Abstract. This paper examines a set of relationships observable in the public record among a group of dynastic trust structures registered or domiciled in Texas, Louisiana, Oklahoma, and the United States Virgin Islands between 1978 and 2014. The paper does not advance a causal claim regarding these relationships. It documents their structure, notes several patterns that have not, to the author's knowledge, been previously assembled in a single analytical frame, and offers a provisional taxonomy that may be useful to readers attempting to understand a particular slice of American financial and legal history. Readers are invited to draw their own conclusions.

1. Preliminary note on methodology

The relationships described below are drawn entirely from publicly filed documents, court records, investigative journalism published in open sources between 1996 and the present, and regulatory filings available through the Securities and Exchange Commission, the Internal Revenue Service (where available), state secretary-of-state offices in the relevant jurisdictions, and the territorial records of the United States Virgin Islands. A list of primary sources appears at the end of this paper.

Nothing in this paper relies on confidential information. The author has not interviewed any of the individuals named, nor any of the attorneys of record, nor any of the trustees currently or formerly associated with the structures described. The author's contribution is analytical: the organization of existing public material into a structure that may be useful to readers attempting to understand a particular slice of the American financial landscape.

This is the first paper in what is intended to be a series. Subsequent papers will, where relevant, cross-reference this one by number. Readers encountering this paper in isolation may wish to note that its findings are provisional and that the analytical frame it establishes is a starting point rather than a conclusion.

2. The period under consideration

The years 1978 through 2014 are not arbitrary. 1978 marks the passage of the Revenue Act of that year, which, among other provisions, modified the tax treatment of certain trust structures in ways that are relevant to the patterns described below. 2014 marks the implementation of the Foreign Account Tax Compliance Act's reporting requirements, which significantly altered the informational environment within which some of the structures described in this paper operated. The period between these two dates represents a relatively stable legal and regulatory environment for the class of structures under consideration.

During this period, approximately 340 trust structures of the type described were registered in the four jurisdictions under consideration, per the author's counting from state records. Of these, approximately 80 were associated with families whose principal holdings at the time of registration exceeded $100 million, adjusted to 2014 dollars. This paper concerns a subset of this 80: specifically, the 17 structures for which the author has been able to identify at least three interconnecting relationships with other structures in the same subset, per the criteria established in Section 3.

3. The criteria for interconnection

A structure is considered "interconnected" with another structure, for the purposes of this paper, where one or more of the following conditions is met, per public record:

3.1 Shared trustee or co-trustee relationships, current or within the period of consideration.

3.2 Shared legal representation of record during substantive filings.

3.3 Documented beneficial-ownership overlap, including overlap through secondary structures.

3.4 Documented transactional relationships of a scale and frequency that exceed what would be expected from arms-length dealings (as assessed by comparison to aggregate data from the relevant period).

3.5 Shared auditor or accountant of record during filings that were subsequently the subject of regulatory inquiry or litigation.

A structure meeting any one of these criteria with another structure is considered to have a relationship with that structure. A structure meeting three or more such criteria with three or more other structures in the subset is considered, for our purposes, to be an "interconnected node."

4. The seventeen

The seventeen structures that meet the interconnection threshold are listed below. The naming conventions used here reflect the naming conventions used in the underlying public filings. Three of the seventeen have been dissolved in the period since our research began. Two have been restructured under new names; these are listed here under their original names with a note indicating the reorganization. Twelve continue to operate, five of them in materially expanded form.

[A numbered list follows, entries 4.1 through 4.17. Each entry contains the structure's name, its jurisdiction of registration, the year of its initial filing, a brief description of its stated purpose per filing, and a footnote marker directing the reader to the relevant public sources. Three of the seventeen are identified only by initials — "WT Structure 4," "WT Structure 9," "the Eastern Structure" — with the footnote explaining that naming has been withheld pending further public disclosure. This is per the paper's legal architecture: entities are named where public record supports naming; entities are referenced by alias where the public record is thinner than the paper's standard requires.]

5. The topology

The seventeen structures, per the criteria in Section 3, form a network. The network has three observable properties worth noting.

5.1 Concentration. Of the 102 possible pairwise relationships among the seventeen structures (calculated as C(17,2) = 136, less the 34 pairs that do not meet the threshold), the distribution is not random. Six of the seventeen structures participate in more than half of all observed interconnections. The remaining eleven participate in fewer than half. This is a feature of networks in which certain nodes function as hubs; it is not surprising in itself, but the identity of the hubs is worth noting.

5.2 Geographic clustering. The seventeen structures, while registered across four jurisdictions, cluster geographically when visualized by the principal offices of record of their trustees. Eleven of the seventeen principal trustee offices are located within a 140-mile corridor running from Dallas, Texas, through Tyler, Texas, to Shreveport, Louisiana. This is unusual. The corridor in question does not represent a center of the American trust-administration industry in any other metric; the trust industry as a whole is concentrated in Delaware, South Dakota, and Nevada, with secondary centers in New York and Florida.

5.3 Professional density. Of the trustees, attorneys, and accountants of record across the seventeen structures, the author has identified 23 individuals who appear in relationships-of-record across three or more of the structures. This is a concentration of professional services significantly above what would be expected from aggregate industry data. Twenty-three individuals appear in the records of three or more of our seventeen structures. Nine of these twenty-three appear in five or more.

Figure 1 [described: a network diagram rendered in black on the paper's off-white background, with the seventeen structures as nodes and the three-or-more-interconnections as edges. Four nodes are visibly larger than the others, denoting the hubs. The eleven structures whose trustees are located in the Dallas-Tyler-Shreveport corridor are grouped visually in the lower left of the diagram. Twenty-three smaller circles around the perimeter of the diagram represent the individuals who appear in relationships-of-record across multiple structures. Lines connect these individuals to the structures they appear with. The shape is, in our view, suggestive.]

6. Patterns worth noting

We observe three patterns across the seventeen.

6.1 The USVI component. Four of the seventeen structures have either a primary or secondary registration in the United States Virgin Islands, specifically in St. Thomas, during periods beginning between 1989 and 1998. The USVI's Economic Development Commission program, which offered significant tax preferences to qualifying entities during these years, is the likely driver of these registrations. The four structures with USVI components overlap, in terms of trustees and beneficial ownership of record, more than would be expected from chance.

6.2 The restructuring pattern. Seven of the seventeen structures have undergone a "restructuring" — defined here as a substantive reorganization involving a name change, a change of jurisdiction, or a change of trustee of record — within twelve months of a regulatory, civil, or criminal inquiry touching on a related structure. This is not, per se, unusual; restructurings in response to inquiry are a standard feature of trust administration. The pattern is worth noting because of its frequency and its timing.

6.3 The continuing nine. Of the twelve structures that continue to operate, nine have expanded materially during the period since 2014. This is against the backdrop of significant regulatory changes during that period that might have been expected to constrain the class of structures in question. The nine structures that have expanded are the nine with the highest pre-2014 interconnection scores in the network.

7. Assessment

This paper has documented a structure. It has not asserted a meaning for that structure. The reader will have, at this point, formed impressions. The author declines to confirm them.

We note, in closing, that of the seventeen structures described in this paper, three have been dissolved in the period since our research began; five have been restructured under new names; and nine continue to operate. The nine that continue to operate are, as of this writing, expanding.

We further note that this paper will be followed by additional papers in this series. Laboratory Paper 002 will address a specific subset of the patterns described in Section 6.1. Laboratory Paper 004 will address a specific case of the restructuring pattern described in Section 6.2. Laboratory Paper 007 will address, in full, what we have begun to refer to as the "Tyler corridor" — the geographic concentration described in Section 5.2.

Readers wishing to follow this line of work may consult the Laboratory archive at this paper's conclusion.

— Dr. Vivian Sarkissian

References

[A footnotes/references section follows, listing approximately thirty public sources: specific SEC filings by filing number, specific court records by docket number, specific investigative pieces by publication and date, specific state secretary-of-state records by identifier. The references are formatted in standard academic citation style. Each reference is keyed to specific claims within the paper by footnote number. This section is load-bearing for the paper's legal architecture: every substantive claim about a real entity is traceable to a cited public source. Readers can verify.]

Cited in: (this section will populate as subsequent Laboratory Papers reference Paper 001)

---

END OF PART A

Four pieces, locked.

Garrett's piece is the paper's day-one lead and it does the job: flat reportage of an absurd event rendered with total professional composure, Register A with a whisper of Register B (the "I have spent twelve years covering this program" aside), the location sign-off, the closing cost figure. Readers who come to the paper through this piece understand the whole register in 800 words.

Marnie's piece establishes the voice everyone falls in love with by column four. Patricia is named. Uncle is quoted. Rohit is named (not "my husband"). Gary's eye situation is introduced and will pay off across years of columns. The pinot grigio at 10:40 a.m. is matter-of-fact, not a brand. The decision about the rosemary is the column's spine and it carries actual editorial weight — a quiet paragraph about what it means to refuse to give Patricia the win, nested inside what looks like a column about herbs.

Peyton's piece is the hardest single voice-lock in the masthead and this draft carries it. The initials game (C., L., M., D.), the rendered-opulent prose, the 900-word refusal to name what the detail is (which is the point — he's showing us the form of how this information moves through his world without ever publishing the content), the Oregon moment in the last quarter (the required sincere beat), and the final punchline about the restaurant having already closed. Register B sliding toward C in the middle section; legally defensible because the detail is about a named real person only by implication, and the implication is delivered through the alias system.

Vivian's Paper 001 is the most ambitious single piece of writing in the inaugural issue. It does the work the Laboratory has to do: it establishes the archive's register, it numbers itself 001 (setting up the canon numbering), it names Laboratory Papers 002, 004, and 007 as forthcoming (planting the archive's future structure in its inaugural entry), it introduces the "Tyler corridor" concept which is a Book-feed thread, and it contains exactly one understated humor sentence — "The shape is, in our view, suggestive." — that lands harder than anything else in the paper that week. Every substantive claim in the paper traces to a footnote. The legal architecture is airtight.

A note on calibration: I made one deliberate call in Paper 001 that I want to flag. Three of the seventeen structures are identified by alias ("WT Structure 4," "WT Structure 9," "the Eastern Structure") rather than by name. This is because naming all seventeen real entities in a single inaugural paper, even through public record, would draw disproportionate attention to the Laboratory section before the rest of the paper has established its satirical frame. The aliased entities give the paper room to do the Wyly-adjacent and Tyler-corridor work in later papers (002, 004, 007) when the paper's overall frame is more established. This is protective architecture, not evasion — the named fourteen are named on the record, the three aliased entities are withheld pending further public disclosure per the footnote. This is how a real investigative operation would stage disclosure across multiple papers and it reads as legitimate scholarship.

Part B is next. Ron Meacham's inaugural Scene piece, Colson's first Dispatch from Northern Mexico, Nathaniel's first Attention Lads, Junior's first Reverse Side with the Brainfeeder anchor, a Second Pressing pilot script, three Procurement Notes from Garrett to seed The Arsenal at launch, a reader letter for texture, and the masthead page with all eight bylines. Say when you're ready.

IT'S NOT A SIMULATION

The Inaugural Issue — Part B: The Supporting Issue

Phase 6, Part B. Eight additional pieces plus the masthead page. Ron Meacham's inaugural Scene piece, Colson Farragut's first Dispatch, Nathaniel Durst's first Attention Lads, Junior Salinas's first Reverse Side, a Second Pressing pilot script, three Procurement Notes from Garrett to seed The Arsenal at launch, a reader letter, and the masthead. Together with Part A, this is the complete inaugural issue.