Parāśara wrote Bṛhat Parāśara Horā Śāstra during a yuga-shift, and somewhere in the Aṣṭakavarga section he says something startling — he says this tool is necessary in the Kali Yuga because the human intellect has degraded through sinful deeds.1 He is not flattering his future readers. He is saying: by the time my book reaches you, your minds will be too compromised to use the subtle dṛṣṭi and Avasthā readings reliably. So I'm leaving you a more mechanical, numerical, brute-force tool. Add up the dots. Count the rekhā. The arithmetic will catch what your degraded subtlety misses.
That tool is Aṣṭakavarga. Aṣṭa = eight. Ka = the Brahmā-bīja (creation-syllable, also coded as kṣetra, the chart itself). Varga = division. Eight divisions of the rāśi chart, generated by contributions from the seven grahas plus the Lagna.2 The nodes (Rāhu and Ketu) are excluded — they cause rebirth and are themselves shadows; their creation (the Lagna) takes their place in the eight-fold scheme. The system distributes 386 rekhā (lines, blessings) across 96 chart-positions (12 houses × 8 vargas), producing per-house strength scores that quantify each bhāva's potential.
Larsen lays out three operational levels of Aṣṭakavarga — SAV (Samudaya Aṣṭakavarga), BAV (Bhinna Aṣṭakavarga), and PAV (Prastara Aṣṭakavarga) — each with different predictive utility. The architecture turns out to be a sharpening lens: SAV gives the broad overview, BAV gives the per-graha-area read, and PAV — only PAV — actually produces predictively accurate Daśā-level predictions. The first two are diagnostic; the third is operational.
Rekhā and Bindu: Each of the 96 chart-positions receives either a rekhā (line, blessing, potential-fulfilled) or a bindu (dot, lack-of-potential, energy-trapped). The total rekhā distributed across the chart is 386 — averaging about 28 per house, but distributed unevenly per the graha-contribution rules.
The eight contributing vargas (per graha contribution-total):
SAV (Samudaya Aṣṭakavarga) — the collective view3 Sum all eight vargas' rekhā for each house. Average per house: 386 ÷ 12 = ~32. Above 32 = auspicious for that house. Below 32 = inauspicious. (Some lineages following Varāhamihira exclude the Lagna varga, giving an average of 337 ÷ 12 = ~28.) The SAV tells you where the native's overall energy concentrates — more rekhā in the 10th = karma-yogi. More in the 11th = bhogi (one who strives to enjoy fruits). Useful for overall character-assessment. Not predictively accurate for specific events.
BAV (Bhinna Aṣṭakavarga) — the per-graha view4 Each of the eight vargas deals with one specific life-area, indicated by that graha's kārakatva. Find the kāraka of the house-of-focus. Open that kāraka's BAV. Read the contribution to the house-of-focus. Maximum 8 rekhā per house; average 4.
Example: Sun is the kāraka for the Lagna (health). The native's Lagna sign has 2 rekhā in Sun's BAV — below average — so the native's health is unstable. The mother (4th house) has 4 rekhā in Sun's BAV — average. The father (9th house) has 3 rekhā in Sun's BAV — below average; father has below-average health.
Similarly Mars is the kāraka of co-born. Treat Mars as Lagna, read the 3rd from Mars in Mars's BAV. The rekhā count gives the maximum number of co-born the native may have. Mars in Taurus, 3rd from Taurus is Cancer, Cancer has 4 rekhā in Mars's BAV → native has 4 brothers.
BAV is more sharply diagnostic than SAV. Still not predictively accurate for specific events — but excellent for category-level reads.
PAV (Prastara Aṣṭakavarga) — the spread-out view5 Prastara means "spread-out." This is the BAV unpacked — each rekhā in a BAV is shown which graha contributed it. The PAV table for, say, Sun shows: "of the 8 possible rekhā Sun can give to each house, which planets are contributing each rekhā?" The breakdown matters because each contributing-planet represents a person who is bringing energy to that house in the native's life.
PAV is the only level conducive to predictive accuracy. Method: identify the Daśā running. Open the PAV of the relevant kāraka (Sun's PAV for health, Lagna's PAV for intelligence, etc.). See whether the Antaradaśā graha contributes to the house-in-focus. If yes — predict that the house will activate during that Antaradaśā. If no — predict that the Antaradaśā will not significantly affect the house.
Example: Pisces Lagna, only 2 grahas contribute to the Lagna in Sun's BAV. Open Sun's PAV. The two contributing grahas are Sun and Mars. During Jupiter Mahādaśā, the Sun Antaradaśā and Mars Antaradaśā are the periods when Sun's BAV-contributors are activated. The native had exceedingly good health and took up physical exercise during those specific Antaradaśās. PAV-prediction confirmed.
The three levels are not redundant — they read at three different resolutions:6
SAV is a broad-spectrum scan. It tells you the general energy-distribution of the chart. Good for character-overview. Bad for events because it averages-out the planet-specific contributions.
BAV is a per-area diagnostic. It tells you, for any specific life-area, how much overall potential is concentrated there from the relevant kāraka's perspective. Good for medical/relational/career assessment. Bad for events because it doesn't tell you which planets are contributing — and Daśās run on specific planets.
PAV is the Daśā-mechanism reveal. By breaking down each rekhā into its contributing graha, it lets you align the Daśā schedule with the bhāva-activation schedule. The Daśā graha is either a BAV-contributor to the bhāva-of-focus or it isn't. If it is, the bhāva activates during that Daśā. If it isn't, the Daśā passes without much effect on that bhāva. This is the level where the system actually works for prediction.
Larsen is brutally honest about this: he calls SAV and BAV "not conducive to predictive accuracy." Most popular Vedic astrology textbooks teach only SAV (or SAV + BAV) and stop there. The result: students who know Aṣṭakavarga vaguely as "those rekhā numbers" but who cannot actually use it for Daśā-prediction. Larsen pushes past the popular ceiling to PAV — the level where Aṣṭakavarga becomes the predictive tool Parāśara intended.
This page connects to Vimśottari Daśā — PAV is specifically a Daśā-mechanism tool, and the Vimśottari schedule is what activates the contributing-graha-Antaradaśās PAV identifies.
It connects to Multi-Lagna Interpretive Sphere — each Lagna (Lagna, Ārūḍha Lagna, Kāraka Lagna, Navāmsa Lagna) has its own Aṣṭakavarga reading, and the practitioner can run PAV-checks from multiple reference points.
It connects to Avasthā Three Systems — the contributing-graha's Avasthā state during the relevant Antaradaśā shapes how the bhāva-activation actually unfolds. PAV tells you whether the house will activate; Avasthā tells you in what mood.
The 386-total-rekhā distribution is mathematically elegant — it's not arbitrary. Each graha contributes a specific rekhā-count determined by the classical 8-by-12 distribution tables. Jupiter contributes most (56) because Jupiter is the most universally-beneficial graha. Mars and Saturn contribute least (39 each) because malefics distribute fewer blessings.
Pisces Lagna native. The question: when will the native's health crises occur during their life? Standard reading: look at the 6th house, 8th house, and Sun's placement. You add Aṣṭakavarga PAV to sharpen.
You open Sun's BAV. Sun's contribution to the Lagna (Pisces) is 2 rekhā — below the average of 4. Health is below average overall.
You open Sun's PAV. The two grahas contributing those 2 rekhā to the Lagna are Sun itself and Mars.
Now you scan the Daśā schedule. Vimśottari running: Jupiter Mahādaśā active. You list the Antaradaśās:
You predict: during the Jupiter Mahādaśā, the only two periods when health will substantially activate (in either direction — and given low SAV, this means opportunity for strengthening rather than crisis) are Sun-Antaradaśā and Mars-Antaradaśā. The native took up physical exercise during exactly these periods.7
This is a far sharper prediction than "Jupiter Mahādaśā = good period, Saturn Mahādaśā = hard period." The PAV gives you Antaradaśā-level resolution on a specific bhāva.
Saturday afternoon. The native wants to know when to launch a new business. You scan the chart. Lagna lord Mercury, Ārūḍha Lagna in Virgo, 10th house unoccupied. You could read this generally — "good Mercury periods" — but you want specificity.
You open Mercury's BAV. The 10th house of the chart receives 5 rekhā from Mercury — above-average. Good news, the chart supports career-launches generally.
You open Mercury's PAV. You see which grahas contribute those 5 rekhā to the 10th. The contributors are: Sun, Jupiter, Venus, Saturn, and Mercury itself. Five grahas confirming the 10th-house career.
You check the Vimśottari schedule. The native is entering Saturn Mahādaśā. You list the first five Antaradaśās of Saturn Mahādaśā:
You predict: the native should launch in Saturn-Saturn or Saturn-Mercury — first two-and-a-half years of the Mahādaśā. They should expect a quieter period in Saturn-Ketu (year ~3-4). Then re-activation in Saturn-Venus (years ~4-7) and Saturn-Sun (year ~7-8).
The native asks why you're so specific. You show them the PAV table. They can see the math. The prediction isn't mystical — it's mechanical. Parāśara's brute-force Kali-Yuga tool, working as intended.
You stop at SAV and call it a day. You see the 10th house has 35 rekhā in SAV (above 32 average) and predict a strong career. You miss that which planets contribute those rekhā matters more than the total. The Daśā might cycle through five years of non-contributors before hitting a contributor.
You read BAV and skip the per-graha-PAV breakdown. You see 6 rekhā in Mercury's BAV for the 5th house (intelligence) and predict strong intellect. But you don't know which six grahas contribute, so when Mahādaśā schedules through, you can't predict when intellectual breakthroughs will actually occur.
You apply Aṣṭakavarga without checking which contributing-graha is currently in Avasthā during the Antaradaśā. PAV says Sun activates the Lagna during Sun Antaradaśā — but Sun is debilitated (Khala-Avasthā) and combust (Kopa). The activation happens but with shaking-and-rage texture rather than confidence. Pure Aṣṭakavarga reading without Avasthā layering misses the quality of the activation.
You include the nodes in BAV. The nodes are excluded from the eight-fold scheme. Some modern textbooks add Rāhu and Ketu BAVs — Larsen explicitly says this does not adhere to the views of the ancient astrologers. More graha-schemes means more than eight Vargas, which contradicts the word Aṣṭa itself.
You confuse Varāhamihira's variant (excludes Lagna varga; total 337) with the Parāśara default (includes Lagna varga; total 386). The averages differ — 28 vs 32 per house. Apply the wrong average and your "above average" calls are off by 4 rekhā in either direction.
Parāśara is the cited primary source (BPHS, Aṣṭakavargādhyāya). Larsen reproduces the rekhā-distribution and operational rules without modification. The three-level architecture (SAV → BAV → PAV) is implicit in the classical text and made explicit in Larsen's exposition.
Larsen's claim that "SAV and BAV are not conducive to predictive accuracy, only PAV is" is operationally important and is not universally taught in popular Vedic astrology. Many practitioners use SAV for "transit predictions" and treat it as predictively useful. Larsen's stance is that the transit application of SAV (Saturn transit producing different results based on the SAV of the transit-house) is a downstream BPHS application that does work — but for natal prediction, only PAV is reliable.
Rao's parallel exposition adds two layers worth holding alongside Larsen's. First, there is a layer beyond BAV that the Larsen treatment does not develop: the Sodhya Piṇḍa apparatus.8 Starting from each graha's BAV, two reductions are applied sequentially — Trikoṇa Śodhana (subtract the lowest value of each set of mutual trines from all three values; if all three are equal, zero them all) and Ekādhipatya Śodhana (a co-ownership reduction on the pairs of signs owned by the same planet, with specific rules for occupied-vs-empty rāśis). The result is the Śodhita Aṣṭakavarga (SoAV), and from SoAV one computes the Sodhya Piṇḍa of each graha by summing per-rāśi-multiplier products (Rāśi Piṇḍa) and per-graha-multiplier products (Graha Piṇḍa). The Sodhya Piṇḍa numbers are then used in transit-event-timing — a parallel predictive-apparatus to PAV that operates on transits rather than Daśās. See Sodhya Piṇḍa Event Timing for the operational mechanics. Second, Rao explicitly catalogues the Parāśara-vs-Varāhamihira controversies — three specific divergences in Moon's and Venus's benefic-house lists between BPHS (which Rao and Larsen both follow) and Varāhamihira (c. 600 AD).9 (1) For Moon: Parāśara holds Moon benefic in the 9th from Moon and malefic in the 9th from Mars; Varāhamihira inverts both. (2) For Moon: Parāśara holds Moon benefic in the 2nd from Jupiter and malefic in the 12th from Jupiter; Varāhamihira inverts both. (3) For Venus: Parāśara holds Venus benefic in the 4th from Mars and malefic in the 5th; Varāhamihira inverts both. Rao's argument for following Parāśara is technical: Parāśara transmits the Aṣṭakavarga rules three times in three different verses (count-of-malefic-references, list-of-malefic-references, list-of-benefic-references) creating a checksum-architecture resistant to manuscript-transmission corruption — a single-word change would have to be replicated in three consistent places to alter the apparatus. Rao reads this redundancy as the Sage's deliberate corruption-resistance design, and concludes the Varāhamihira variants are likely transmission errors rather than authentic alternative tradition. The vault holds the Parāśara version as canonical while flagging that the controversy exists and that the question is empirically open — both Larsen and Rao acknowledge that conclusive resolution would require systematic event-prediction testing against both rule-sets, which has not been done.
Open question: does Aṣṭakavarga's effectiveness vary by yuga as Parāśara implies? If the tool was specifically designed for degraded Kali-Yuga intellects, should we expect it to work better now (deeper into Kali Yuga) than in classical times? This is an interesting historical-claim that's not directly testable but informs interpretation.
Cross-lineage tension: Kurczak/Fish use Aṣṭakavarga sparingly, treating Vimśottari-Daśā mechanics as the primary timing tool. Pai's Frawley-lineage uses it more centrally. Foss focuses on per-graha bīja-mantra-architecture and treats Aṣṭakavarga as secondary. Larsen elevates PAV specifically to first-class predictive status. The cross-lineage convergence: Aṣṭakavarga exists and is classical. The disagreement: how prominently to feature it as a predictive workhorse.
Foss and Larsen are both Sanjay-Rath-direct disciples. Foss's Yoga of the Planets does not develop Aṣṭakavarga extensively — his pedagogical center-of-gravity is per-graha mantra-architecture. Larsen brings PAV to operational prominence. The two emphases are complementary: Foss tells you which mantras to recite for each graha; Larsen tells you which Daśā-Antaradaśā periods will activate which bhāvas via PAV-contribution analysis. A complete reading runs both — identify the PAV-activation periods (Larsen), prescribe the mantra-remedies (Foss).
Pai uses Aṣṭakavarga more centrally than Foss, but typically at the SAV/BAV level rather than PAV. Pai's Frawley-Trivedi lineage teaches Aṣṭakavarga as one of several diagnostic tools alongside per-nakṣatra reads. Larsen makes PAV the predictive workhorse.
Kurczak/Fish's Kriya-Yoga lineage is most skeptical of Aṣṭakavarga's centrality — they prefer Vimśottari-Daśā mechanics, Avasthā readings, and direct chart-analysis. They do not deny Aṣṭakavarga's validity; they de-emphasize it operationally.
DiCara's Gauḍīya Vaiṣṇava lineage treats Aṣṭakavarga more as background diagnostic than primary tool; per-nakṣatra Devatā readings carry the operational weight.
The cross-lineage map: all five lineages know Aṣṭakavarga exists and is classical. Pai and Larsen elevate it (Larsen specifically to PAV-level prominence). Foss, Kurczak/Fish, and DiCara use it as secondary. None deny it. The disagreement is operational prominence, not validity.
The PAV architecture — which specific contributors activate which specific outcomes — is not just astrological. It maps onto recurring patterns in network analysis, organizational dynamics, and complex-systems modeling.
behavioral-mechanics: Contributor Tracking as Influence Architecture — modern influence engineering (Hughes' Behavior OPS Manual, Cialdini's Pre-Suasion, Greene's strategic frameworks) all rely on the principle that outcomes are produced by specific contributors, not by aggregate forces. The popular reading of a political win is "the candidate was strong overall." The operator's reading is "X donors, Y demographic shifts, and Z media moments contributed the marginal votes; the strategy was built around activating those specific contributors at specific times." Larsen's PAV is the chart-form of this operator-reading. Each rekhā in a BAV is a specific contributor (a planet, coding for a person-type in the native's life). Each Daśā-Antaradaśā activates a specific subset of contributors. The handshake reveals that what Vedic astrology figured out about prediction-resolution (you need per-contributor breakdown to predict accurately) is what modern influence engineering rediscovered through network analysis: aggregate metrics like SAV always under-perform per-contributor metrics like PAV when you want actionable predictions. The cross-domain insight is that the resolution at which you read the system determines the quality of prediction you can make — both for charts and for influence campaigns. SAV-level thinking produces vague predictions. PAV-level thinking produces specific ones. The discipline transfers directly.
psychology: Which Internal Voice Contributes Which Outcome — Internal Family Systems (Schwartz) and Voice Dialogue (Stone) both treat the psyche as a parliament of parts, each contributing different motivational rekhā to different life-outcomes. The depressed-self contributes to the "stay in bed" outcome. The achiever-self contributes to the "answer the email" outcome. The artist-self contributes to the "make the painting" outcome. Reading the psyche aggregate-only ("he's feeling depressed today") is SAV-level. Reading per-contributor ("the depressed-self is contributing 5 rekhā to the social-withdrawal outcome while the achiever-self is contributing 2 rekhā") is PAV-level. The handshake reveals that depth psychology has independently arrived at the architectural principle Larsen names: outcomes are aggregate effects of specific contributor-parts, and intervention is more effective when targeted at the contributing-part than at the aggregate-mood. The cross-domain insight: both Vedic prediction-craft and modern psychotherapy improve their precision by resolving the aggregate into per-contributor signals. The chart and the parts-work are doing the same thing at different scales.
The Sharpest Implication
If only PAV is predictively accurate, then most popular Vedic astrology predictions are operationally one-level too coarse. The popular reader runs SAV, sees high totals in the 11th house, predicts wealth, and stops. The chart's actual predictive content lives one level deeper: which planets contribute the rekhā to the 11th, and when those planets become Antaradaśā active. Going to PAV multiplies the prediction-resolution by roughly 8x — instead of one "this Daśā is good for wealth" prediction, you get 8 distinct "this specific Antaradaśā activates this specific contributor-graha for wealth" predictions. The implication for practice: any consultation that doesn't run PAV-checks on the relevant bhāva is leaving 7/8 of the available predictive signal on the table. The remediation is operational: just run PAV. The information is in the chart. You just have to look at the right resolution.
Generative Questions