Allika through its neighbours: what the 1935 plans changed
Two plans recognise the same neighbouring boundary from opposite sides. What does this check add to the history of Allika A10, and why must a detached plot be distinguished from the present yard?
Checking one boundary with two documents
A farm's history can begin in the wrong place. The document may be genuine and the map carefully drawn, yet neither belongs to the yard being researched. That is what happened when our investigation connected present-day Allika with Piiritusevabriku A11. The working identification now follows Allika A10, with an important check supplied by the boundaries of neighbouring land.
This article continues and corrects our earlier account of the 1935 boundary plans. Its new question is how one plan can test another: which identifiers agree, which neighbours recur, and where the evidence stops. The place under investigation is Allika in Oandu, in today's Lüganuse municipality, Ida-Viru County. Keeping that location explicit prevents a familiar name from carrying us into another property's history.
Why the name alone is not enough
Finding “Allika” in an archive feels like arriving at an answer. But a historic name does not by itself identify a modern farmyard. The terrain, parcel designation, property number, date and purpose of the record must be considered together. A mismatch need not make a document useless: it may describe another Allika or a detached part of the same holding.
Our earlier identification of Piiritusevabriku A11, property 12526, with the present farm has been withdrawn. Buildings, areas and economic functions shown on that plan cannot be transferred to today's Allika. The plan remains evidence about its own site. Authenticity of a document and accuracy of our identification are two separate questions.
That correction has a practical consequence. Future accounts must start from the revised evidence, while earlier accounts must be read with their correction notice. An attractive story cannot justify retaining a connection that no longer withstands comparison.
Two land changes and a precise identifier
Two 1935 plans in the Rahvusarhiiv catalogue give the investigation a more specific starting point. ERA.T-3.22.2534 concerns Lit. C, separated from Maidla manor land, property 80, and attached to Allika no. A10, property 2067. ERA.T-3.22.2535 describes the opposite direction: Lit. A10a is separated from Allika A10 and attached to the manor land.
The complete wording matters. Both titles pair A10 with 2067. They describe changes to identified pieces of land, rather than merely mentioning a resident or a nearby village. The documentary designation Allika A10 / 2067 is therefore a verified fact in these records.
The wording about attachment does not, on its own, establish when every legal step was completed. A plan identifies the subject of a land arrangement. A history of ownership needs the relevant legal records as well as the drawing. Nor do these two titles explain the participants' motives or authorise us to invent a negotiation between neighbours.
Reading the edge of a larger property
A boundary plan may concentrate on a piece of land rather than show the entire farm. For someone seeking a house, this can seem frustrating. Yet the edge contains useful tests: adjacent designations, a road, and the way several boundaries meet. These details offer more than a shared name.
On the Lit. C plan, the relationship between A10, neighbouring designations and the road supports the current spatial line of inquiry for the farmyard. This is a spatial argument, not a continuous legal chain to the modern cadastral parcel. It should remain open to testing against further records.
The distinction lets us make progress without pretending to have answered everything. We can improve an identification of land while still lacking its legal owner's name in a particular year. We can investigate continuity of a yard without dating every surviving building. A map is valuable because it answers a limited question clearly.
A neighbour checks the neighbour
A further check comes from the plan for Õunapuu A3, property 3961, after the separation of Kooli A19: ERA.T-3.22.1768. The catalogue dates its preparation to 1932–1933 and its approval to 25 January 1934. In its detail, the detached A3c plot has A10a beside it. On the plan of the land separated from Allika, A10a in turn has A3c as a neighbour.
This is a reciprocal check. The same pair appears from opposite sides in two documents. It helps test whether a designation has been read correctly and whether it belongs in the proposed spatial context. Unlike a loose resemblance between names, the comparison can be repeated on the plans themselves.
Its limit is equally important. A3c and A10a are detached plots; recognising their shared boundary does not locate today's Allika house on either one. A holding's separate allotment and its principal yard can belong to one land history while occupying different places. The check supports the identity of the neighbouring plots, not an inference about buildings that the plans do not establish.
What the comparison resolves
We now have a documented name-and-number pair, Allika A10 / property 2067; plans describing boundary changes; and reciprocal adjacency between A10a and A3c. The Lit. C plan provides a separate spatial line of inquiry for the present yard. Together these results narrow the next search beyond the vague instruction to find anything called Allika.
Ownership transfers, residents by year and the full relationship to modern boundaries remain open. The catalogue records EAA.4187.1.10761 and ERA.62.4.4668 are leads for further reading. Their existence does not disclose the contents of unread pages, the names of owners or the terms of transactions.
Earlier numbers also require care. A parcel marked 10 is not automatically A10, nor does 10a necessarily mean A10a. Dates, neighbours, scale and the purpose of each map must be compared. A difference in area needs an explanation supported by evidence; it cannot simply be smoothed away to make two records agree.
Why publish a correction?
Restoration is visible in photographs. Historical research changes less visibly: sometimes its most useful result is removing a mistaken connection and replacing it with a question that can actually be tested. Recording that change helps the next reader avoid repeating it.
The author's winter photograph shows the modern landscape that motivates this research. It does not show the boundaries of 1935. Historic plans and present-day photographs serve different purposes, and their roles should stay clear when they appear in the same article. We link to the archival originals rather than reproduce scans with unresolved publication rights.
If you have a record naming Allika A10, property 2067 or related boundaries in this Oandu, a precise archive reference, date and link to an accessible copy would be especially useful. Please send them through the site's contact page. A recollection can also suggest a direction, provided it is identified as a recollection. The next useful chapter will come from a source that answers one of these remaining questions.
The titles of ERA.T-3.22.2534 and ERA.T-3.22.2535 explicitly pair Allika A10 with property 2067.
The two inspected plans show A3c and A10a as reciprocal neighbours; these are detached plots.
The identification of the present yard with A10 remains a spatial inference; the former A11 / 12526 identification is withdrawn.
The continuous chain of legal ownership and the relationship between earlier and later numbers need further evidence.