthew 5:17 is the HRM's most cited verse. "Do not think that I have come to abolish the Law or the Prophets; I have not come to abolish them but to fulfil them." The entire HRM case for Torah observance leans heavily on this verse. You would expect, therefore, that the movement would have a clear and settled answer to the obvious question: what does "fulfil" (plēroō) mean here?

It does not. The HRM holds at least four mutually incompatible definitions, deployed depending on which one is most resistant to the challenge being made.

Version A - Fulfil means confirm as permanently valid. Jesus fulfilled the law by ratifying it, by placing his authority behind it, by declaring it binding for all time. Fulfil = affirm. This is the simplest version and the one most commonly stated in introductory HRM teaching.

Version B - Fulfil means fully preach. Jesus fulfilled the law by teaching it in its complete depth - expounding it, revealing its full meaning, bringing out what had been obscured by tradition and misreading. He filled it up with meaning. Fulfil = expound comprehensively.

Version C - Fulfil means live out as our example. Jesus fulfilled the law by keeping it perfectly. He is the model Torah-keeper whose example we are to imitate. He showed us what faithful Torah observance looks like so that we can follow his pattern. Fulfil = demonstrate through practice.

Version D - Fulfil means bring to its fullest expression without ending it. Jesus fulfilled the law the way a musician fulfils a composition - performing it completely, bringing out its depths, not cancelling it but honouring it in the most complete way possible. Fulfil = bring to completeness while maintaining.

These are not four shades of one meaning. They are four different claims about what Jesus did to the Torah. Version A says he ratified it. Version B says he taught it fully. Version C says he modelled it. Version D says he completed it while maintaining it. A person can hold Version A without holding B, C, or D. A person who holds Version C is committed to the claim that Torah observance is normative because Jesus modelled it - but then has to explain why Jesus also fulfilled Isaiah 53, Psalm 22, and Micah 5:2 as our example to imitate. A person who holds Version D is closest to the mainstream reading and, if consistent, would have to acknowledge that things can be completed without continuing - which is what mainstream theology says.

Here is the decisive test. Matthew uses plēroō twelve other times in his Gospel. In every single other use, without exception, it describes a prophecy arriving at its appointed completion - "this happened to fulfil what was spoken by the prophet." Not confirming. Not teaching fully. Not modelling. Arriving at completion. The word Matthew uses everywhere else for a prophecy meeting its appointed end is the same word Jesus uses in 5:17. The HRM needs it to mean something different in 5:17 than it means in all twelve other uses in the same book - and produces four different alternatives, none of which matches Matthew's own consistent usage. At best, one of the four versions is right. The other three are wrong. And all four are wrong about what Matthew means by the word.

Area Two: What "Works of the Law" Means in Galatians

Galatians is the letter the HRM most needs to neutralise, because it is Paul's most sustained argument against imposing Torah observance on Gentile believers. The phrase at the centre of the letter's argument is erga nomou - "works of the law." Paul says in Galatians 2:16 that a person is not justified by works of the law but through faith in Jesus Christ. The HRM cannot accept the straightforward reading of this - that Torah observance does not justify - and so has developed multiple alternative definitions of what "works of the law" means.

Version A - Works of the law means boundary markers. Drawing on the New Perspective on Paul scholarship of NT Wright and James Dunn, the HRM argues that erga nomou refers specifically to the sociological boundary practices - circumcision, food laws, calendar observance - used by sectarian Jewish groups as conversion requirements and membership badges. Paul opposes these practices as identity markers that divide Jew and Gentile, not Torah observance generally. The Dead Sea Scrolls document 4QMMT uses the phrase in a similar way.

Version B - Works of the law means meriting salvation. Paul opposes using Torah observance as a credit system - performing commandments to earn standing before God. Torah observance as a grateful response to grace is fine and required. Torah observance as the basis of justification is what Paul condemns. Obedience flows from salvation; it does not earn it.

Version C - Works of the law means Oral Torah additions. What Paul calls "works of the law" is not the written Torah at all but the rabbinic fence-laws - the human traditions layered on top of Moses by the Pharisaic tradition, the Talmudic additions that Jesus condemned in Mark 7 as the "tradition of the elders." Paul opposes man-made law, not God's written Torah.

These three versions are mutually exclusive. The scholars the HRM borrows Version A from - Wright and Dunn - do not conclude that Torah observance is therefore binding on Gentile believers. They conclude the opposite: that the boundary markers Paul opposes are precisely the Torah practices the HRM advocates. The HRM takes the scholarly observation (the phrase refers to boundary markers) but not the scholarly conclusion (therefore those boundary markers are not required of Gentiles). Version B is closer to traditional Protestant theology than to anything distinctively HRM - and it cannot explain why Paul treats circumcision in Galatians 5:4 not merely as a works-salvation error but as something that severs a person from Christ entirely, regardless of their motivation. Version C requires Paul to have a word for "Oral Torah" and to have chosen not to use it - and the 4QMMT document the HRM cites uses the phrase for written Torah requirements, not rabbinic additions.

The HRM deploys Version A when engaging with scholarly objections, Version B when engaging with evangelical objections, and Version C when engaging with objections based on Jesus's own Torah observance. The version deployed is determined by the challenge, not by the text.

Area Three: Which Parts of Torah Are Binding

This is the question the HRM most urgently needs to answer and most consistently refuses to answer directly. The Torah contains 613 commandments. The HRM teaches that Torah is binding on all believers. The obvious question is: which commandments? The movement's answers are irreconcilable.

Version A - All 613. Not one jot or tittle passes away. Every commandment God gave remains binding. Jesus confirmed the whole law. Selective observance is not Torah observance.

Version B - Moral law only. The civil legislation of ancient Israel and the ceremonial law were situation-specific and Israel-specific. The moral law - summarised in the Ten Commandments - is universal and permanently binding because it reflects God's unchanging character.

Version C - Everything except what has been explicitly fulfilled. The sacrificial system is fulfilled in Christ's once-for-all sacrifice and is therefore no longer practiced. The temple and priesthood laws are suspended because there is no standing temple. But Sabbath, feast days, dietary laws, and everything else that has not been explicitly superseded remains binding.

Version D - Written Torah minus Oral Torah additions. Everything in the five books of Moses is binding, but the rabbinic commentary, the Talmudic elaborations, and the Mishnaic fence-laws are not. God's written word is the standard; human tradition is not.

Version A is almost never practiced. HRM communities do not sacrifice animals, maintain Levitical purity laws, apply Mosaic civil legislation, observe the Jubilee year, or follow the agricultural laws of the land of Israel - all of which are in the same Torah as the Sabbath and the feast days. Version A is stated in principle and ignored in practice. Version B produces something virtually indistinguishable from Reformed Christianity and renders the HRM's distinctive practices unnecessary - if only the moral law is binding, why are feast days and dietary restrictions required? Version C makes a distinction - fulfilled vs. not yet fulfilled - that the Torah itself does not make. No chapter of Leviticus divides itself into "currently applicable" and "suspended pending temple reconstruction." The distinction is imported from the outside, not derived from the text. Version D requires determining which traditions are human additions, which is a massive interpretive project with no agreed standard - and it produces the irony that the HRM ends up doing exactly what it accuses mainstream Christianity of doing: making interpretive distinctions within Torah that Torah itself does not authorise.

At best one version is right. The other three are wrong. And the one that is right determines a completely different set of practices than the others. The person who holds Version A and the person who holds Version B are not slightly different Torah-observers. They are living entirely different religious lives - and calling both of them "Hebrew Roots" while they contradict each other's foundational commitments reveals that the label covers not a theology but a cultural orientation.

Area Four: What the Sabbath Commandment Requires

The Sabbath is the HRM's most identity-defining practice. It is the point of most visible distinction from mainstream Christianity. You might expect, therefore, that the HRM would have absolute clarity about what the Sabbath requires. It does not.

Version A - Saturday, sundown to sundown, no exceptions. The Sabbath must be observed from Friday evening to Saturday evening. This is the biblical commandment. Any other day, any other calculation, any flexibility in timing is a departure from God's command. Sunday worship is Constantinian replacement.

Version B - The principle matters; the exact timing is flexible. When pressed on the practical difficulty of universal sundown-to-sundown observance across different time zones, or on the lives of shift workers, medical professionals, and those in regions with extreme seasonal daylight variation, some HRM teachers argue that the heart principle - rest, cessation from work, dedication to God - is what the commandment requires. Specific hours are a later rabbinic imposition, not the biblical text.

Version C - The Sabbath must be recalculated from the new moon. A significant minority within the HRM - the Lunar Sabbath movement - argues that the continuous seven-day week is a later invention and that the Sabbath was originally calculated from new moon cycles, making it a different day each month rather than a fixed weekly day. The current Saturday may not be the original seventh day at all.

Version D - The Sabbath is a sign for Israel, but Gentiles share it through grafting. Exodus 31:17 explicitly identifies the Sabbath as "a sign between me and the children of Israel forever." The HRM acknowledges this specificity but argues that Gentiles grafted into Israel inherit the sign along with its obligations.

Versions A and C are directly incompatible - you cannot observe Saturday sundown-to-sundown and simultaneously hold that Saturday is not the correct day. Versions A and B are incompatible - if the principle is flexible, the specific day loses its absolute binding character, which undermines the main argument against Sunday worship. Version D creates a problem from the text it cites: Exodus 31:17 calls the Sabbath a sign specifically between God and Israel. If the Sabbath is Israel's covenant sign, and Gentiles are grafted into Israel, the question becomes whether grafting in brings covenant obligations or covenant blessings - and the New Testament's consistent answer, from Acts 15 to Galatians to Romans, is that Gentile grafting does not bring Torah obligation.

Within HRM communities you will find Version A holders and Version C holders in direct, unresolved disagreement - each certain they are observing the correct Sabbath, each certain the other is not observing it correctly, neither able to appeal to an agreed standard within the HRM framework that settles the question.

Area Five: What "Grafted In" Means in Romans 11

Romans 11:17–24 is the HRM's primary proof text for Gentile Torah obligation. Gentiles are wild olive branches grafted into the cultivated olive tree. They share in the rich root. The argument: the root is Torah, therefore grafted Gentiles are obligated to Torah.

Version A - Grafted into Torah obligation. The olive tree is the Torah-observant covenant community of Israel. Grafting in means joining that community with its full covenant obligations intact.

Version B - Grafted into the blessings and promises. Gentiles share in the covenant blessings - forgiveness, the Spirit, eternal life, the promises to Abraham - without necessarily sharing every covenant obligation specific to ethnic Israel.

Version C - Grafted into the commonwealth of Israel. Ephesians 2:12 says Gentiles were excluded from the commonwealth of Israel. Grafting in restores them to citizenship in Israel, which carries Torah obligations as part of civic membership.

Version D - Grafted in through Abraham, not Moses. The root of the olive tree is Abrahamic faith - the unconditional promise given to Abraham 430 years before Torah. Gentile grafting is into the promise, not into the Mosaic covenant that came later.

Version D is what Paul actually argues in Romans 4 - the same letter. He explicitly identifies the root as the faith credited to Abraham before circumcision (Romans 4:9–12), 430 years before Sinai (Galatians 3:17). Paul's own identification of the root is Version D, not Version A. The HRM needs Version A but Paul provides Version D in the immediate context. Versions B and C make different claims about what "grafting in" transfers - blessings, or obligations, or citizenship - and the text does not adjudicate between them in the way the HRM needs. Version C's appeal to Ephesians 2:12 backfires: the rest of that passage describes the "new man" Christ created - one new entity in place of the two - with no mention of Torah obligation for Gentile members of this new community.

The HRM appeals to Romans 11 as its grafting proof text while consistently ignoring Romans 4, which is in the same letter, by the same author, and identifies the root as Abraham's pre-circumcision, pre-Torah faith.

Area Six: When the Church Was Corrupted

The previous article in this series covered this in detail, but it belongs in any complete accounting of HRM internal contradictions. The corruption narrative - the claim that the original Torah-observant church was replaced by a paganised counterfeit - is the framework's ultimate safety net against historical evidence. And the movement holds six incompatible versions of when this corruption occurred.

Version 1 - Constantine (325 AD): The Council of Nicaea is the turning point. Refuted by Sunday worship documented in 107 AD, 112 AD, and 150 AD.

Version 2 - Proto-Catholic church (2nd–3rd century): The early church fathers corrupted the faith. Refuted by Polycarp knowing John personally - if Polycarp was wrong, the corruption reached John.

Version 3 - Post-70 AD Gentile takeover: Jerusalem's scattering allowed Gentile leadership to replace Hebrew Christianity. Refuted by Paul's letters, written before 70 AD, already saying what mainstream Christianity says they say.

Version 4 - Paul himself: The false apostle introduced Hellenised theology. Self-refuting because the HRM cites Paul selectively - he is reliable when Torah-positive and unreliable when Torah-negative.

Version 5 - The Septuagint (250 BC): Greek translation corrupted the source material. Self-refuting because it predates Christianity entirely and Jesus quotes the Septuagint.

Version 6 - Perpetual structural corruption: Gentile Christianity was never right. Unfalsifiable by design - any evidence of non-Torah-observant early Christianity becomes proof of how thorough the corruption was.

No two versions are compatible. Each one, if true, makes the others false. All six are held simultaneously within the movement, deployed as needed.

The Mathematics of the Situation

Step back from the details of each area and consider the simple arithmetic.

In Area One, the HRM holds four versions of what "fulfil" means. At best, one is right. That means at least three are wrong - held with certainty, taught with authority, used to construct theological conclusions, by people who are confident they have the correct reading of God's word.

In Area Two, the HRM holds three versions of what "works of the law" means. At best, one is right. At least two are wrong.

In Area Three, the HRM holds four versions of which parts of Torah are binding. At best, one is right. At least three are wrong - and the person who holds the wrong version is not observing what they think they are observing. They are performing a religious practice under the conviction that it is divinely required, based on a reading of scripture that is incorrect.

In Area Four, the HRM holds four versions of what the Sabbath commandment requires. At best, one is right. People in the same community may be observing the Sabbath on different days, by different calculations, with different rules about what constitutes rest - and both are certain they are obeying God's command.

In Area Five, the HRM holds four versions of what "grafted in" means. At best, one is right.

In Area Six, the HRM holds six versions of when the church was corrupted. At best, one is right. At least five are wrong - meaning that at least five versions of the history that explains why mainstream Christianity doesn't observe Torah are factually incorrect.

Add it up. Across six theological areas, the HRM holds between three and six versions of each claim. At best, six out of the total twenty-five versions are correct. At least nineteen are wrong. Nineteen wrong versions, held with the same certainty as the potentially correct ones, used to construct the same obligations, applied to the same religious lives.

And here is the deepest problem: the HRM has no agreed internal standard for determining which version is correct. There is no agreed council, no agreed confessional standard, no agreed hermeneutical framework that the movement as a whole has adopted to adjudicate between the versions. Different teachers hold different versions. Different communities follow different teachers. The certainty that characterises each version is not the certainty of having worked through the disagreements and reached a settled conclusion. It is the certainty of having not yet encountered - or not yet taken seriously - the version that contradicts it.

Why This Pattern Exists

The existence of multiple incompatible versions across every major HRM theological claim is not a coincidence. It is the predictable result of a method that begins with a conclusion and works backward to find support, rather than beginning with the text and following it to a conclusion.

When the conclusion is fixed - Torah observance for all believers - and the text is examined to find support for it, the specific support found will vary depending on the examiner's background, the challenges they have encountered, the scholars they have read, and the conversational context in which they are making the argument. Different examiners find different support. Different support produces different versions. The versions differ because the method produces whatever the examiner needs at the moment of examination, not because the text itself teaches different things to different people.

A genuine textual investigation, by contrast, tends toward convergence. When people examine a text carefully, without a predetermined conclusion, the evidence pushes them toward the same place - because evidence has a direction, and that direction is the same regardless of who is reading. The divergence of HRM versions is not the diversity of honest inquiry. It is the fragmentation of motivated reasoning operating in different contexts.

The person genuinely seeking what scripture teaches asks: what does this passage say, and what does that require me to believe? The answer to that question is the same for everyone who reads the passage carefully. The person defending a predetermined conclusion asks: how does this passage support what I already know? The answer to that question varies with every conversation, every challenge, every new front that needs defending. Twenty-five versions, at most six correct, at least nineteen wrong - not because scripture is unclear, but because the method was never designed to let scripture speak.

The Conclusion That Cannot Be Avoided

Here is what the mathematics requires us to say.

If there are four versions of what "fulfil" means in Matthew 5:17, and they are mutually incompatible, then the person who holds any given version has approximately a one-in-four chance of being right - and a three-in-four chance of having built their entire understanding of Matthew 5:17 on a foundation that is wrong.

If there are four versions of which parts of Torah are binding, the person who holds any given version has approximately a one-in-four chance of observing what God actually requires - and a three-in-four chance of performing a religious practice under the sincere but mistaken belief that it is divinely obligated.

The certainty with which these versions are held does not change the arithmetic. Certainty is not evidence. The person who is certainly wrong is still wrong. And the pattern across all six areas - four versions here, three there, six here - is the pattern of a movement that has mistaken emotional and cultural conviction for textual conclusion.

There is a version that is correct - one reading of Matthew 5:17 that reflects what Jesus actually said and meant, one account of which parts of Torah are binding that reflects what God actually requires, one correct history of the church that reflects what actually happened. The existence of that correct version does not validate the twenty-four incorrect ones. It condemns them - because the people who hold the incorrect versions are certain they hold the correct one, and they are wrong.

The appropriate response to discovering that your movement holds six incompatible versions of its most foundational claims is not to redouble commitment to the version you currently hold. It is to ask - with genuine openness, with the willingness to be wrong - which one is actually supported by the text. And to ask that question as a genuine question, not as a search for confirmation.

The Bereans were commended because they examined the scriptures to see if the things they were being taught were so. They were not commended for the certainty of their conclusions. They were commended for the honesty of their examination. Honest examination carries the risk of finding that the things are not so - and that risk, accepted openly, is the beginning of the kind of reading that actually arrives at truth rather than at the version of truth that was already decided before the examination began.

Now these Jews were more noble than those in Thessalonica; they received the word with all eagerness, examining the scriptures daily to see if these things were so. - Acts 17:11