Get ebook downloads in full at ebookmeta.

com

Proofs and Models in Philosophical Logic Greg

Restall

https://ebookmeta.com/product/proofs-and-models-in-
philosophical-logic-greg-restall/

OR CLICK BUTTON

DOWNLOAD NOW

Explore and download more ebook at https://ebookmeta.com

 Recommended digital products (PDF, EPUB, MOBI) that
you can download immediately if you are interested.

Logic and Language Models for Computer Science Fourth

Edition Dana Richards

https://ebookmeta.com/product/logic-and-language-models-for-computer-
science-fourth-edition-dana-richards/

ebookmeta.com

Cavell s Must We Mean What We Say at 50 Cambridge

Philosophical Anniversaries Greg Chase

https://ebookmeta.com/product/cavell-s-must-we-mean-what-we-say-
at-50-cambridge-philosophical-anniversaries-greg-chase/

ebookmeta.com

Seven Myths of the Spanish Conquest Matthew Restall

https://ebookmeta.com/product/seven-myths-of-the-spanish-conquest-
matthew-restall/

ebookmeta.com

Private Actors in International Investment Law 1st Edition

Katia Fach Gómez

https://ebookmeta.com/product/private-actors-in-international-
investment-law-1st-edition-katia-fach-gomez/

ebookmeta.com
Saint s Passage 1st Edition Elizabeth Hunter

https://ebookmeta.com/product/saint-s-passage-1st-edition-elizabeth-
hunter/

ebookmeta.com

Practical SQL: A Beginner's Guide to Storytelling with

Data, 2nd Edition Anthony Debarros

https://ebookmeta.com/product/practical-sql-a-beginners-guide-to-
storytelling-with-data-2nd-edition-anthony-debarros/

ebookmeta.com

Cattywampus 1st Edition Ash Van Otterloo

https://ebookmeta.com/product/cattywampus-1st-edition-ash-van-
otterloo/

ebookmeta.com

Tessa 1st Edition Bayleigh Rae

https://ebookmeta.com/product/tessa-1st-edition-bayleigh-rae/

ebookmeta.com

Sugar Walls A MFM Menage Romance Sweet Treats Book 5 1st

Edition S E Law Law S E

https://ebookmeta.com/product/sugar-walls-a-mfm-menage-romance-sweet-
treats-book-5-1st-edition-s-e-law-law-s-e/

ebookmeta.com
Analytical Lens Design 2nd Edition Rafael G Gonza´Lez-
Acun˜A

https://ebookmeta.com/product/analytical-lens-design-2nd-edition-
rafael-g-gonzalez-acun%cb%9ca/

ebookmeta.com
Philosophy and Logic

Proofs and Models

in Philosophical
Logic

Greg Restall

ISSN 2516-418X (online)

ISSN 2516-4171 (print)
 Elements in Philosophy and Logic
edited by
Bradley Armour-Garb
SUNY Albany
Frederick Kroon
The University of Auckland

P RO O F S A N D M O D E L S I N
PH I L O S O P H I C A L L O G I C

Greg Restall
University of St Andrews
 University Printing House, Cambridge CB2 8BS, United Kingdom
One Liberty Plaza, 20th Floor, New York, NY 10006, USA
477 Williamstown Road, Port Melbourne, VIC 3207, Australia
314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre,
New Delhi – 110025, India
103 Penang Road, #05–06/07, Visioncrest Commercial, Singapore 238467

Cambridge University Press is part of the University of Cambridge.

It furthers the University’s mission by disseminating knowledge in the pursuit of
education, learning, and research at the highest international levels of excellence.

www.cambridge.org
Information on this title: www.cambridge.org/9781009045384
DOI: 10.1017/9781009040457
© Greg Restall 2022
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2022
A catalogue record for this publication is available from the British Library.
ISBN 978-1-009-04538-4 Paperback
ISSN 2516-418X (online)
ISSN 2516-4171 (print)
Cambridge University Press has no responsibility for the persistence or accuracy of
URLs for external or third-party internet websites referred to in this publication
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
 Proofs and Models in Philosophical Logic

Elements in Philosophy and Logic

DOI: 10.1017/9781009040457
First published online: March 2022

Greg Restall
University of St Andrews

Author for correspondence: Greg Restall, [email protected]

Abstract: This Element is an introduction to recent work on proofs and

models in philosophical logic, with a focus on the semantic paradoxes and
the sorites paradox. It introduces and motivates different proof systems
and different kinds of models for a range of logics, including classical logic,
intuitionistic logic, a range of three- and four-valued logics, and
substructural logics. It also compares and contrasts the different
approaches to substructural treatments of the paradox, showing how the
structural rules of contraction, cut and identity feature in paradoxical
derivations. It then introduces model theoretic treatments of the
paradoxes, including a simple fixed-point model construction that
generates three-valued models for theories of truth, which can provide
models for a range of different non-classical logics. The Element closes
with a discussion of the relationship between proofs and models, arguing
that both have their place in philosophers’ and logicians’ toolkits.

Keywords: proofs, models, logic, semantics, paradox

© Greg Restall 2022

ISBNs: 9781009045384 (PB), 9781009040457 (OC)
ISSNs: 2516-418X (online), 2516-4171 (print)
 Contents

1 Context 1

2 Proofs 9

3 Models 38

4 Connections 58

Glossary of Symbols 73

References 77
 Proofs and Models in Philosophical Logic 1

1 Context
As far as academic disciplines go, logic is strange. In the western academy,
its roots go back to Aristotle, to Euclid, to the Stoics, through medievals, the
Arabic world, and into a flowering complexity in the nineteenth and twentieth
centuries, as philosophers and mathematicians grappled with understanding the
power and limits of deductive reasoning. The field we now know as modern
logic took root in the project of systematising and securing the foundations of
mathematics1 and in giving an account of the relationship between those math-
ematical theories and our experience of the world around us. In the twentieth
century, new connections emerged with the nascent fields of linguistics, digital
systems and computer science. There is no way that an Element on the use of
proofs and models in philosophical logic could do justice to anything more than
a tiny fragment of this sprawling edifice.2
So what small fragment of the discipline of logic will this Element address?
As the title states, our focus is philosophical logic and the twin roles of proofs
and models in the development of logic. The philosophical concern will also
be twofold: we will reflect on the application of logic to some questions in
philosophy and, at the same time, consider a philosophical reflection on the
discipline of logic itself. Philosophical logic provides both a set of sensibilities
and processes and tools for application in philosophical discourse (among other
kinds of discourse), and at same time, it is a site of philosophical reflection. We
will maintain these dual perspectives on our topic throughout this Element.
In this first section, I set the scene by way of an introduction to how we can
use the different tools of proofs and models to form judgements about logical
validity and invalidity. Then I outline how attention to proofs and models plays
a role in some of the current debates in philosophical logic, concerning the
semantic paradoxes and vagueness. I then end the section by looking ahead to
the argument of the remainder of the Element.

1.1 Proofs and Models

There are many ways to look at logic and the constellation of concepts that
logicians have attempted to analyse using proofs and models. One way to do

1 J. Alberto Coffa’s The Semantic Tradition from Kant to Carnap (1993) tells the compelling
story of the growth of modern logic beyond its Aristotelian bounds as Bolzano, Weierstrass and
others attempted to make sense of the mathematically important notions of convergence and
continuity.
2 So I will not cover the rich tradition of proof complexity, Gentzen’s consistency proof for arith-
metic, the connections between proof search and decidability and many more interesting topics
in proof theory. Neither will I discuss a great deal of model theory, such as significant meta-
theoretical results including compactness, cardinality of models or ultrapowers, and other model
construction techniques. This Element is only so long.
2 Philosophy and Logic

this is to focus on the production and the evaluation of argumentation, reasoning

and inference. Let’s start with a simple example, involving two mathematicians,
who are reasoning about some newfangled binary relation R they are exploring.
They have discovered that the relation R is transitive (i.e., for any objects x, y
and z if R relates x to y and R relates y to z, then R relates x to z too, so is
older than is an example of a transitive relation, while is a parent of is not)
and symmetric (if R relates x to y, then R relates y back to x too, so neither is a
parent of nor is older than are symmetric relations but is a sibling of is), and
finally, the relation is directed (for each object x, there is some object y where
R relates x to y, so, on the collection of non-negative natural numbers (natural
numbers for short) 0, 1, 2, . . ., the relation is smaller than what is directed, since
for every number x we can find some whole number y where x is smaller than
y, but the relation is larger than what is not directed if we restrict our attention
to the natural numbers. There is no natural number y where 0 is larger than y).
So our two mathematicians agree that this newfangled relation R is transitive,
symmetric and directed. One of our mathematicians exclaims: ‘The relation
R is reflexive, too!’ (A relation R is reflexive if, for every object x, R relates
that object to itself ). Our second mathematician asks: why is that? The first
responds:

(1) R is transitive. It’s symmetric. It’s directed. It must be reflexive, too!

The second mathematician doesn’t see why this is the case, so they ask for the
reasoning to be spelt out. Can the leap from the premises to the conclusion be
broken down into smaller steps? It can. Our quick thinker responds:

(2) Take an object a. Since R is directed, there is some object b where R

relates a to b. Since R is symmetric, R also relates b to a. Now, R relates
a to b and R relates b to a, so since R is transitive, R relates a to a. So
we have just shown that for any object a at all, R relates a to itself. That
is, R is reflexive.

This elaboration is one way to spell out the jump from the premises to the con-
clusion. It is what we call a proof. It fills out that large jump in thought in
terms of smaller steps. In this case, the smaller steps involve the applications
of agreed-upon definitions (unpacking the definitions of the terms reflexivity,
transitivity, directedness and symmetry), the operations of individual logical
concepts like the universal and existential quantifiers (in the concept of direct-
edness, e.g., for every x, there is some y where R relates x to y), and other logical
concepts like conjunction (the and in the definition of transitivity) and condi-
tionality (the if in the definitions of transitivity and symmetry). If you were
to question this proof at any of the steps in the explanation, it would seem to
 Proofs and Models in Philosophical Logic 3

be a very different problem than not understanding a large leap in thought. It

would be a problem of understanding the concepts in use and not a problem
of understanding how those concepts are combined. (That is the idea, anyway.
Exactly how proofs work, and what options we have in understanding them is
the topic of the next section.)
Proofs are one side of the logical coin. Models are the other. To explain
models, we might consider another possible transition in thought: suppose our
mathematicians have another newfangled relation S, and they have concluded
that S is transitive and symmetric, but they do not know whether S is directed
or not. They wonder, does it follow from these two properties (transitivity and
symmetry) that S is reflexive? We can see that the reasoning supplied previ-
ously concerning R does not apply in the case of S, since we do not know that
S is directed. But having one potential proof that doesn’t work does not mean
that there isn’t another proof that does. Is it the case that when a relation S is
symmetric and transitive, then it must also be reflexive? Our mathematicians
think for a while, and sketching some ideas on a blackboard, they draw the
following diagram:

This diagram represents a way a relation S could be. Each dot is a different
object in the domain of the relation, and a line connecting dot x with dot y
indicates that S relates object x to object y. By design, the relation depicted is
symmetric, since if there is a line connecting x with y it is line that connects y
with x. We can see, too, that whenever we can get from x to y and from y to
z by lines, there is a direct line from x to z. This holds even in the case where
we can get from x to y and back – whenever there is a line from x to anywhere
at all, there is a line looping around from x to itself. So the relation depicted
here is transitive too. However, it is not reflexive because we have an isolated
dot in our diagram. This lonely object is not a counterexample to the claim of
symmetry or of transitivity for S, but it shows that if S were like this, it would
not be reflexive. We have a counterexample to the rule that symmetric and
transitive relations must be reflexive.
This counterexample is a model. It is not a claim about how the origi-
nal relation S is. It is a sketch, a representation, showing that if we want an
explanation why a relation S is reflexive, we cannot appeal merely to its transi-
tivity and its symmetry, since the argument breaks down in circumstances like
those depicted in this model. A counterexample is a barrier through which our
argument cannot pass.
4 Philosophy and Logic

Here, in a nutshell, we have the distinction between proofs and models. For
the first argument, we have provided a proof, leading from the premises to the
conclusion, and for the second, we have provided a counterexample, a model
that renders the premises true and the conclusion false.3
The proofs and models we have seen so far are relatively simple, involving
reasoning with the quantifiers all and some and familiar logical connectives
like if and and. There are important questions about the role of these concepts
in our argumentation and in our construction of proofs and models. There is
broad agreement that these quantifiers and connectives are important. There is
less agreement over whether there is anything categorically distinctive about
those concepts.4
 

However, there are concepts other than the familiar connectives and quantifiers
that have proved important for philosophical logic, and which are amenable to
treatment by way of proofs and models. One example is provided by modal
concepts, such as possibility and necessity. However, the proofs and models
appropriate for modal operators seem qualitatively different to the models we
have seen so far. They do not just represent a way things could have been but
also represent more than one such way that things could be. To see how they
arise, consider the difference between two different arguments:

(3) It’s possible that either p or q. So either it’s possible that p or it’s
possible that q.

We can fill in this reasoning into a proof in the following way:

(4) Since it’s possible that either p or q, we grant some possibility where
either p or q holds. Suppose it’s p. In that case, we can conclude that
(back where we started) it’s possible that p, and so it’s either possible
that p or it’s possible that q. On the other hand, suppose that the pos-
sibility we granted makes q hold. In that case (also, back where we
started), it’s possible that q, and again, either it’s possible that p or it’s
possible that q. So, in either case, it’s possible that p or it’s possible that
q, and we’re done.

3 Distinguishing validity as defined by way of proofs and validity defined by way of models was
a great conceptual advance in the twentieth century. Consult Zach (1999) for a discussion of the
early days of that development of the distinction.
4 There is extensive literature attempting to characterise the logical constants from other concepts.
We will not explore it here. Gila Sher’s The Bounds of Logic is a good place to start (1991).
 Proofs and Models in Philosophical Logic 5

In this proof, we broke down the leap from premise to conclusion into smaller
steps, using more fundamental principles governing possibility and disjunction.
This is a good candidate for being a proof.

Suppose, on the other hand, we tried a similar argument, with necessity in

place of possibility.

(5) It’s necessary that either p or q. So either it’s necessary that p or it’s
necessary that q.

This argument is less convincing. We can propose a model as a counterexample.

(6) Suppose we have a range of possibilities, where some of them (not all)
make p true, and the all the others (again, not all of possibilities) make
q true. In that case, in each possibility, we have either p or q – so from
the point of view of any possibility at all, it is necessary that either p or
q. Nonetheless, we don’t have that it is necessary that p (since in some
possibilities, p fails), and we also do not have that it is necessary that q
(since in other possiblities, q fails). So, in any possibility in our model,
we do not have either that it is necessary that p or that it is necessary
that q. So this model, at any possibility, the premise is true, and the
conclusion is not.

Modal reasoning works just like the other reasoning we have seen. Valid argu-
ments can be broken down into proofs, while invalid arguments can be given
models as counterexamples. The models have a richer structure than the mod-
els we saw at first. We used not only a representation of one way that things
might be but also a range of such representations, a system of different possible
worlds.
There are many more concepts that can be rigorously explored with proofs
and models, like the identity predicate; definite and indefinite descriptions;
quantifiers (over objects) beyond the existential and universal quantifier; quan-
tifiers ranging over other domains, such as functions, propositions or properties
and much more. However, this will be more than enough to be going on with
for what follows.

1.2 Paradoxes
As soon as the field developed accounts of proofs and models – in fact, before
these tools took distinct shapes – some natural questions arose. How do we
evaluate those tools? Do they distinguish good and bad arguments correctly
(whatever that would mean), or should the main candidates for the correct
account of proofs or the correct account of models be revised or rejected? Some
6 Philosophy and Logic

of the most active revisionary arguments concerning proofs and models have
involved different kinds of paradoxes. After all, a paradoxical argument is one
where the premises seem true, the argument seems valid and the conclusion
seems false. If we want to find a good reason to take some argument that is
traditionally thought to be valid to be, in fact, invalid, then the paradoxes are
where we should look.

Example 1: The liar paradox

Consider this sentence, which says of itself that it is not true.

(7) Sentence (7) is not true.

It seems that we can reason like this. Suppose (7) is true. Then, since (7) says
that (7) is not true, then it would follow that (7) is not true, which would contra-
dict (7) being true. In other words, if (7) is true, we have a contradiction. This
means that (7) is not true, since we reduced the supposition that (7) is true to a
contradiction. We’ve refuted it. But this means that we have proved that (7) is
not true and that is what (7) itself says. So we’ve proved (7). It’s true. And so,
it isn’t. We have proved a contradiction.
This is the liar paradox, one example of a semantic paradox, and a paradox
of self-reference.5 As we will see in the next section, we have used very few
logical principles in this line of reasoning, and it very much looks like we have
made some kind of mistake, though it has proved very difficult to locate the
mistake to everyone’s satisfaction. For some, the semantic paradoxes like the
liar have been seen as reasons to curtail our rules of proof for the logical con-
nectives in some way or other, so as to stop the contradictory conclusion or to
render the contradictory conclusion palatable. For others, the fact that logical
principles like these are involved in the proof means that the problem must lie
elsewhere, either in the so-called definition of the liar sentence (say, we attempt
to ban self-reference) or to say that despite appearances, the logic of the truth
predicate cannot satisfy the rules used here in the derivation of the paradoxes.
There are many different kinds of response to the liar paradox, and we will dis-
cuss a representative sample of these in the coming sections, since doing so will
give us a range of perspectives on what we are doing when we use the logical
tools of proofs and models.

5 Another example, which we will also consider, is Curry’s paradox, which uses the conditional,
where the liar uses negation. Pick some statement p. Consider c, the class of all classes x where
if x is a member of itself, then p. Suppose c is a member of itself. Then, it follows that if c is a
member of itself, then p. So, combining those two facts, we have p. In other words, we have just
proved that if c is a member of itself, then p. So, it follows that c is a member of itself. Again,
putting these together, we have proved p. And we made no assumption about p at all.
 Proofs and Models in Philosophical Logic 7

Example 2: The sorites paradox

Consider a colour strip of colour, shading evenly from red on the left to yellow
on the right. Let’s divide the strip up into 10 000 evenly sized tiny patches, from
left to right, labelled 1 to 10 000. For each number n from 1 to 10 000, consider
the claim that patch number n looks red (to me). The first such claim, ‘patch 1
looks red (to me)’, is true. The last such claim, patch 10 000 looks red (to me), is
false. The claim ‘if patch 1 looks red to me, so does patch 2’ also seems true, not
just because both patches look red but also because they look indistinguishable
to me. This generalises: three distinct features conspire to make it that each
claim of the form ‘if patch n looks red to me, so does patch n + 1’ seems just
as true. First, the strip shades evenly from red to yellow, with no sharp changes
in observable colour. Second, we chose very many subdivisions, so each patch
differs from its neighbours by at most a tiny difference, and third, my powers of
visual discrimination have their limits. So the premises of this argument seem
true:
(8) Patch 1 looks red to me.
If patch 1 looks red to me, so does patch 2.
If patch 2 looks red to me, so does patch 3.
..
.
If patch 9 999 looks red to me, so does patch 10 000.
Therefore, patch 10 000 looks red to me.
From these premises, we can draw the conclusion that patch 10 000 looks red
to me, using one very simple principle of logic, the inference rule of modus
ponens, which takes us from a conditional claim of the form if A then B and
its antecedent A to its consequent, B. Unfortunately, for us, we seem to have
a logically valid argument from premises that seem true to a conclusion that
seems false.

 

If we wish to find a counterexample to the validity of the sorites argument, we

need to find some model in which the premises hold and in which the con-
clusion fails. If there is no such counterexample, then either we grant that the
argument is valid or we reject the constraint that invalidity must be witnessed
by a model as a counterexample. Classical ‘two-valued’ logical systems have
proved difficult to adapt to this task. A two-valued model will assign ‘true’ or
‘false’ to each sentence ‘patch n looks red to me’, which will involve assign-
ing either ‘true’ to each such claim (so this model represents the whole strip as
8 Philosophy and Logic

looking red to me) or ‘false’ to each claim (so the model represents the whole
strip as not looking red to me), or there are two adjacent patches, and the model
represents one as looking red to me and the other as not looking red to me. But,
as we said, the set-up is designed to make each patch indistinguishably differ-
ent from its neighbours. So, if that is unpalatable, a natural reaction involves
expanding the picture of semantic evaluation to allow for more than the two
values of ‘true’ and ‘false’: logics with truth-value gaps or a whole panoply
of degrees of truth might provide ways to understand the sorites paradoxes. In
Section 3, we will examine options for the sorites paradox, as well as other
reflections on models that have proved fruitful in philosophical logic in recent
decades.

1.3 The Plan

The semantic paradoxes and the sorites paradox are two examples of paradoxes
over which a great deal of ink has been spilt in recent decades. The philo-
sophical literature concerning the paradoxes provides one entry point – among
many – to the different approaches to understanding the foundations of logical
consequence, and this is our entryway into the broader landscape of the use of
proofs and models in philosophical logic. The paradoxes are sites where what
seemed for all the world to be fundamental principles about proofs and about
models come into tension and give rise to what seem to be absurd conclusions.
Different proposals for revising those fundamental principles or for defusing
the tension provide different approaches to understanding these principles of
logic, and they will provide a suitable set of lenses through which to view key
ideas in logic as they have developed.

 

In Section 2, we will discuss logic from the standpoint of proof, giving a quick
introduction to the kinds of techniques philosophical logicians have adopted
in the study of proof and its application to issues in semantics, epistemology
and metaphysics. In this section, we will keep an eye on the kinds of responses
people have made to the semantic paradoxes, as these paradoxes provide ample
motivation for us to inquire into the costs and benefits of different fundamental
proof-theoretical principles. Then, in Section 3, we will do the same thing for
models, introducing not only the debate over the applicability of the standard
two-valued ‘classical’ semantic picture in the light of the paradoxes, but also
our sights that will involve discussions of other kinds of models of use in phil-
osophical logic, such as models featuring possible worlds, which have proved
so useful, and so controversial, in giving an account of the meanings of modal
expressions.
 Proofs and Models in Philosophical Logic 9

After those two sections, we will wrap up with Section 4, in which we explore
not only the ways that these tools are used in the discussions of the paradoxes
but also some other natural questions, including the relationship between proofs
and models themselves. In particular, we will ask which proofs and models
should be taken to be fundamental. But, first, let us turn our attention to proofs.

2 Proofs
Let’s start with the proof labelled (2) on page 2. That is a proof that R is reflex-
ive. It has three premises: R is directed, R is symmetric and R is transitive.
It lays out a path from those premises to the conclusion, leaving nothing out.
The aim of a proof is not just to convince us that some conclusion is true but
also, in some sense, to make explicit how that conclusion follows from the
premises.
Notice that this proof does not just start from the premises and lead to the
conclusion, with each intermediate step following from the ones granted before
it.6 There are some other important features of our reasoning that are worth
examining. First, our proof includes an imperative: ‘take an object a’. This
sentence is not a premise, nor is it a conclusion, and it is not another statement
that follows from the premises. It is an invitation. It cannot be true or false. We
cannot assert or deny it. Second, the term ‘b’ in the proof also has an interesting
status. We moved from the claim that R is directed (so, in particular, there must
be some object to which R relates a) to calling one such object b. The fact that
R relates a to a given object b does not logically follow from the claim that R is
directed. R could be directed without R relating a to this particular b (whichever
b it happens to be).7 So much more is going on in this proof than simply working
out conclusions from things we have granted. There are different steps where
objects are given names, and speech acts, other than asserting, are involved
as well. Proofs have complex structure. A crucial constraint, though, is that a
proof is not simply a statement of the premises and the conclusion – at least,
not in most cases. To prove some conclusion C from some premises P1 , . . . ,
Pn , you must somehow trace the connection from P1, . . . , Pn to C.

6 Proofs with that direct ‘linear’ structure Hilbert Proofs. We will see these in the next section.
7 If you find this puzzling, think of a concrete case. I have a son. As a matter of fact, Zachary is
my son. The fact that Zachary is my son does not follow as a logical consequence of the fact
that I have a son because there are other ways I could have had a son, other than Zac. Similarly,
if R relates a to some object, in any particular circumstance in which that is true, we could call
that object b. Given that choice, it would still not follow that in every circumstance where R
relates a to something, that R must relate a to that object b. It might have related a to some other
object instead.
Other documents randomly have
different content
 « Quid agēs ? respondet alter : prīmam quamlibet nāvem, quae
hinc Hamburgum tendit, sine morā cōnscendēs ; tunc re [10]versus ad
tuōs peccātī veniam piē rogābis, pollicitus tē numquam posteā in
similī culpā fore. — Sed plānē careō pecūniā, ait Rōbinson. — Ēn
quattuor guineās, excēpit nauta, quās ego tibi commodābō, licet ipse
parvō, quod mihi superest, aegrē caream. Hīs tū adjūtus ad portum
tē cōnfer ; sit Deus tibi magis propitius redeuntī quam nōbīs fuit
nāvigantibus. » Hīs dictīs, manum benevolē jūnxit, atque faustum iter
ipsī precātus est. Abiit Rōbinson.

Dum ille portum peteret, varia sēcum in animō volvēbat :

« Quōmodo meī reducem mē excipient ? Castīgābunt sānē propter
tāle dēlictum. Sodālēs vērō, et tam multī aliī mē irrīdēbunt, quod tam
citō redierim. » Sīc diū dubitāns quidnam cōnsiliī caperet, ad portum
pergit ; sed ibi audiit, summā quidem cum voluptāte, nūllam adesse
nāvem, quae Hamburgum tenderet. Quī autem hāc dē rē eum
certiōrem fēcerat, ūnus ē praefectīs [11]eārum nāvium erat, quae ad
Guinean proficīscuntur.

Cum Rōbinson inter cōnfābulandum dīxisset, sē nōn dolēre

quod nūlla sibi Hamburgum redeundī adesset opportūnitās, quia
mīrā flagrābat cupiditāte peregrīnandī, praefectus nāvis eī auctor fuit
itineris ad Guineam faciendī. Quō audītō, prīmum Rōbinson
obstupuit. Sed cum praefectus itineris eī dēclārāvisset, iter hoc fore
jūcundissimum, sēque ut habēret ipse quīcum versārētur, grātīs eum
exceptūrum, ac praetereā rem eam esse unde quaestum ille nōn
mediocrem faceret, tum vērō tanta eum invāsit cupiditās
proficīscendī, ut subitō ē memoriā exciderit quidquid bonus nauta
Hamburgēnsis eum admonuerat.
 Sed rē paululum cōnsīderātā, « Equidem, ait Rōbinson, quattuor
tantummodo guineās habeō. Ecquid ego commerciī cum hāc exiguā
rē eō locō faciam, quō tū proficīsceris ? — Sex īnsuper guineās ego
tibi commodābō, respondit praefectus. [12]Nec majōre pecūniā tibi
opus est ad emendum unde multās in Guineā opēs cōnsequāris.
Quotiēs ab exiguīs initiīs rēs māximae profectae sunt !

« Sed quid ego hīs emam ? Rōbinson interrogat. — Merās

nūgās, respondit praefectus ; vitrēs, torquēs, cultrōs, forficēs,
secūrēs, taeniās, etc., quibus nigrī Āfricae incolae tantopere
gaudent, ut tibi vim aurī eborisque centiēs majōrem prō iīs datūrī
sint. »

Nec jam diūtius sibi Rōbinson temperāre potuit ; sed oblītus

parentum, amīcōrum, patriae, exclāmat : « Ēn ego tibi comes itineris
praestō sum. — Agedum, » respondit praefectus, dextrīsque jūnctīs
rem pacīscuntur.

Rōbinson itaque decem guineīs dīves ad urbem properat,

comparat variās mercēs, utī nāvarchus praecēperat, comparātāsque
in nāvem trānsportandās cūrat. Paucīs diēbus ēlāpsīs, ventō
favente, praefectus nāvis ancorās solvī, atque vēla ventīs darī jussit.
[13]
 Caput secundum.

Rōbinson pergit iter. — Mala ōmina. — Nāvis incēnsa. — Alia

flūctibus jactāta. — Advehitur ad īnsulās Canāriās. — Dēscrīptiō locī
illīus amoenissimī. — Inde profectus ad Americam naufragium facit.

N ovum hoc Rōbinsōnis iter faustissimum initium habuit. Jamque

incolumēs fretum Calesium trānsierant, et in ipsō Atlanticō marī
versābantur : tum vērō per plūrimōs diēs continuōs ventō reflante
nāvis Americam versus abrepta est.

At ecce vespere quōdam gubernātor dēclārāvit sē flammās

relūcentēs ā longinquō cōnspicere ; cumque īnspectā tabulā nauticā
intellegeret vel ad centum mīliāria nihil terrārum esse, conjectābat
hunc [14]ignem nihil aliud esse nisi nāvem incendiō flagrantem.

Vix haec dīxerat, cum in aurās tollī vīsa est nāvis, terribilī cum
fragōre, et mox tōta undīs obruta est. Cum nox supervēnisset, nihil
amplius cernī poterat. Prīmā autem lūce appārent duae scaphae,
cum flūctibus colluctantēs, quae rēmīs pertinācius ēverberābant
mare, ut ad nāvem accēderent. Extemplō praefectus vēxillum
nauticum explicuit, significāns sē ad opem illīs ferendam parātum
esse. Nāvis ipsa omnibus vēlīs ad eōs tendit, atque intrā dīmidium
hōrae ad miserōs pervenit.
 Sexāgintā erant, virī, mulierēs et puerī, quī omnēs nāve exceptī
sunt. Erat rēs omnīnō miserābilis, cum īnfēlīcēs illī faucibus mortis
sē ēreptōs vidērent. Aliī enim flēre prae gaudiō, aliī clāmāre, quasi
perīculum nunc prīmum imminēret ; illī exsultāre saltū lymphāticō, hī
vērō pallidī manūs torquēre. Nōnnūllī stāre mūtī et stupentibus
similēs. Nēmō quoque inter nau [15]tās tam ferreus, quīn hīs
cōnspectīs commōtus ipse lacrimārētur.

Cum eīs paulātim rediisset animus, ūnus nārrāvit quid miserīs

accidisset :

« Nāvis incēnsa magna quaedam nāvis erat mercātōria

Francōrum, quae ad īnsulam Mārtīnicam tendēbat. Exārserat ignis in
cubiculō gubernātōris, flammā tam vēlōciter grassante, ut nūlla ejus
exstinguendae spēs adforet. Vix per tempus licēbat sē in cymbās
recipere, et ab incēnsā nāve paululum discēdere, cum, flammā
cameram, in quā pulvis tormentārius sēpositus est, corripiente,
discerpta nāvis dissiluit. »

Dum haec nārrārentur, sedēbat in angulō Rōbinson silēns,

pallidusque, similis hominī quem malefactī remordet cōnscientia.
« Bone Deus ! ait ille sēcum, sī cum hominibus illīs, inter quōs sunt
profectō quīdam longē quam ego meliōrēs, tam male agitur, heu !
quid mihi ex [16]spectandum, quī sīc ergā parentēs peccāverim ? »

Cum illī cibō aliquō refectī fuissent, tum ūnus quī dignitāte
cēterīs praestāre vidēbātur, praefectum adiit, crumēnamque aureīs
nummīs plēnam porrigēns, « Ego inquit, miseram illam nāvem
īnstrūxī ; hoc sōlum ēripuī, ōrōque ut pignus grātī prō salūte animī
accipiās. »
 Tum vērō fuit rēs spectāculō digna, certāmen ambōs inter virōs,
uter hinc grātior, inde honestior foret.

« Absit sānē, respondet praefectus, ut mūnus tuum accipiam.

Cum ego vōbīs opem tulī, nihil aliud ēgī quam quod ā nātūrā hominī
ergā hominem praescrīptum est, scīlicet ut homō hominī, quīcumque
sit, cōnsulat, ob eam causam quod is homō sit. »

Frūstrā ille praefectum urgēre, ut oblātum mūnus acciperet ;

persevēranter negāvit, rogāns, ut rem plānē dīmitteret. [17]Tum
dēlīberātur, quōnam servātī hominēs essent vehendī. Duplex erat
causa cūr ad Guineam nōn dūcerentur. Prīmō quidem minimē opus
erat longum iter in eam regiōnem facere, nihil ibi negōtiī habentibus.

Nec deinde tanta aderat in nāve cibōrum cōpia, unde tot

nāvigantēs in itinere vīctitārent.

Praefectus tandem statuit, omissā propriā ūtilitāte, centum et

amplius mīliāribus ā viā rēctā dēflectere, eōsque in Terram Novam
trānsportāre, ubi spērābat fore, ut sēsē offerret opportūnitās in
Galliam cum piscātōribus asellōrum redeundī. Eō igitur cursum
dīrēxit ; cumque advēnisset, nāvēs Gallicās nactus est, quae
miserōs recēpērunt. Hīs perāctīs, iter suum ad Guineam prōsecūtus
est.

Tum nāvis celerrimō cursū, aquās secat ; quā nāvigandī

vēlōcitāte Rōbinson noster mīrē dēlectātus est. Post aliquot diēs,
ecce magnam nāvem cōnspiciunt, ad sē ten [18]dentem. Mox vērō
audiunt sonitum tormentōrum perīculī īnstantis indicem, et
animadvertunt nāvem esse duōbus mālīs orbātam. Cum ad eam
propius accessissent, quī in eā vehēbantur, sublātīs manibus,
miserābiliter exclāmant : « Servāte īnfēlīcēs quibus omnibus
pereundum est, nisi vōs eōrum miserēbit. »

Tum ex iīs quaesītum est quid malī accidisset ; quīdam vērō ex

illīs sīc ōrsus est :

« Anglī sumus, ex īnsulā Jamaicā, saccharī vectūram inde

dēportātūrī. Ibi dum nāvis stāret ancorīs alligāta, praefectus cum
gubernātōre in terram dēscendit, ad mercēs aliquās īnsuper
ēmendās.

« Intereā exorta est tempestās, tantō turbine, ut, fūne disruptō,

nāvis ex portū in altum prōpellerētur. Tempestās ista trēs diēs
noctēsque saeviit : tum mālīs omnibus āmissīs, centum et plūra
mīliāria abreptī sumus. Accēdit ad hanc nostram calamitātem, quod
nēmō nostrōrum artis nauticae perītus sit : novem jam [19]integrās
hebdomadēs hinc et inde jactātī, vīctum omnem cōnsūmpsimus, et
nostrī plērīque jacent famē exhaustī. »

Bonus itaque praefectus statim cymbam expōnī jussit,

assūmptāque sēcum idōneā cibōrum cōpiā, ipse cum Rōbinsōne ad
nāvem accēdit. Erat sānē lūgenda prōrsus omnium condiciō quī in
illā vehēbantur : ūniversī inediā quasi cōnsūmptī ; nōnnūllī vix
poterant pedibus stāre.

Cum autem cubiculum nauticum intrāssent, horrendum vīsū !

jacēbant humī māter, fīlius et servula, quī omnēs famē ēnectī
vidēbantur. Mox autem dēprehēnsum est, reliquiās in eīs animae
vītālis superesse. Postquam enim nōnnūllae sūcī ē carne expressī
guttae in ōs cujusque īnstillātae sunt, lūcem oculīs quaerere
coepērunt.
 Māter prae nimiā vīrium imbēcillitāte nihil quidquam absorbēre
poterat ; cum autem innuisset, ut tantummodo fīliō suō cōnsulerētur,
mox illa exspīrāvit.
[20]

Inter haec duo reliquī animam recēperant ; atque ut erant aetāte

rōbustiōrēs, praefectī dīligentiā fēlīcius servātī sunt. Cum autem
juvenis, oculīs in mātrem conjectīs, mortuam esse intellēxisset,
tantus eum invāsit dolor, ut dēficientibus iterum vīribus aegrē ad
vītam revocārī posset. Vīcit tamen cūra, atque ille cum servā quoque
ē faucibus mortis ēreptus est. Deinde praefectus nāvem omnī
genere alimentōrum īnstrūxit, mālōs ā fabrīs suīs reficiendōs cūrāvit,
ignārīsque maris perītum nautam dedit, quī nāvem regeret ; atque ad
terram proximam ipse tendit, ut novam cibāriōrum comparātiōnem
faceret. Haec erat īnsula Madeira, ūna ē Canāriīs. Rōbinson ūnā
cum praefectō in terram dēscendit, atque ibi laetissimō fortūnātae
illīus īnsulae adspectū satiārī nōn potuit. Scīlicet terra partim in
plānitiem porrēcta, partim in colliculōs molliter assurgēns, vernantem
frūgiferārum arborum cōpiam explicat. Caelī admodum jūcunda
temperiēs : nūllus [21]enim hīc hiemis rigor ; ita vim omnem frīgoris
retundunt clēmentiōrēs sōlis radiī mollēsque favōniōrum animae,
quārum flābellīs regiō tōta circumquāque ventilātur. Erant in oculīs,
quam longē patēbat prōspectus, segetēs in agrīs bene pinguibus
diffūsae, juga montium continuīs vītibus cōnsita. Ut arrīdēbant
Rōbinsōnī nostrō pendentēs racēmī ! Ō quālēs ille hausit dēliciās,
cum praefectus aere datō impetrāvit, ut juvenī licēret ūvīs prō libīdine
satiārī. Postquam ibi aliquantum temporis morātus esset nāvis
reficiendae causā, ancoram solvit.

Per plūrēs continuōs diēs fēlīcissimō cursū ūsī sunt.

 At ecce repente vehemēns tempestās ab austrō oritur.
Spūmābant flūctūs, et in immēnsam surgēbant altitūdinem. Nāvis
tamen stetit invicta ; sed cum sex diēs continuōs atrōx tempestās
saeviisset, illa tam longē ā viā dējecta est, ut nec praefectus, nec
gubernātor amplius intellegerent, quō [22]nam in locō versārentur.
Arbitrābantur tamen sē ab īnsulīs, quae Caraibae vocantur, nōn
multum distāre. Septimō autem diē, cum prīmum illūcēsceret, ē
nautīs quīdam, ingentī omnium laetitiā, terram inclāmat. Tum omnēs
in stegam cōnscendere, vīsūrī quaenam sit terra illa ad quam
appulsūrī sunt. Sed nunc laetitia in summum terrōrem mūtāta est.
Ecce enim nāvis impingitur : quō ictū quī stābant in stegā excussī
omnēs plānē corruērunt.

Scīlicet nāvis in syrtem tam vehementī impetū allīsa erat, ut

haerēret quasi affīxa. Cōnfestim vērō flūctūs spūmantēs tantam vim
aquae in stegam profūdērunt, ut omnibus in cubiculō fuerit
cōnfugiendum, nē ipsī abriperentur.

Tum flēbilis nautārum orīrī clāmor ; tum ejulātus hinc Deum

ōrantium, inde clāmantium : aliī dēspērantēs sibi crīnēs āvellere,
quīdam jacēre sēmianimī.

Hōs inter Rōbinson erat exspīrantī similis ; ac repente exclāmant

nāvem dēhīscere. [23]Itaque in stegam omnēs accurrere, et
dēmissam quam celerrimē in scapham omnēs dēsilīre. Tanta autem
erat multitūdō incōnsultē ruentium, ut vix scapha palmae lātitūdine
undās superāret. Cum vērō terra procul obesset, cūnctī dē ōrā
umquam assequendā, saeviente adeō tempestāte, dēspērābant.

At ecce ingēns flūctus ad scapham volvī, montis īnstar : quō

omnēs cōnspectō obstupēscere, rēmōsque dīmittere. Nunc, nunc
īnstat perīculum. Et jam aquae mōns scapham assequitur
subvertitque ; tum ūniversī hauriuntur.
[24]
 Caput tertium.

Sēra Rōbinsōnis paenitentia. — Dēspērātiō. — Vītam miserē

sustentat. — Habitat in spēluncā.

F ēlīcī fātō contigit, ut ingēns ille flūctus quī Rōbinsōnem

absorpserat, īdem vī magnā reciprocāns, hominem in vadō
exspueret. Hic cum praeter exspectātiōnem in siccō versārētur, tum
extrēmīs vīribus ūsus est, ut summam in ōram adscenderet ;
cumque eō pervēnisset, coepit circumspicere. Ēheu ! quālia oculīs
obversantur ! scapha, nāvis, sociī, omnia dēmersa. Nihil omnīnō
reliquum praeter āvulsās tabulās, quae flūctuantēs ad terram
ferēbantur. Ūnus ipse, ūnus mortem effūgerat. Gaudiō igitur et
terrōre trepidāns, [25]in genua prōcubuit, manibusque sublātīs,
lacrimīs perfūsus, Deō prō salūtis mīrāculō grātiās ēgit. Cum vērō
nihil praeter arbusta aut arborēs dēprehenderet, nec ūlla vestīgia,
unde intellegere posset hanc regiōnem ab hominibus incolī, jam
gravissimum illī vidēbātur vītam sīc in sōlitūdine dēgere. Vēnit
quoque in mentem esse forte vagantēs bēluās, aut ferōs hominēs ;
tum multō majōrī horrōre perculsus est. Itaque prīmum prae timōre
nec stāre ausus, nec prōgredī, circumspectāre omnia, et vel minimō
strepitū expavēscere. Sed mox quā ārdēbat sitis torpōrem istum
excussit ; quam cum diūtius tolerāre nōn posset, fontem aut rīvum
indāgāre coepit, et sīc vestīgantī occurrit mānāns per herbās
rīvulus : quem ille secūtus, invēnit fontem liquidissimum, quō valdē
recreātus est.

Nec illum famēs tunc māximē urgēbat ; quippe cui angor

terrorque omnem cibī appetītum exēmerant. Sed erant adeō
ex [26]haustae illīus vīrēs, ut multō magis requiēscere optāret.
Attamen caput erat locum invenīre, in quō tūtus pernoctāret. Nūlla
domus, nūllum tugurium, nūllum specus erat in cōnspectū. Sīc diū
cōnsiliī inops flēvit. Tandem avēs imitārī, hospitiumque in arbore
quaerere statuit. Ac brevī ūnam prōspexit tam dēnsam opācamque,
ut commodē īnsidēre dorsumque acclīnāre posset : hanc adscendit
rēptandō, piēque Deum precātus, cum sē composuisset, statim
obdormīvit. Inter somnum, illīus mentī, ut erat cūrārum aestū jactāta,
rērum eārum, quae prīdiē vigilantī acciderant, recursābant imāginēs.
Vidēre sibi vidēbātur mōtum et agitātiōnem flūctuum nāvem
haurientium, nautārumque ejulātūs exaudīre. Sibi in animō fingēbat
parentēs lūctū ac maerōre cōnfectōs, quī fīliī vicem dolentēs,
manibus ad caelum sublātīs, flēbant, et omne prōrsus sōlācium
āversābantur. Tunc altā vōce exclāmat, inter somniandum : « Adsum,
adsum, parentēs [27]cārissimī ! » Dumque amplexū mātrem petit,
dēcidit ex arbore. Sed forte quod ipsī faustum fuit, sēdem in locō nōn
ita sublīmī posuerat, solumque tam dēnsō grāmine vestītum erat, ut
ille cāsū nōn ita gravī dēcideret. Iterum arborem suam cōnscendit, in
quā manibus retinēns surculōs circā ēminentēs, ad ortum sōlis
remānsit.

Prīmā lūce coepit cōgitāre, quā ratiōne vīctum sibi parāret.

Omnēs vērō arborēs quās hāctenus cōnspexerat, erant ex eārum
genere quās campêche vocāmus ; atque illae folia quidem, sed
frūctūs ferunt omnīnō nūllōs.
 Nec satis illī liquēbat, quid sibi faciendum esset. Ex arbore
tamen dēscendit. Cum autem prīdiē nihil omnīnō comēdisset, coepit
famē saevissimā labōrāre : itaque hūc illūc per aliquot mīlia passuum
discurrit ; sed discurrendō nihil invēnit, nisi arborēs sterilēs et
grāmina.

Tunc summō angōre excruciātus famem [28]et ultima sibimet

ōminābātur : « Pereundum mihi erit famē ! » exclāmat, miserābiliter
ejulāns. Dēspērantī vērō necessitās satis animī vīriumque adjēcit, ut
sē ad lītus maris cōnferret, investīgāretque dīligentius, sī quid forte
alimentī invenīret. Sed sua eum spēs fefellit. Hīc vērō dēbilitātus
frāctusque prōcubuit, faciē in terram dēmissā, flentemque miserē
paenituit, quod nōn fuisset flūctibus et ipse obrutus. Et jam statuerat
in hōc miserābilī rērum articulō exspectāre, dum lenta famēs morte
omnium taeterrimā cōnsūmeret vītam, cum forte conversus, ecce
cōnspexit nōnnūllās in harēnā squāmās ostreārum. Avidus itaque eō
accurrere, circumspicere, scrūtārī, num forte plēnās quāsdam
reperīret, repertīsque exsultāre.

Ac prīmō quidem illās nōn sine difficultāte aperuit ; sed īnsertō

plānō eōdemque acūtō silice perfēcit, ut trīgintā circiter haustīs
ventrem paululum plācāret. [29]Cum nunc certior esset, sē posse
quamvīs miserē sustentāre vītam, jam māximē sollicitus erat, quō sē
ā ferīs hominibus aut bēstiīs tūtum reciperet.

Tum in montem summum, aegrē, ut erat satis excelsus,

adscendit, unde quam lātissimē circumspicere poterat. Sed quantus
eum terror invāsit, cum vīdisset sē esse in īnsulā, neque ūllam
omnīnō, quam longē aciem oculōrum intendere posset, terram
appārēre, praeter trēs parvās īnsulās, quae, spatiō aliquot mīliārium
interjectō, ex aequore ēminēbant !
 « Heu ! mē miserum ! exclāmat, manibus ad caelum miserābiliter
sublātīs. Mē nē ab hominum commerciō ita sējūnctum esse, ut mihi
in hāc sōlitūdine vīvendum sit velut ferae bēstiae ! Heu ! parentēs
optimī, numquam igitur ego vōs revīsam ! Numquam licēbit culpae
meae veniam exōrāre ! »

Et sīc dolēns dēscendit dē monte, cir [30]cumspiciēns, ubi posset

in tūtō sē collocāre. Diū autem frūstrā indāgāvit ; tandem ad locum
ēditōrem pervēnit, ex adversā parte, īnstar parietis, praeruptum.
Cum accūrātius explōrāret, locum invēnit leviter excavātum, satis
angustō aditū. Hīc sī uncum, sī scalprum, aut alia quaevīs
īnstrūmenta habuisset, facilī negōtiō cavum in saxō excīsum per
fragmenta habitāculō idōneum fēcisset. Sed cum haec omnia
dēsīderāret, huic inopiae medendum erat.

Cum diū multumque dēlīberāsset, sīc sēcum statuit : « Arborēs,

quās hīc videō, salicibus patriīs similēs sunt, quae, sī trānsferantur,
novās facile rādīcēs agunt. Ēruam igitur manibus satis magnam
eārum cōpiam, quae, ante cavum dēnsō ōrdine cōnsitae, erunt prō
mūnīmentō ; cum illae rādīcēs ēgerint, ibi licēbit clausō et sēcūrō
mihi quasi in domō dormīre. »
[31]

Tunc fēlīcī hōc inventō laetus, ad opus statim sē accīnxit. Sed

quantō majōre gaudiō exsultāvit, cum huic locō proximum fontem
liquidissimum prōspexisset ē monte exsilientem ! Eō statim
properāvit, quia, cum hūc illūc fervente sōle discurrisset, multum
aestuābat.

Posteā vērō coepit manibus et multō sūdōre tenerās aliquot

arborēs exstirpāre, quās eō trānstulit, ubi sēdem sibi dēstināverat,
atque ibi quoque cavum fodere manibus oportuit, in quō arborēs
īnsereret : hōc vērō multum temporis requīrente, vesper advēnit, vix
quīnque aut sex numerō arboribus cōnsitīs. Famē compulsus, ōram
maris rūrsus petiit, ut novās conquīreret ostreās. Sed cum aestus
rediisset, tōtum lītus occupāverant undae : aegrē igitur ūnam aut
alteram repperit, et sīc male pāstus cubitum īre coāctus est.
Reversus est igitur ad arborem suam, ibi pernoctātūrus, dōnec
hospitium sibi [32]tūtum cōnfēcisset. Nē autem hāc nocte idem ipsī
accideret quod superiōrī, genuālibus sē eī rāmō religāvit, ad quem
dorsum prius acclīnāverat, cumque summō rērum creātōrī sē
commendāsset, tranquillē obdormīvit.
[33]
 Caput quārtum.

Rōbinson reperit pōma eximiae magnitūdinis. — Sibi cōnficit

varia īnstrūmenta. — Fūniculōs. — Strātum. — Umbellam. — Pēram.
— Kalendārium.

H āc nocte Rōbinson nōn dēcidit ex arbore ; sed placidē ūsque

ad ortum sōlis quiēvit.

Prīmā lūce, lītus petiit, ut, collēctīs ibi ostreīs, ad opus

inchoātum reverterētur. Cum eō aliā viā tenderet, euntī contigit
invenīre arborem quae pōma inūsitātae magnitūdinis ferēbat. Etsī
eōrum indolem ignōrābat, spērāns tamen ea fore quibus vēscī
posset, ūnam ex iīs dēcussit. Erat autem nux oblonga nec minor
capite [34]puerulī, quae vulgō in Americae īnsulīs vocātur cocossa.
Erat nucleus quasi medulla et sūcus quī sapōrem oleī ex dulcī
amygdalā expressī referēbat, atque in mediō lac exquīsītum.

Cum stomachum lātrantem ūna nōn plācāret, secundam, tertiam

quoque dēcussit, quibus abundē satiātus, tamen lītus maris petiit ;
cumque mare reddidisset terrās paulō ante salō mersās, nōnnūllās
passim in siccō ostreās repperit. Magnam quoque et plānam in lītore
concham invēnit, quā prō spathā ūsus est ; atque illud ipsī nōn
mediocrī fuit adjūmentō. Nec multō post plantam dēprehendit fibrōsō
caule, quālis est apud nōs līnī aut cannabis.
 Haec ille et tālia quōvīs aliō tempore neglēxisset ; nunc vērō
nihil omittēbat, intentus ad singula, sī forte inde aliquid ūtilitātis
caperet.

Itaque spērāns id plantae genus nōn minus sibi fore

opportūnum, quam līnum [35]nōbīs, aut cannabem, magnam ejus
cōpiam ēvulsit, colligātōsque fasciculōs in aquam dēmersit.

Diēbus aliquot perāctīs, cum animadvertisset exteriōrem

crustam aquā jam satis ēmollītam, fasciculīs inde sublātīs, stirpēs
molliōrēs factās ad sōlem aprīcum explicuit. Quibus vix mediocriter
siccātīs, tentāvit num hoc quoque līnī īnstar repetītō fustis ictū
conterī posset. Cumque rēs ex sententiā cessisset, cōnfestim
fūniculōs facere tentat tantā firmitāte, at concham baculō affīxam
alligāre posset. Et sīc habuit īnstrūmentum haud absimile ligōnī
holitōris.

Tum vērō opus assiduē persecūtus est, dēnsāsque cōnseruit

arborēs, dōnec āreolam ante prōpositum sibi habitāculum omnīnō
saepsisset ; atque inde quasi pariēs exstitit adeō validus, ut nōnnisi
vī multā expugnārī posset.

Illās veniente diē, illās dēcēdente irrigābat aquā ē fonte proximō

haustā. Ac [36]brevī magnō cum gaudiō tenerās vīdit arborēs suās
virēscentēs, crēscentēsque adeō fēlīciter, ut mīra esset in ipsō
adspectū locī amoenitās.

Cum saeptum perfēcisset, continuum restī torquendae impendit

diem, ex quā sibi scālam cōnficeret : nōluerat enim habitāculum
suum portā claudere, nē satis firma nōn foret adversus irruentis
ferae impetum.
 Quae rūpēs habitāculō imminēbat, habēbat circiter duodecim
pedum altitūdinem : stābat in summā arbor, cui scālam ita alligāvit, ut
in terram dēpendēret. Tum ipse cōnscendere rūpem aggressus est ;
quae rēs ex vōtō cessit.

Hīs omnibus cōnfectīs, sēcum dēlīberāre coepit quōmodo

cavum augēret, ita ut esset spatium in quō habitāret. Atque facile
intellēxit tāle opus sōlīs manibus nōn esse suscipiendum.
Excōgitandum erat īnstrūmentum, quod foret adjūmentō.
[37]

Eā mente regiōnem lūstrāvit, et post multās investīgātiōnēs

nactus est lapidem secūris fōrmam referentem, praeacūtum, atque in
eō forāmen in quō manūbrium intrūdī posset. Quae rēs quoque
multō labōre, et lapide aliō adhibitō fēlīciter successit. Cumque
fustem alligāsset fūniculīs quōs propriā arte parāverat, haerēbat
manūbrium quasi clāvīs affīxum.

Dum pergit quaerere, duōs invēnit lapidēs aliōs variīs operibus

aptōs : alter īnstar erat tudiculae ; alter ita fōrmātus, ut aciem cuneī
habēret. Utrōque Rōbinson assūmptō, tam alacrī animō reversus
est, ut extemplō sē ad opus accingeret.

Quī quidem tum cuneō, tum tudiculā sēnsim multa rūpis

fragmenta dēcussit. Rēs erat plēna sānē labōris. Sed diēbus aliquot
in opere cōnsūmptīs tantopere prōcesserat, ut spatium cubiculō
idōneum exīstimāret. Tunc āvulsō grāmine siccātōque, fēnum
cōnfēcit, lectulum inde commodum sibi parātūrus.
[38]

Nihil igitur fuit impedīmentō, quōminus in suggestū mollī somnō

indulgēret, postquam, ālitum mōre, sedentī in arbore octo et amplius
noctēs fuerant trādūcendae. Ō quantā voluptāte perfūsus est, cum
licuit fessōs artūs in lectulō compōnere !

Dominica diēs proxima erat. Quam Rōbinson ōrandō et

meditandō cōnsūmpsit, rogāns Deum sibi peccātōrum veniam,
parentibus vērō miserīs praesidium et sōlācia.

Atque nē diērum ōrdō excideret ē memoriā, diem vērō

dominicam semper dignōsceret, vēnit in mentem Rōbinsōnī sibi
kalendārium condere ; nōn illud quidem tam accūrātum, nec typīs
impressum, quālī nōs in Eurōpā ūtimur ; ejus modī tamen, ut ad diēs
numerandōs sufficeret. Quod quidem tālī modō assecūtus est.
Quattuor arboribus contiguīs cortice quam lēvissimō ēlēctīs, ūnam,
quae magnitūdine cēterās superābat, quōque vespere līneolā incīsā
notāvit, quae diem absolūtum significāret. Post incīsās septem
[39]līneolās, erat fīnis hebdomadae, et hunc indicābat exarāta in
proximā arbore līnea.

Quotiēs in alterā hāc arbore quattuor aut quīnque līneolās

īnsculpserat, in tertiā arbore similī līneā notābat integrum mēnsem
ēlāpsum. Duodecim autem līneīs mēnstruīs absolūtīs, in quārtā
arbore signābat annum perāctum. Cum autem mēnsēs quīque nōn
sint pārī diērum numerō, vicibus alternīs, trīgintā ūnī, trīgintā et ūnum
alterī incīdēbat, exceptō mēnse februāriō, quem octo et vīgintī
tantum līneolīs dēsignābat.

Hīs ergō annālibus, Rōbinson temporum ōrdinem servāvit,

diemque dominicam ā cēterīs dignōscere potuit, quam mōre
chrīstiānōrum rīte celebrāret.

Interim erat cōnsūmpta māxima pars nucum cocossae ejus,

quam ūnam hūc ūsque invēnerat ; atque in lītore tam rārae
appārēbant ostreae, ut hominī alendō nōn sufficerent. Itaque dē
vīctū iterum sollicitus esse coepit.
[40]

Hāctenus quidem nōn longē ab habitāculō suō discesserat,

metū ferārum aut hominum. Jam vērō necessitāte coāctus, īnsulam
perlūstrāre ausus est, ut nova quaereret alimenta ; cōnstituitque diē
proximā facere excursiōnem. Sed, ut sē ā nimiō sōlis aestū
dēfenderet, vesperam istam umbellae cōnficiendae impendit. Cum
habēret neque linteum, neque virgulās bālaenārum, nec forficēs,
neque acum, neque fīla, ē vīminibus salignīs tegmen strūxit, et
īnsertum in mediā parte baculum fūniculīs alligāvit. Tum ex arbore
cocossae folia sūmpsit lātiōra, quae tegminī spīnīs affīxit. Sīcque
umbellam cōnfēcit tantā dēnsitāte, ut sōlis radiīs esset omnīnō
impervia. Atque eādem arte textīs fūniculīs fabricāvit sibi pēram, in
quā et vīctum sēcum portāret, et conderet alimenta, sī modo
quaedam invenīre sibi contigisset.

Tum vērō tantopere laetātus est opere perfectō, ut noctem prae

gaudiō īnsomnem trādūxerit.
[41]
 Caput quīntum.

Rōbinson īnsulam perlūstrat. — Magnus terror. — In gaudium

vertitur. — Dēscrīptiō lamae. — Ūnum occīdit. — Sed igne caret. —
Carnem mōre Tartarōrum coquit.

R ōbinson vix ortum diēī exspectāre poterat ; ita praevertit lūcem

ipsam. Pēram fūnemque corporī circumdedit ; secūrīque prō
gladiō cīnctus, et umbellā umerīs impositā, in viam alacriter sē dedit.
Ac prīmum arborem cocossam adiit, ut ūnam aut alteram nucem
dēcerperet ; tum lītus quoque revīsit, ostreās collēctūrus. Cumque
utrīusque generis satis magnam cōpiam sibi comparāvisset, iter
ingressus est.

Mīra caelī mātūtīnī erat serēnitās. Sua redeunte lūce singulīs

rēbus fōrma redierat. [42]Sōl oriēns clārissimō fulgēbat splendōre,
quasi ex ōceanō ēmergeret, arborumque cacūmina velutī inaurābat.
Multae undique avēs variīs mīrīsque pictae colōribus carmen
canēbant mātūtīnum, aurōram salūtantēs. Pūrō frīgidōque āere,
flōrēs suāvissimum odōrem exhālābant.

Rōbinson gaudiī et grātitūdinis sēnsū perfūsus : « Hīc quoque,

ait, hīc Deus sē benignissimum dēclārat. » Tum vōcem ipse suam
cantibus avium miscuit, verbaque cecinit, quae pietās suggerēbat.