Two practitioners, one birth certificate, two different readings. It happens often enough in BaZi to be worth a proper explanation rather than an apology, and the explanation is more interesting than either “one of them is a fraud” or “it is all interpretation”.
The short version: the canon openly contradicts itself, the disagreements are centuries old and written down, and several of them happen before anybody has interpreted anything.
First, the Word School
流派, school, is doing two jobs in this material, and running them together is where most popular writing goes wrong.
- Some are genuine textual traditions with doctrinal commitments you find in named surviving books. They disagree in print, and have for centuries.
- Some are modern brands and teaching curricula, dating from the 1990s onwards, whatever their marketing says about lineage.
- And some things called schools are not schools at all. Climate adjustment, image reading and symbolic stars are components every tradition knows about. What differs is how much weight each one gets.
An ethnographer who spent years with practitioners in Taipei, Beijing and Kaifeng reached a conclusion worth carrying through everything below: the vocabulary is shared, the detail and the interpretation vary, and resistance to standardisation is a feature of these arts rather than a fault in them.
A Problem at the Root
The tradition’s foundational text is credited to a Tang official. The imperial editors who catalogued it in the eighteenth century rejected the attribution in print. Their reasoning is worth repeating: the man’s own epitaph, written by a famous contemporary, praises his mastery of the five phases and never mentions a book. Older editions credited somebody else entirely with him as commentator. Their conclusion: a Tang text likely existed, and Song astrologers interpolated their own material into it under his name, to make their own art seem impressive.
The figure the modern method is named after fares no better. Textual scholars have argued he never existed at all, on the grounds of a commentary attributed to him dating to a century after his supposed lifetime. One scholar holds the legends about him to be an academic fabrication built up in layers. Traditional accounts place him in seclusion on a mountain with a cast of Daoist immortals, which is hagiography rather than biography.
None of this makes the method worthless. It does mean the tradition’s own claims about its origins were failing scholarly scrutiny two hundred and forty years ago, and a lineage claim is not the thing to judge a practitioner by.
Four Books, and Four Different Answers
The real disagreements are in print, in four classics, which is what makes them checkable rather than a matter of opinion.
淵海子平: the Song compilation, the first comprehensive treatment. One of its six chapters is given over to symbolic stars.
三命通會: the Ming encyclopedia. Inclusive, carrying the full traditional apparatus and older material alongside the newer method.
滴天髓: cryptic, famous, and disputed as to authorship. The commentary most people read in practice carries 512 worked chart examples.
子平真詮: the Qing text of structure theory, reportedly written as private notes with no intention of publishing them.
The tradition’s own reference works classify the methods into four: support and suppress, structure, sound based five phases, and climate. It is an internal taxonomy rather than a modern list, and a better starting point than any roll of schools.
Structure
格局, structure or pattern, classifies a chart into a configuration and reads it accordingly. The founding commitment is compact and quotable: the useful element is sought exclusively in the month command, which is the month branch, treated as the seat of seasonal authority over the whole chart.
Eight regular patterns are named by the relationship between the Day Master and the dominant element of the branch. A crisp rule governs what to do with each: four of the relations are benign and are supported, four are difficult and are controlled.
The tradition also criticises itself here, and the criticism is fair. Structure practitioners are fluent on charts forming a clean pattern and often stuck on charts which do not, and most ordinary people’s charts do not. The difficulty is given inside the tradition as the reason the strength approach exists at all.
Strength
旺衰, strength and decline, asks how well supported the Day Master is by everything else, and reads the chart to restore a balance. The assessment runs in order: does the season of the month support it, does it have roots among the hidden stems, is a hidden element also revealed openly in a stem, and how much is draining or controlling it.
The pair of ideas worth learning is rooting and revealing. An element with a root has depth. An element revealed at the stem but rooted nowhere is on the surface. An element rooted but not revealed is latent.
Two practitioners weighting roots and reveals differently will rank the same element as dominant or negligible, all before any interpretation begins. Numerical scoring schemes circulate, assigning points to each position. They are modern teaching devices, they vary by teacher, and none of them is classical.
The Useful God, and Where the Disagreement Bites Hardest
用神, the useful god, is the element the chart most needs, and it is the pivot the whole reading turns on. The tradition itself names five ways of choosing one.
- 扶抑, support or suppress: strengthen a weak Day Master, drain a strong one.
- 調候, regulate the climate: with a chart at an extreme of cold, heat, dryness or damp, fix the extreme first.
- 通關, open the pass: when two forces are locked in opposition, choose what mediates.
- 病藥, disease and medicine: find what obstructs the needed element, and take the remedy as the useful god.
- 專旺, concentrated flourishing: where the chart is overwhelmingly one directional, follow it rather than fight it.
Now the structural point, and it is the single clearest answer to the question in this article’s title. In the structure tradition the useful god is fixed by the month command, so it is decided by where in the solar year you were born. In the strength tradition any one of the eight characters might be the useful god, decided by a balance judgement across the whole chart.
Those are not different emphases. They select different elements, sometimes opposite ones, and every downstream statement about favourable years, favourable directions and favourable choices follows from the selection.
One historical note cuts against the rival schools framing, though. Strength and structure were originally one family, and took the same route in the earliest compilation. The split is a later hardening.
Climate, and a Conflict with No Tiebreak
調候 adjusts for extremes of cold, heat, dryness and damp, judged mainly from the month of birth. The classic behind it treats the cosmos as a furnace with the four seasons for fuel, and offers practical maxims: a summer chart should not be without water, a winter chart needs to see fire.
The priority claim is simple. Climate is urgent, so where a chart sits at a climatic extreme the climate fix comes first and the structure or strength choice waits.
Here is the live conflict. When the climate useful god and the structure or strength useful god point at different elements, and they routinely do, there is no agreed tiebreak. Two competent practitioners name opposite elements as favourable, and each cites a classic. Nothing in the tradition adjudicates.
Following Patterns, and a Threshold Nobody Agrees On
從格, the following patterns, cover charts where the Day Master is so unsupported it is read as surrendering to the dominant force rather than being propped up.
The judgement inverts the entire reading. In an ordinary chart a weak Day Master wants support. In a following chart, support is destructive. One threshold call therefore flips every favourable element to unfavourable, and the threshold is not agreed.
The distinction between true following and partial following exists precisely because the boundary is fuzzy. There is also a classical asymmetry worth knowing: the yang day stems will not abandon themselves while any support remains anywhere in the chart, and the yin stems enter a partial following pattern more readily.
The Blind School
盲派 needs care, because three different claims usually arrive fused together.
Blind diviners as a profession are documented. Guilds of blind musicians and fortune tellers operated in China into the twentieth century, and blind practitioners worked by touch and by voice. The guilds claimed continuous existence back two thousand years, and the claim is the guilds’ own.
The origin story tying the method to a Warring States figure is legend, and it carries the marker Chinese sources use for legend. Notice the pattern: the imperial editors found the same figure attached to the root text and judged it a device for making one’s own art seem divine. Attributing a method to him is a recognisable move rather than a historical claim.
The lineage claim is the weakest part. The Chinese reference article on the subject has carried a banner saying it has no sources whatsoever since 2012, and it names no historical blind practitioner and no documented lineage holder. What is verifiable: the books putting the method into print appeared in the early 2000s.
And yet this has the best claim of anything here to being a genuinely different analytical route. It largely bypasses both strength assessment and structure classification, the two things the sections above are entirely about, and works instead by extracting concrete images and scenarios directly from the stem and branch structure, leaning on branch interactions to call specific events. Its own sympathisers note the inconsistency: different masters read one chart in sharply different ways.
So the position worth holding is an uncomfortable one. The best documented methodological distinctness in the subject sits with the least documented lineage.
Image Reading
象法 reads a stem and branch combination as a concrete picture rather than resolving it into quantities of elemental strength. An officer sitting on a seal is read directly as an official post.
It is not a separate school. It is a mode of reading which the blind school foregrounds and the strength and structure traditions subordinate. Chinese classification has long separated calculation based arts from sign based ones, so reading by image has old standing and is not an invention of one lineage.
Symbolic Stars, and an Eight Hundred Year Old Demotion
神煞 are named symbolic stars derived by table lookup: peach blossom for attraction, the travelling horse for movement, the canopy for solitary scholarship, the nobleman for rescue, the blade for force, the void for things failing to materialise. The compendia catalogue more than a hundred.
This is the best evidenced split in the whole subject, because the books are on record on both sides. The structure classic holds them irrelevant to whether a pattern succeeds: a chart full of inauspicious stars will not damage a sound pattern, and every auspicious star in heaven will not rescue a broken one. Another classic dismisses several by name as empty talk, and its major commentator used not one of them across 512 worked charts. On the other side, the Song compilation gives them a whole chapter and the Ming encyclopedia carries the full apparatus.
The historical fact behind it is better than the argument. Symbolic stars were central to the older method, the one using year and month stems, and the Zi Ping reform demoted them to an auxiliary role about eight hundred years ago in favour of the Ten Gods. The classics rejecting them are continuing a reform rather than starting a fight.
Two of the lookups are themselves disputed, which is a sharper problem than whether to use them.
- The nobleman star has two competing mnemonics differing by a single character. One reading includes a stem in the first group and the other excludes it, and the two produce different answers from the same five line verse.
- The blade star is agreed for the yang stems and genuinely contested for the yin ones, with two published tables disagreeing and some schools holding the yin stems have no blade at all. A third lookup circulating in popular articles is a corruption.
Practitioners split them into those derivable from the underlying logic and those inherited wholesale from an older astrology, of which it is openly said they work sometimes and not others. A common position inside the practice holds symbolic stars to carry no force of their own, being symbols rather than causes.
Hidden Stems, the Disagreement Which Propagates
This one is the most consequential and the least visible, because it happens before any reading starts.
Each branch contains one to three hidden stems, and practitioners do not agree on the contents.
- Two recensions of the standard verse differ for five branches, and the best known single disagreement concerns whether the Pig branch carries a third stem.
- Two incompatible orderings exist: by strength, and by the order in which each stem takes command through the month.
- The command scheme assigns days of rule to each hidden stem after the month opens, and different classics give different day counts for the same branch.
- There is an explicit claim the two systems cannot correspond one to one, and structure practitioners largely ignore the command scheme even though every classic prints the table.
Hidden stems feed rooting. Rooting feeds strength. Strength feeds the useful god. A disagreement at the table level therefore arrives at the final verdict having been amplified twice, and the practitioners involved might never discuss the table.
Percentage weightings for hidden stems circulate widely. None of them is classical, and one popular set does not even sum to one hundred.
Timing
Two documented disagreements, both anchored in classics, and both changing when an event is predicted rather than whether.
The first: does a ten year luck pillar split into two halves, the stem governing the first five years and the branch the second? The Ming encyclopedia says yes. Another classic says flatly to read one luck pillar as ten years and never to cut it top from bottom, because stem and branch are one body. Those are contradictory instructions from two canonical books.
The second is about weighting: a common maxim gives the branch priority in the luck pillar and the stem priority in the annual one, on the reasoning of the branch being root and the stem leaf.
There is also a smaller cycle for the years before the first luck pillar opens, and it has two incompatible methods. One steps from the hour pillar, one starts from a fixed position by sex. The classical text gives the fixed branch and derives the stem per year, which is neither of the two methods circulating. Many practitioners drop it altogether once the luck pillars start.
And Some Disagreements Are About the Data
This is a different kind of thing from everything above, and it deserves its own beat. The disputes so far are about how to read a chart. These are about what the chart IS.
- The year boundary is a three way argument between Li Chun, the first day of the first lunar month and the winter solstice, and the second of those is the Chinese national standard.
- The hour before midnight belongs to the outgoing day or the incoming one depending on the rule a practitioner follows, which changes the day pillar, which changes the Day Master, which changes every one of the Ten Gods.
- True solar time corrections for daylight saving, longitude and the annual wobble between clock noon and solar noon compound past an hour, and a two hour branch flips on about an hour of error.
The mechanics of all three, and why the hour pillar is the weakest input in the chart, are in what is BaZi? a guide to the Four Pillars of Destiny.
A practitioner who tells you which conventions they use has given you more information than one who tells you their lineage.
The Modern Picture
Three Republican era writers in Shanghai shaped how the classics reach modern readers, and one of them produced the standard editions of two of the four books, so most people meet those texts through his commentary.
After the economic reforms the mainland field split three ways: a traditional stream, a new school, and the blind school. The new school simplified radically, dropping symbolic stars and sound based five phases entirely, reducing every chart to one of four types and reading favourable elements straight off it. Its distinctive doctrines include a rule against non-adjacent stems and branches interacting at all, which traditional theory rejects outright.
There is also a substantial commercial layer of paid certification academies, mostly Southeast Asian, which shapes most English language BaZi vocabulary. Lineage claims in the sector are generally not independently documented. One practising consultant’s own framing is the fairest available: each practitioner has their own line of training and experience, doing things differently does not make one of them right or wrong, and the situation frustrates anybody trying to learn.
And the presentation of this subject as a rational or scientific discipline is itself recent. Both the mainland and Taiwan revivals date from the 1980s and 1990s, after decades in which these arts were officially classed as superstition, and the drive to legitimise them academically belongs to the revival rather than to the tradition.
So Why Do Two Trained Practitioners Disagree?
In rough order of how much they change the reading.
- Different doctrine for choosing the useful god, fixed by the month command or judged across all eight characters, which sometimes selects opposite elements.
- A conflict between the climate choice and the structure or strength choice, with no agreed tiebreak.
- A following pattern threshold call, which inverts every favourable element in the chart.
- Symbolic stars used heavily or not at all, with canonical support on both sides.
- Different hidden stem tables and orderings, which propagate up through rooting and strength.
- Luck pillar reading, split into halves or read whole, stem weighted or branch weighted.
- Bypassing strength and structure altogether in favour of images and branch interactions.
- Disagreements about the chart itself, upstream of all interpretation.
- And syncretism, because most working practitioners blend, and the proportions of the blend are personal and usually unstated.
None of These Schools Is the Correct One
This guide does not rank them, and the reason is not diplomacy. The tradition has no adjudicating authority, no body of outcome data for testing rival doctrines against, and a canon whose own core books contradict one another in print.
Under those conditions divergence is not a defect in the practice. It is the predictable result of the situation, and a practitioner able to tell you which doctrine they follow, and why, has given you the most useful thing available.
Most working practitioners are syncretic anyway. A common modern description has structure for the skeleton, strength for the flesh, climate for the blood and image reading for the nerves, which is an honest account of how a chart usually gets read, and a reminder of the neat taxonomy being more real in books than at the table.
What to do when you have two readings and they conflict is set out in how to tell a careful practitioner from a confident one.