stanfordnlp/CoreNLP · error · RuntimeException

: Term lacks morpho analysis

Error message

%s: Term lacks morpho analysis: %s

What it means

transformTree adds morphosyntactic features to POS tags, but only if the terminal's label is a CoreLabel carrying a non-null originalText holding the morpho analysis string (lemma+morph separated by the morph separator). If the tag spec is active yet the leaf lacks that analysis, the parser cannot derive features and throws with the offending tree.

Solutions

  1. Run the morpho annotation pass (MorphoTreeTransformer with -morphoFile) on the trees before transformTree.
  2. Ensure leaves are CoreLabels that preserve originalText (label annotations must not be stripped).
  3. Disable the morpho feature option (tagSpec) if no morphological analysis is available.

Example fix

// before
LexicalizedParser.parseTrees(op.tlpParams, trees) // trees lack morpho annotation
// after
MorphoTreeTransformer mt = new MorphoTreeTransformer(morphSpec);
List<Tree> annotated = trees.stream().map(mt::transformTree).collect(Collectors.toList());
Defensive patterns

Strategy: validation

Validate before calling

for (Tree t : trees) {
    for (Tree leaf : t.getLeaves()) {
        if (!(leaf.label() instanceof CoreLabel) || ((CoreLabel) leaf.label()).originalText() == null)
            throw new IllegalStateException("Leaf lacks morpho analysis: " + leaf);
    }
}

Try / catch

try { tree = params.transformTree(tree); } catch (RuntimeException e) { log.error(e.getMessage()); throw e; }

Prevention

When it happens

Trigger: Parsing/training with -tagPairedMorphoFeatures (or equivalent tagSpec) while the input trees' terminals were not annotated with morpho strings via a MorphoTreeTransformer / -morphoFile preprocessing step.

Common situations: Feeding raw French treebank trees to a morphologically annotated model, or forgetting the morpho annotation pass so originalText is null on the leaves.

Understand the failure class

Background: "missing required argument" and "the following required arguments were not provided": what required-argument errors mean and how to fix them — this error's family across 20 libraries.

Related errors

AI-assisted analysis of stanfordnlp/CoreNLP@1b7edd19c4 (2026-09-10). Data as JSON: /api/errors/c5c1c57476f39391. Report an issue: GitHub.

Appendix: source

Thrown at src/edu/stanford/nlp/parser/lexparser/FrenchTreebankParserParams.java:590

    return (readPennFormat) ? new FrenchTreeReaderFactory() : new FrenchXMLTreeReaderFactory(false);
  }

  public List<HasWord> defaultTestSentence() {
    String[] sent = {"Ceci", "est", "seulement", "un", "test", "."};
    return SentenceUtils.toWordList(sent);
  }

  @Override
  public Tree transformTree(Tree t, Tree root) {
    // Perform tregex-powered annotations
    t = super.transformTree(t, root);

    String cat = t.value();

    //Add morphosyntactic features if this is a POS tag
    if(t.isPreTerminal() && tagSpec != null) {
      if( !(t.firstChild().label() instanceof CoreLabel) || ((CoreLabel) t.firstChild().label()).originalText() == null )
        throw new RuntimeException(String.format("%s: Term lacks morpho analysis: %s",this.getClass().getName(),t.toString()));

      String morphoStr = ((CoreLabel) t.firstChild().label()).originalText();
      Pair<String,String> lemmaMorph = MorphoFeatureSpecification.splitMorphString("", morphoStr);
      MorphoFeatures feats = tagSpec.strToFeatures(lemmaMorph.second());
      cat = feats.getTag(cat);
    }

    //Update the label(s)
    t.setValue(cat);
    if (t.isPreTerminal() && t.label() instanceof HasTag)
      ((HasTag) t.label()).setTag(cat);

    return t;
  }


  private void loadMWMap(String filename) {
    mwCounter = new TwoDimensionalCounter<>();

View on GitHub (pinned to 1b7edd19c4)