The Company Behind the Company

The standard story says La Réunion failed as a company. It didn’t — not the company that mattered. On paper there were two: a Brussels land-holding company barred by its own charter from operating anything, and a Texas operating company built eleven months later to do exactly what the parent legally could not. Only one of them collapsed. The other ran its charter to term, sold its remaining land in the charter’s final year, and never itself failed at all.

Two companies, one failure

The Société de Colonisation Européo-Américaine au Texas — SCEAT, trading as Bureau, Guillon, Godin & Co. — was constituted at Brussels on 26 September 1854, capitalized at $1,000,000 (5,400,000 fr., Art. 8), chartered to run twenty-one years, to 31 December 1875 (Arts. 1, 6). Article 3 is the load-bearing clause: it forbids the Society “any permanent exploitation, on its own account,” of the ground it holds. SCEAT was built, by its own charter, to buy land and resell it — not to farm it, not to run a colony on it.

Eleven months later, on 7 August 1855, a second company was stood up on the ground in Texas: the Société de Réunion. Its recovered charter — preserved only as principal articles quoted inside Savardan’s Naufrage au Texas, ch. VII; no full text survives in the record — states its object plainly (Art. 2): acquire the Réunion domain and develop it as “a first center of colonization.” That is precisely the operating role Article 3 forbade the parent. The subsidiary exists, structurally, because the holding company was barred from doing what a colony actually required someone to do.

Put the two side by side and the shape of the thing is visible at a glance:

Considerant himself sat on both sides of the seam. He was the movement’s founder, SCEAT’s salaried Executive Agent in Texas with full power of attorney (Title XIV, Art. 68), and — through a same-day Brussels side instrument, the Administrative and Distributive Committee — holder of five-sixths of his own founder’s appreciation-share, routed back toward the creditors of his earlier 1840 and 1843 ventures. On-the-ground authority and the Gérance’s legal authority were built to friction from day one: Cantagrel, the subsidiary’s own director, pushed for a 5% capital-advance rate at the drafting and was refused. He resigned within the year. Considerant later told Savardan and Cantagrel exactly why the rate — 3⅓%, half what Cantagrel asked — was set so low: “I fixed this rate expressly to force them to go away, if they do not know how to live on it.” That is not a founder describing an accident. It is a founder describing a mechanism.

What the money actually did

The founding threshold — $100,000 subscribed — was declared met on the incorporation date itself, 26 September 1854, on pledges that included $20,000 each from Albert Brisbane and Jean-Baptiste Godin, with Considerant personally standing surety (Art. 31). What came in after that is where the sources start to disagree, and the disagreement itself is a finding: Pratt says Au Texas “had stimulated subscriptions of 650,000 francs” — roughly $130,000 — describing what reads as an early wave. Verlet, quoting the supervisory board’s own report, says that of the $1,000,000 called for, “only a fifth would in the end be subscribed” — roughly $200,000. Neither source dates its figure precisely enough to prove Pratt’s number is simply an earlier snapshot of Verlet’s, but they agree on the order of magnitude: somewhere around a fifth, or less, of the authorized capital was ever raised. Two identifiable streams within that total: Bürkli raised over 150,000 francs in Switzerland; Goetseels put in roughly 30,000 francs (about $6,000) — the largest individual colonist stake on record.

The one accounting that survives for where the money went is Bürkli’s own 1858 apologia, and it should be read as an apologia. Of nearly 2,000,000 francs, he writes, 120,000 fr. were “squandered” at the outset on travel, preliminary work, “unavoidable misfortune” — and, he concedes, self-inflicted blunders: a roughly 6% write-off he treats as the ordinary cost of getting an expedition off the ground. His 2,000,000-franc denominator is larger than either subscription figure above, which suggests it folds in the colonists’ own belongings and savings alongside the company’s capital — a base chosen, whether by design or not, to make the loss look smaller than it was against the company’s own money alone. Bürkli expected the land itself, bought cheap in a money-short Texas, to “cover the loss incurred in the first years” through resale. That expectation is the entire SCEAT business model in one sentence, and it is also, per Article 3, the only thing the parent company was ever legally allowed to do with the ground it held.

Meanwhile Savardan counted the shareholder presence in Texas directly: the shareholder element at Réunion “comprised only a score of persons,” and only ten of them held the $125 in stock needed to sit in a general assembly. The capital stayed in Europe. The emigration was of labor. Spending on the ground, glimpse by glimpse, all from Savardan unless noted: $1,000 (5,200 fr.) to provision the avant-garde’s canteen with 7,000 lbs of goods; $100 for Considerant’s own carriage and $400 for his horses; upward of $3,000 sunk into the great garden against under $300 of produce out of it; a first wheat crop grown at a cost of $3 a bushel against a 75¢ market (Sandell); a $2,000 réméré loan extended to the land speculator Cordova — company money lent out at interest while, by the colonists’ own account, productive works on the ground were refused “a few thousand dollars”; repatriation advances of $100–200 per family, sized to how loudly each family demanded one.

The wind-down prices close the loop. Colonists offered to buy back the Horton tract at $8 an acre; Cousin refused and held out for $15, against Paris’s own stated wishes (Pratt). When liquidation finally came, stockholder-colonists were credited land at $4 an acre (Sandell, twice). And in 1875 — the charter’s last legal year, not a moment of crisis — Cantagrel bought what remained of SCEAT’s land. The company did not die in the wreck. It sold its inventory on schedule, twenty years later, exactly as Article 3 always said a land-resale company would.

The cession that mattered

The hinge is 7 August 1855, the day SCEAT ceded the whole domain to the new Société de Réunion. Considerant, as the parent’s own executive agent, signed the cession and personally subscribed “the complement of the sum fixed as necessary for the constitution” of the subsidiary he was creating — the parent’s own man underwriting the very entity that would absorb the operating risk (Art. 18). Per the supervisory board’s own report, quoted in Verlet: the domain, plus “an important sum of our social fund,” went to the subsidiary, and in exchange the parent “received shares of the new Society for a sum of 50,000 dollars” — leaving SCEAT’s Brussels shareholders with, in the board’s own words, “a mere claim on the subsidiary” where land title had stood a day earlier. The board was “moved”; management disapproved; the transfer went through regardless.

Savardan, writing as a contemporary and not a historian, registered two objections at the time that the record has since confirmed as structurally important rather than incidental: the cession handed over roughly thirteen thousand acres “when a third of it would have perfectly sufficed,” and it did so “without estimate, inventories, or verification.” No one priced what was being transferred before it moved off the parent’s books and onto the subsidiary’s. That is not bookkeeping carelessness. After 7 August 1855, whatever happened to that land could only reach Brussels’ shareholders through $50,000 of paper in a company that collapsed within two years — a firewall, whether anyone intended it as one or not.

Five tensions

Article 3 versus the Société de Réunion. The parent’s charter forbids it any permanent exploitation of its own land; a second company was built the same year to do precisely that. “The colony failed” and “the company did exactly what its charter said” are both true at once — they just describe two different corporations.

Considerant’s three hats. Founder, salaried agent, and beneficiary of his own appreciation-share meant that the man setting the subsidiary’s capital-advance rate at 3⅓% instead of the 5% his own director asked for was, by his own later admission, using that rate to drive people away. The subsidiary’s terms functioned as an instrument of attrition, not of shared enterprise.

Charter life versus operating life. Chartered to 1875; operational collapse by roughly 1857. It is the subsidiary that collapses. The parent survives as exactly what its charter always said it was — a land company holding for resale — and its 1875 sale of the remaining acreage is a scheduled wind-down, not a failure.

Capital and labor were different people. The subsidiary’s statutes built an elaborate mechanism for splitting products and appreciation between “providers of funds” and “providers of associated labor” (Arts. 10–11) — a mechanism that presumes both are present. In practice, some twenty shareholders reached Réunion, ten of them quorum-qualified, while capital itself stayed in Europe and only labor emigrated. The machinery built to reconcile two classes had almost no one on one side of it.

The cession severed the shareholders from the land. Before 7 August 1855, SCEAT’s Brussels shareholders held a company that owned Texas land outright. After it, they held $50,000 of stock in a subsidiary — “a mere claim,” in their own board’s words. Whatever went wrong in Texas after that date was, legally, someone else’s balance sheet.

What’s still missing

This argument is built entirely on what the sources above actually say, but three gaps in the record keep parts of it provisional, and one date conflict is unresolved rather than settled.

The Société de Réunion’s full statutes are not in this archive. Savardan preserves the principal articles he chose to quote, not the instrument itself; Cordillot refers to the statutes as having been published, which means a fuller text likely survives somewhere — the Bulletin de la colonisation is the leading candidate. The subscription total is bracketed, not settled: Pratt’s 650,000 francs and Verlet’s “a fifth” of $1,000,000 are probably an early wave and a final figure rather than two estimates of the same number, but neither source dates itself precisely enough to prove it. Founding land prices are likewise open — the 1855 Horton, Coombs, McCracken and Sheppard deeds carry no consideration amounts in the record as transcribed, so the land side of “what was deployed” still rests on Bürkli’s qualitative “at high prices” rather than a figure. And the founding date itself is contested: the statutes place constitution on 26 September 1854, matching the project’s working timeline, while Cordillot gives 14 September 1854 — most likely a deed-versus-constitution gap (the acts were deposited with notary Heetveld on 4 October 1854), but that is a plausible reconciliation, not a confirmed one.

The fuller source apparatus behind every figure above — article numbers, witness reliability, what each source is positioned to know — is in the findings.